diff --git a/tests/data/azure-devops.req-310-expected.json b/tests/data/azure-devops.req-310-expected.json index d914d42..a422091 100644 --- a/tests/data/azure-devops.req-310-expected.json +++ b/tests/data/azure-devops.req-310-expected.json @@ -126,12 +126,12 @@ "type": "pypi", "namespace": null, "name": "azure-core", - "version": "1.33.0", + "version": "1.34.0", "qualifiers": {}, "subpath": null, "primary_language": "Python", - "description": "Microsoft Azure Core Library for Python\n# Azure Core shared client library for Python\n\nAzure core provides shared exceptions and modules for Python SDK client libraries.\nThese libraries follow the [Azure SDK Design Guidelines for Python](https://azure.github.io/azure-sdk/python/guidelines/index.html) .\n\nIf you are a client library developer, please reference [client library developer reference](https://github.com/Azure/azure-sdk-for-python/blob/main/sdk/core/azure-core/CLIENT_LIBRARY_DEVELOPER.md) for more information.\n\n[Source code](https://github.com/Azure/azure-sdk-for-python/blob/main/sdk/core/azure-core/)\n| [Package (Pypi)][package]\n| [Package (Conda)](https://anaconda.org/microsoft/azure-core/)\n| [API reference documentation](https://learn.microsoft.com/python/api/overview/azure/core-readme)\n\n## Getting started\n\nTypically, you will not need to install azure core;\nit will be installed when you install one of the client libraries using it.\nIn case you want to install it explicitly (to implement your own client library, for example),\nyou can find it [here](https://pypi.org/project/azure-core/).\n\n## Key concepts\n\n### Azure Core Library Exceptions\n\n#### AzureError\n\nAzureError is the base exception for all errors.\n\n```python\nclass AzureError(Exception):\n def __init__(self, message, *args, **kwargs):\n self.inner_exception = kwargs.get(\"error\")\n self.exc_type, self.exc_value, self.exc_traceback = sys.exc_info()\n self.exc_type = self.exc_type.__name__ if self.exc_type else type(self.inner_exception)\n self.exc_msg = \"{}, {}: {}\".format(message, self.exc_type, self.exc_value) # type: ignore\n self.message = str(message)\n self.continuation_token = kwargs.get(\"continuation_token\")\n super(AzureError, self).__init__(self.message, *args)\n```\n\n*message* is any message (str) to be associated with the exception.\n\n*args* are any additional args to be included with exception.\n\n*kwargs* are keyword arguments to include with the exception. Use the keyword *error* to pass in an internal exception and *continuation_token* for a token reference to continue an incomplete operation.\n\n**The following exceptions inherit from AzureError:**\n\n#### ServiceRequestError\n\nAn error occurred while attempt to make a request to the service. No request was sent.\n\n#### ServiceResponseError\n\nThe request was sent, but the client failed to understand the response.\nThe connection may have timed out. These errors can be retried for idempotent or safe operations.\n\n#### HttpResponseError\n\nA request was made, and a non-success status code was received from the service.\n\n```python\nclass HttpResponseError(AzureError):\n def __init__(self, message=None, response=None, **kwargs):\n self.reason = None\n self.response = response\n if response:\n self.reason = response.reason\n self.status_code = response.status_code\n self.error = self._parse_odata_body(ODataV4Format, response) # type: Optional[ODataV4Format]\n if self.error:\n message = str(self.error)\n else:\n message = message or \"Operation returned an invalid status '{}'\".format(\n self.reason\n )\n\n super(HttpResponseError, self).__init__(message=message, **kwargs)\n```\n\n*message* is the HTTP response error message (optional)\n\n*response* is the HTTP response (optional).\n\n*kwargs* are keyword arguments to include with the exception.\n\n**The following exceptions inherit from HttpResponseError:**\n\n#### DecodeError\n\nAn error raised during response de-serialization.\n\n#### IncompleteReadError\n\nAn error raised if peer closes the connection before we have received the complete message body.\n\n#### ResourceExistsError\n\nAn error response with status code 4xx. This will not be raised directly by the Azure core pipeline.\n\n#### ResourceNotFoundError\n\nAn error response, typically triggered by a 412 response (for update) or 404 (for get/post).\n\n#### ResourceModifiedError\n\nAn error response with status code 4xx, typically 412 Conflict. This will not be raised directly by the Azure core pipeline.\n\n#### ResourceNotModifiedError\n\nAn error response with status code 304. This will not be raised directly by the Azure core pipeline.\n\n#### ClientAuthenticationError\n\nAn error response with status code 4xx. This will not be raised directly by the Azure core pipeline.\n\n#### TooManyRedirectsError\n\nAn error raised when the maximum number of redirect attempts is reached. The maximum amount of redirects can be configured in the RedirectPolicy.\n\n```python\nclass TooManyRedirectsError(HttpResponseError):\n def __init__(self, history, *args, **kwargs):\n self.history = history\n message = \"Reached maximum redirect attempts.\"\n super(TooManyRedirectsError, self).__init__(message, *args, **kwargs)\n```\n\n*history* is used to document the requests/responses that resulted in redirected requests.\n\n*args* are any additional args to be included with exception.\n\n*kwargs* are keyword arguments to include with the exception.\n\n#### StreamConsumedError\n\nAn error thrown if you try to access the stream of `azure.core.rest.HttpResponse` or `azure.core.rest.AsyncHttpResponse` once\nthe response stream has been consumed.\n\n#### StreamClosedError\n\nAn error thrown if you try to access the stream of the `azure.core.rest.HttpResponse` or `azure.core.rest.AsyncHttpResponse` once\nthe response stream has been closed.\n\n#### ResponseNotReadError\n\nAn error thrown if you try to access the `content` of `azure.core.rest.HttpResponse` or `azure.core.rest.AsyncHttpResponse` before\nreading in the response's bytes first.\n\n### Configurations\n\nWhen calling the methods, some properties can be configured by passing in as kwargs arguments.\n\n| Parameters | Description |\n| --- | --- |\n| headers | The HTTP Request headers. |\n| request_id | The request id to be added into header. |\n| user_agent | If specified, this will be added in front of the user agent string. |\n| logging_enable| Use to enable per operation. Defaults to `False`. |\n| logger | If specified, it will be used to log information. |\n| response_encoding | The encoding to use if known for this service (will disable auto-detection). |\n| raw_request_hook | Callback function. Will be invoked on request. |\n| raw_response_hook | Callback function. Will be invoked on response. |\n| network_span_namer | A callable to customize the span name. |\n| tracing_attributes | Attributes to set on all created spans. |\n| permit_redirects | Whether the client allows redirects. Defaults to `True`. |\n| redirect_max | The maximum allowed redirects. Defaults to `30`. |\n| retry_total | Total number of retries to allow. Takes precedence over other counts. Default value is `10`. |\n| retry_connect | How many connection-related errors to retry on. These are errors raised before the request is sent to the remote server, which we assume has not triggered the server to process the request. Default value is `3`. |\n| retry_read | How many times to retry on read errors. These errors are raised after the request was sent to the server, so the request may have side-effects. Default value is `3`. |\n| retry_status | How many times to retry on bad status codes. Default value is `3`. |\n| retry_backoff_factor | A backoff factor to apply between attempts after the second try (most errors are resolved immediately by a second try without a delay). Retry policy will sleep for: `{backoff factor} * (2 ** ({number of total retries} - 1))` seconds. If the backoff_factor is 0.1, then the retry will sleep for [0.0s, 0.2s, 0.4s, ...] between retries. The default value is `0.8`. |\n| retry_backoff_max | The maximum back off time. Default value is `120` seconds (2 minutes). |\n| retry_mode | Fixed or exponential delay between attempts, default is `Exponential`. |\n| timeout | Timeout setting for the operation in seconds, default is `604800`s (7 days). |\n| connection_timeout | A single float in seconds for the connection timeout. Defaults to `300` seconds. |\n| read_timeout | A single float in seconds for the read timeout. Defaults to `300` seconds. |\n| connection_verify | SSL certificate verification. Enabled by default. Set to False to disable, alternatively can be set to the path to a CA_BUNDLE file or directory with certificates of trusted CAs. |\n| connection_cert | Client-side certificates. You can specify a local cert to use as client side certificate, as a single file (containing the private key and the certificate) or as a tuple of both files' paths. |\n| proxies | Dictionary mapping protocol or protocol and hostname to the URL of the proxy. |\n| cookies | Dict or CookieJar object to send with the `Request`. |\n| connection_data_block_size | The block size of data sent over the connection. Defaults to `4096` bytes. |\n\n### Async transport\n\nThe async transport is designed to be opt-in. [AioHttp](https://pypi.org/project/aiohttp/) is one of the supported implementations of async transport. It is not installed by default. You need to install it separately.\n\n### Shared modules\n\n#### MatchConditions\n\nMatchConditions is an enum to describe match conditions.\n\n```python\nclass MatchConditions(Enum):\n Unconditionally = 1 # Matches any condition\n IfNotModified = 2 # If the target object is not modified. Usually it maps to etag=\n IfModified = 3 # Only if the target object is modified. Usually it maps to etag!=\n IfPresent = 4 # If the target object exists. Usually it maps to etag='*'\n IfMissing = 5 # If the target object does not exist. Usually it maps to etag!='*'\n```\n\n#### CaseInsensitiveEnumMeta\n\nA metaclass to support case-insensitive enums.\n\n```python\nfrom enum import Enum\n\nfrom azure.core import CaseInsensitiveEnumMeta\n\nclass MyCustomEnum(str, Enum, metaclass=CaseInsensitiveEnumMeta):\n FOO = 'foo'\n BAR = 'bar'\n```\n\n#### Null Sentinel Value\n\nA falsy sentinel object which is supposed to be used to specify attributes\nwith no data. This gets serialized to `null` on the wire.\n\n```python\nfrom azure.core.serialization import NULL\n\nassert bool(NULL) is False\n\nfoo = Foo(\n attr=NULL\n)\n```\n\n## Contributing\n\nThis project welcomes contributions and suggestions. Most contributions require\nyou to agree to a Contributor License Agreement (CLA) declaring that you have\nthe right to, and actually do, grant us the rights to use your contribution.\nFor details, visit [https://cla.microsoft.com](https://cla.microsoft.com).\n\nWhen you submit a pull request, a CLA-bot will automatically determine whether\nyou need to provide a CLA and decorate the PR appropriately (e.g., label,\ncomment). Simply follow the instructions provided by the bot. You will only\nneed to do this once across all repos using our CLA.\n\nThis project has adopted the\n[Microsoft Open Source Code of Conduct](https://opensource.microsoft.com/codeofconduct/).\nFor more information, see the\n[Code of Conduct FAQ](https://opensource.microsoft.com/codeofconduct/faq/)\nor contact [opencode@microsoft.com](mailto:opencode@microsoft.com) with any\nadditional questions or comments.\n\n\n[package]: https://pypi.org/project/azure-core/\n\n\n# Release History\n\n## 1.33.0 (2025-04-03)\n\n### Features Added\n\n- Added native OpenTelemetry tracing to Azure Core which enables users to use OpenTelemetry to trace Azure SDK operations without needing to install a plugin. #39563\n - To enable native OpenTelemetry tracing, users need to:\n 1. Have `opentelemetry-api` installed.\n 2. Ensure that `settings.tracing_implementation` is not set.\n 3. Ensure that `settings.tracing_enabled` is set to `True`.\n - If `setting.tracing_implementation` is set, the tracing plugin will be used instead of the native tracing.\n - If `settings.tracing_enabled` is set to `False`, tracing will be disabled.\n - The `OpenTelemetryTracer` class was added to the `azure.core.tracing.opentelemetry` module. This is a wrapper around the OpenTelemetry tracer that is used to create spans for Azure SDK operations.\n - Added a `get_tracer` method to the new `azure.core.instrumentation` module. This method returns an instance of the `OpenTelemetryTracer` class if OpenTelemetry is available.\n - A `TracingOptions` TypedDict class was added to define the options that SDK users can use to configure tracing per-operation. These options include the ability to enable or disable tracing and set additional attributes on spans.\n - Example usage: `client.method(tracing_options={\"enabled\": True, \"attributes\": {\"foo\": \"bar\"}})`\n - The `DistributedTracingPolicy` and `distributed_trace`/`distributed_trace_async` decorators now uses the OpenTelemetry tracer if it is available and native tracing is enabled.\n - SDK clients can define an `_instrumentation_config` class variable to configure the OpenTelemetry tracer used in method span creation. Possible configuration options are `library_name`, `library_version`, `schema_url`, and `attributes`.\n - `DistributedTracingPolicy` now accepts a `instrumentation_config` keyword argument to configure the OpenTelemetry tracer used in HTTP span creation.\n\n### Breaking Changes\n\n- Removed automatic tracing enablement for the OpenTelemetry plugin if `opentelemetry` was imported. To enable tracing with the plugin, please import `azure.core.settings.settings` and set `settings.tracing_implementation` to `\"opentelemetry\"`. #39563\n- In `DistributedTracingPolicy`, the default span name is now just the HTTP method (e.g., \"GET\", \"POST\") and no longer includes the URL path. This change was made to converge with the OpenTelemetry HTTP semantic conventions. The full URL is still included in the span attributes.\n- Renamed span attributes in `DistributedTracingPolicy`:\n - \"x-ms-client-request-id\" is now \"az.client_request_id\"\n - \"x-ms-request-id\" is now \"az.service_request_id\"\n\n### Bugs Fixed\n\n- Fixed an issue where the `traceparent` header was not being set correctly in the `DistributedTracingPolicy`. The `traceparent` header will now set based on the context of the HTTP client span. #40074\n\n### Other Changes\n\n- Added `opentelemetry-api` as an optional dependency for tracing. This can be installed with `pip install azure-core[tracing]`. #39563\n\n## 1.32.0 (2024-10-31)\n\n### Features Added\n\n- Added a default implementation to handle token challenges in `BearerTokenCredentialPolicy` and `AsyncBearerTokenCredentialPolicy`.\n\n### Bugs Fixed\n\n- Fixed an issue where the `tracing_attributes` keyword argument wasn't being handled at the request/method level. #38164\n\n### Other Changes\n\n- Log \"x-vss-e2eid\" and \"x-msedge-ref\" headers in `HttpLoggingPolicy`.\n\n## 1.31.0 (2024-09-12)\n\n### Features Added\n\n- Added azure.core.AzureClouds enum to represent the different Azure clouds.\n- Added two new credential protocol classes, `SupportsTokenInfo` and `AsyncSupportsTokenInfo`, to offer more extensibility in supporting various token acquisition scenarios. #36565\n - Each new protocol class defines a `get_token_info` method that returns an `AccessTokenInfo` object.\n- Added a new `TokenRequestOptions` class, which is a `TypedDict` with optional parameters, that can be used to define options for token requests through the `get_token_info` method. #36565\n- Added a new `AccessTokenInfo` class, which is returned by `get_token_info` implementations. This class contains the token, its expiration time, and optional additional information like when a token should be refreshed. #36565\n- `BearerTokenCredentialPolicy` and `AsyncBearerTokenCredentialPolicy` now first check if a credential has the `get_token_info` method defined. If so, the `get_token_info` method is used to acquire a token. Otherwise, the `get_token` method is used. #36565\n - These policies now also check the `refresh_on` attribute when determining if a new token request should be made.\n\n### Other Changes\n\n- The Azure Core OpenTelemetry tracing plugin will now be the preferred tracing plugin over the OpenCensus plugin. If both plugins are installed and `opentelemetry` is imported, then OpenTelemetry will be used to trace Azure SDK operations. #35050\n\n## 1.30.2 (2024-06-06)\n\n### Features Added\n\n- Tracing: `DistributedTracingPolicy` will now set an attribute, `http.request.resend_count`, on HTTP spans for resent requests to indicate the resend attempt number. #35069\n\n### Bugs Fixed\n\n- Raise correct exception if transport is used while already closed #35559\n\n### Other Changes\n\n- HTTP tracing spans will now include an `error.type` attribute if an error status code is returned. #34619\n- Minimum required Python version is now 3.8\n\n## 1.30.1 (2024-02-29)\n\n### Other Changes\n\n- Accept float for `retry_after` header. #34203\n\n## 1.30.0 (2024-02-01)\n\n### Features Added\n\n- Support tuple input for file values to `azure.core.rest.HttpRequest` #33948\n- Support tuple input to `files` with duplicate field names `azure.core.rest.HttpRequest` #34021\n\n## 1.29.7 (2024-01-18)\n\n### Other Changes\n\n- Removed dependency on `anyio`. #33282\n\n## 1.29.6 (2023-12-14)\n\n### Bugs Fixed\n\n- Adjusted `AsyncBearerTokenCredentialPolicy` to work properly with `trio` concurrency mechanisms. ([#33307](https://github.com/Azure/azure-sdk-for-python/pull/33307))\n\n### Other Changes\n\n- Added dependency on `anyio` >=3.0,<5.0\n- Bumped minimum dependency on `requests` to 2.21.0.\n\n## 1.29.5 (2023-10-19)\n\n### Bugs Fixed\n\n- Fixed an issue with `multipart/form-data` in the async transport where `data` was not getting encoded into the request body. #32473\n\n### Other Changes\n\n- Use ssl context from aiohttp by default.\n\n## 1.29.4 (2023-09-07)\n\n### Bugs Fixed\n\n- Fixed the issue that some urls trigger an infinite loop. #31346\n- Fixed issue where IndexError was raised if multipart responses did not match the number of requests. #31471\n- Fixed issue unbound variable exception if dict is invalid in CloudEvent.from_dict. #31835\n- Fixed issue asyncBearerTokenCredentialPolicy is not backward compatible with SansIOHTTPPolicy. #31836\n- Fixed issue mypy complains with new version of azure-core. #31564\n\n## 1.29.3 (2023-08-22)\n\n### Bugs Fixed\n\n- Typing fix: `message` cannot be `None` in `AzureError`. #31564\n\n## 1.29.2 (2023-08-14)\n\n### Bugs Fixed\n\n- Added a default implementation for `AsyncTokenCredential.__aexit__()` #31573\n\n### Other Changes\n\n- Bumped `typing-extensions` version to 4.6.0.\n\n## 1.29.1 (2023-08-09)\n\n### Bugs Fixed\n\n- Not pass `enabled_cae` unless it is explicitly enabled.\n\n## 1.29.0 (2023-08-03)\n\n### Features Added\n\n- A keyword argument `enable_cae` was added to the `get_token` method of the `TokenCredential` protocol. #31012\n- `BearerTokenCredentialPolicy` and `AsyncBearerTokenCredentialPolicy` now accept `enable_cae` keyword arguments in their constructors. This is used in determining if [Continuous Access Evaluation (CAE)](https://learn.microsoft.com/azure/active-directory/conditional-access/concept-continuous-access-evaluation) should be enabled for each `get_token` request. #31012\n\n## 1.28.0 (2023-07-06)\n\n### Features Added\n\n- Added header name parameter to `RequestIdPolicy`. #30772\n- Added `SensitiveHeaderCleanupPolicy` that cleans up sensitive headers if a redirect happens and the new destination is in another domain. #28349\n\n### Other Changes\n\n- Catch aiohttp errors and translate them into azure-core errors.\n\n## 1.27.1 (2023-06-13)\n\n### Bugs Fixed\n\n- Fix url building for some complex query parameters scenarios #30707\n\n## 1.27.0 (2023-06-01)\n\n### Features Added\n\n- Added support to use sync credentials in `AsyncBearerTokenCredentialPolicy`. #30381\n- Added \"prefix\" parameter to AzureKeyCredentialPolicy #29901\n\n### Bugs Fixed\n\n- Improve error message when providing the wrong credential type for AzureKeyCredential #30380\n\n## 1.26.4 (2023-04-06)\n\n### Features Added\n\n- Updated settings to include OpenTelemetry as a tracer provider. #29095\n\n### Other Changes\n\n- Improved typing\n\n## 1.26.3 (2023-02-02)\n\n### Bugs Fixed\n\n- Fixed deflate decompression for aiohttp #28483\n\n## 1.26.2 (2023-01-05)\n\n### Bugs Fixed\n\n- Fix 'ClientSession' object has no attribute 'auto_decompress' (thanks to @mghextreme for the contribution)\n\n### Other Changes\n\n- Add \"x-ms-error-code\" as secure header to log\n- Rename \"DEFAULT_HEADERS_WHITELIST\" to \"DEFAULT_HEADERS_ALLOWLIST\". Added a backward compatible alias.\n\n## 1.26.1 (2022-11-03)\n\n### Other Changes\n\n- Added example of RequestsTransport with custom session. (thanks to @inirudebwoy for the contribution) #26768\n- Added Python 3.11 support.\n\n## 1.26.0 (2022-10-06)\n\n### Other Changes\n\n- LRO polling will not wait anymore before doing the first status check #26376\n- Added extra dependency for [aio]. pip install azure-core[aio] installs aiohttp too.\n\n## 1.25.1 (2022-09-01)\n\n### Bugs Fixed\n\n- Added @runtime_checkable to `TokenCredential` protocol definitions #25187\n\n## 1.25.0 (2022-08-04)\n\nAzure-core is supported on Python 3.7 or later. For more details, please read our page on [Azure SDK for Python version support policy](https://github.com/Azure/azure-sdk-for-python/wiki/Azure-SDKs-Python-version-support-policy).\n\n### Features Added\n\n- Added `CaseInsensitiveDict` implementation in `azure.core.utils` removing dependency on `requests` and `aiohttp`\n\n## 1.24.2 (2022-06-30)\n\n### Bugs Fixed\n\n- Fixed the bug that azure-core could not be imported under Python 3.11.0b3 #24928\n- `ContentDecodePolicy` can now correctly deserialize more JSON bodies with different mime types #22410\n\n## 1.24.1 (2022-06-01)\n\n### Bugs Fixed\n\n- Declare method level span as INTERNAL by default #24492\n- Fixed type hints for `azure.core.paging.ItemPaged` #24548\n\n## 1.24.0 (2022-05-06)\n\n### Features Added\n\n- Add `SerializationError` and `DeserializationError` in `azure.core.exceptions` for errors raised during serialization / deserialization #24312\n\n## 1.23.1 (2022-03-31)\n\n### Bugs Fixed\n\n- Allow stream inputs to the `content` kwarg of `azure.core.rest.HttpRequest` from objects with a `read` method #23578\n\n## 1.23.0 (2022-03-03)\n\n### Features Added\n\n- Improve intellisense type hinting for service client methods. #22891\n\n- Add a case insensitive dict `case_insensitive_dict` in `azure.core.utils`. #23206\n\n### Bugs Fixed\n\n- Use \"\\n\" rather than \"/n\" for new line in log. #23261\n\n### Other Changes\n\n- Log \"WWW-Authenticate\" header in `HttpLoggingPolicy` #22990\n- Added dependency on `typing-extensions` >= 4.0.1\n\n## 1.22.1 (2022-02-09)\n\n### Bugs Fixed\n\n- Limiting `final-state-via` scope to POST until consuming SDKs has been fixed to use this option properly on PUT. #22989\n\n## 1.22.0 (2022-02-03)\n_[**This version is deprecated.**]_\n\n### Features Added\n\n- Add support for `final-state-via` LRO option in core. #22713\n\n### Bugs Fixed\n\n- Add response body to string representation of `HttpResponseError` if we're not able to parse out information #22302\n- Raise `AttributeError` when calling azure.core.pipeline.transport.\\_\\_bases__ #22469\n\n### Other Changes\n\n- Python 2.7 is no longer supported. Please use Python version 3.6 or later.\n\n## 1.21.1 (2021-12-06)\n\n### Other Changes\n\n- Revert change in str method #22023\n\n## 1.21.0 (2021-12-02)\n\n### Breaking Changes\n\n- Sync stream downloading now raises `azure.core.exceptions.DecodeError` rather than `requests.exceptions.ContentDecodingError`\n\n### Bugs Fixed\n\n- Add response body to string representation of `HttpResponseError` if we're not able to parse out information #21800\n\n## 1.20.1 (2021-11-08)\n\n### Bugs Fixed\n\n- Correctly set response's content to decompressed body when users are using aiohttp transport with decompression headers #21620\n\n## 1.20.0 (2021-11-04)\n\n### Features Added\n\n- GA `send_request` onto the `azure.core.PipelineClient` and `azure.core.AsyncPipelineClient`. This method takes in\nrequests and sends them through our pipelines.\n- GA `azure.core.rest`. `azure.core.rest` is our new public simple HTTP library in `azure.core` that users will use to create requests, and consume responses.\n- GA errors `StreamConsumedError`, `StreamClosedError`, and `ResponseNotReadError` to `azure.core.exceptions`. These errors\nare thrown if you mishandle streamed responses from the `azure.core.rest` module\n- add kwargs to the methods for `iter_raw` and `iter_bytes` #21529\n- no longer raise JSON errors if users pass in file descriptors of JSON to the `json` kwarg in `HttpRequest` #21504\n- Added new error type `IncompleteReadError` which is raised if peer closes the connection before we have received the complete message body.\n\n### Breaking Changes\n\n- SansIOHTTPPolicy.on_exception returns None instead of bool.\n\n### Bugs Fixed\n\n- The `Content-Length` header in a http response is strictly checked against the actual number of bytes in the body,\n rather than silently truncating data in case the underlying tcp connection is closed prematurely.\n (thanks to @jochen-ott-by for the contribution) #20412\n- UnboundLocalError when SansIOHTTPPolicy handles an exception #15222\n- Add default content type header of `text/plain` and content length header for users who pass unicode strings to the `content` kwarg of `HttpRequest` in 2.7 #21550\n\n## 1.19.1 (2021-11-01)\n\n### Bugs Fixed\n\n- respect text encoding specified in argument (thanks to @ryohji for the contribution) #20796\n- Fix \"coroutine x.read() was never awaited\" warning from `ContentDecodePolicy` #21318\n- fix type check for `data` input to `azure.core.rest` for python 2.7 users #21341\n- use `charset_normalizer` if `chardet` is not installed to migrate aiohttp 3.8.0 changes.\n\n### Other Changes\n\n- Refactor AzureJSONEncoder (thanks to @Codejune for the contribution) #21028\n\n## 1.19.0 (2021-09-30)\n\n### Breaking Changes in the Provisional `azure.core.rest` package\n\n- `azure.core.rest.HttpResponse` and `azure.core.rest.AsyncHttpResponse` are now abstract base classes. They should not be initialized directly, instead\nyour transport responses should inherit from them and implement them.\n- The properties of the `azure.core.rest` responses are now all read-only\n\n- HttpLoggingPolicy integrates logs into one record #19925\n\n## 1.18.0 (2021-09-02)\n\n### Features Added\n\n- `azure.core.serialization.AzureJSONEncoder` (introduced in 1.17.0) serializes `datetime.datetime` objects in ISO 8601 format, conforming to RFC 3339's specification. #20190\n- We now use `azure.core.serialization.AzureJSONEncoder` to serialize `json` input to `azure.core.rest.HttpRequest`.\n\n### Breaking Changes in the Provisional `azure.core.rest` package\n\n- The `text` property on `azure.core.rest.HttpResponse` and `azure.core.rest.AsyncHttpResponse` has changed to a method, which also takes\nan `encoding` parameter.\n- Removed `iter_text` and `iter_lines` from `azure.core.rest.HttpResponse` and `azure.core.rest.AsyncHttpResponse`\n\n### Bugs Fixed\n\n- The behaviour of the headers returned in `azure.core.rest` responses now aligns across sync and async. Items can now be checked case-insensitively and without raising an error for format.\n\n## 1.17.0 (2021-08-05)\n\n### Features Added\n\n- Cut hard dependency on requests library\n- Added a `from_json` method which now accepts storage QueueMessage, eventhub's EventData or ServiceBusMessage or simply json bytes to return a `CloudEvent`\n\n### Fixed\n\n- Not override \"x-ms-client-request-id\" if it already exists in the header. #17757\n\n### Breaking Changes in the Provisional `azure.core.rest` package\n\n- `azure.core.rest` will not try to guess the `charset` anymore if it was impossible to extract it from `HttpResponse` analysis. This removes our dependency on `charset`.\n\n## 1.16.0 (2021-07-01)\n\n### Features Added\n\n- Add new ***provisional*** methods `send_request` onto the `azure.core.PipelineClient` and `azure.core.AsyncPipelineClient`. This method takes in\nrequests and sends them through our pipelines.\n- Add new ***provisional*** module `azure.core.rest`. `azure.core.rest` is our new public simple HTTP library in `azure.core` that users will use to create requests, and consume responses.\n- Add new ***provisional*** errors `StreamConsumedError`, `StreamClosedError`, and `ResponseNotReadError` to `azure.core.exceptions`. These errors\nare thrown if you mishandle streamed responses from the provisional `azure.core.rest` module\n\n### Fixed\n\n- Improved error message in the `from_dict` method of `CloudEvent` when a wrong schema is sent.\n\n## 1.15.0 (2021-06-04)\n\n### New Features\n\n- Added `BearerTokenCredentialPolicy.on_challenge` and `.authorize_request` to allow subclasses to optionally handle authentication challenges\n\n### Bug Fixes\n\n- Retry policies don't sleep after operations time out\n- The `from_dict` methhod in the `CloudEvent` can now convert a datetime string to datetime object when microsecond exceeds the python limitation\n\n## 1.14.0 (2021-05-13)\n\n### New Features\n\n- Added `azure.core.credentials.AzureNamedKeyCredential` credential #17548.\n- Added `decompress` parameter for `stream_download` method. If it is set to `False`, will not do decompression upon the stream. #17920\n\n## 1.13.0 (2021-04-02)\n\nAzure core requires Python 2.7 or Python 3.6+ since this release.\n\n### New Features\n\n- Added `azure.core.utils.parse_connection_string` function to parse connection strings across SDKs, with common validation and support for case insensitive keys.\n- Supported adding custom policies #16519\n- Added `~azure.core.tracing.Link` that should be used while passing `Links` to `AbstractSpan`.\n- `AbstractSpan` constructor can now take in additional keyword only args.\n\n### Bug fixes\n\n- Make NetworkTraceLoggingPolicy show the auth token in plain text. #14191\n- Fixed RetryPolicy overriding default connection timeout with an extreme value #17481\n\n## 1.12.0 (2021-03-08)\n\nThis version will be the last version to officially support Python 3.5, future versions will require Python 2.7 or Python 3.6+.\n\n### Features\n\n- Added `azure.core.messaging.CloudEvent` model that follows the cloud event spec.\n- Added `azure.core.serialization.NULL` sentinel value\n- Improve `repr`s for `HttpRequest` and `HttpResponse`s #16972\n\n### Bug Fixes\n\n- Disable retry in stream downloading. (thanks to @jochen-ott-by @hoffmann for the contribution) #16723\n\n## 1.11.0 (2021-02-08)\n\n### Features\n\n- Added `CaseInsensitiveEnumMeta` class for case-insensitive enums. #16316\n- Add `raise_for_status` method onto `HttpResponse`. Calling `response.raise_for_status()` on a response with an error code\nwill raise an `HttpResponseError`. Calling it on a good response will do nothing #16399\n\n### Bug Fixes\n\n- Update conn.conn_kw rather than overriding it when setting block size. (thanks for @jiasli for the contribution) #16587\n\n## 1.10.0 (2021-01-11)\n\n### Features\n\n- Added `AzureSasCredential` and its respective policy. #15946\n\n## 1.9.0 (2020-11-09)\n\n### Features\n\n- Add a `continuation_token` attribute to the base `AzureError` exception, and set this value for errors raised\n during paged or long-running operations.\n\n### Bug Fixes\n\n- Set retry_interval to 1 second instead of 1000 seconds (thanks **vbarbaresi** for contributing) #14357\n\n\n## 1.8.2 (2020-10-05)\n\n### Bug Fixes\n\n- Fixed bug to allow polling in the case of parameterized endpoints with relative polling urls #14097\n\n\n## 1.8.1 (2020-09-08)\n\n### Bug fixes\n\n- SAS credential replicated \"/\" fix #13159\n\n## 1.8.0 (2020-08-10)\n\n### Features\n\n- Support params as list for exploding parameters #12410\n\n\n## 1.7.0 (2020-07-06)\n\n### Bug fixes\n\n- `AzureKeyCredentialPolicy` will now accept (and ignore) passed in kwargs #11963\n- Better error messages if passed endpoint is incorrect #12106\n- Do not JSON encore a string if content type is \"text\" #12137\n\n### Features\n\n- Added `http_logging_policy` property on the `Configuration` object, allowing users to individually\nset the http logging policy of the config #12218\n\n## 1.6.0 (2020-06-03)\n\n### Bug fixes\n\n- Fixed deadlocks in AsyncBearerTokenCredentialPolicy #11543\n- Fix AttributeException in StreamDownloadGenerator #11462\n\n### Features\n\n- Added support for changesets as part of multipart message support #10485\n- Add AsyncLROPoller in azure.core.polling #10801\n- Add get_continuation_token/from_continuation_token/polling_method methods in pollers (sync and async) #10801\n- HttpResponse and PipelineContext objects are now pickable #10801\n\n## 1.5.0 (2020-05-04)\n\n### Features\n\n- Support \"x-ms-retry-after-ms\" in response header #10743\n- `link` and `link_from_headers` now accepts attributes #10765\n\n### Bug fixes\n\n- Not retry if the status code is less than 400 #10778\n- \"x-ms-request-id\" is not considered safe header for logging #10967\n\n## 1.4.0 (2020-04-06)\n\n### Features\n\n- Support a default error type in map_error #9773\n- Added `AzureKeyCredential` and its respective policy. #10509\n- Added `azure.core.polling.base_polling` module with a \"Microsoft One API\" polling implementation #10090\n Also contains the async version in `azure.core.polling.async_base_polling`\n- Support kwarg `enforce_https` to disable HTTPS check on authentication #9821\n- Support additional kwargs in `HttpRequest.set_multipart_mixed` that will be passed into pipeline context.\n\n## 1.3.0 (2020-03-09)\n\n### Bug fixes\n\n- Appended RequestIdPolicy to the default pipeline #9841\n- Rewind the body position in async_retry #10117\n\n### Features\n\n- Add raw_request_hook support in custom_hook_policy #9958\n- Add timeout support in retry_policy #10011\n- Add OdataV4 error format auto-parsing in all exceptions ('error' attribute) #9738\n\n## 1.2.2 (2020-02-10)\n\n### Bug fixes\n\n- Fixed a bug that sends None as request_id #9545\n- Enable mypy for customers #9572\n- Handle TypeError in deep copy #9620\n- Fix text/plain content-type in decoder #9589\n\n## 1.2.1 (2020-01-14)\n\n### Bug fixes\n\n- Fixed a regression in 1.2.0 that was incompatible with azure-keyvault-* 4.0.0\n[#9462](https://github.com/Azure/azure-sdk-for-python/issues/9462)\n\n\n## 1.2.0 (2020-01-14)\n\n### Features\n\n- Add user_agent & sdk_moniker kwargs in UserAgentPolicy init #9355\n- Support OPTIONS HTTP verb #9322\n- Add tracing_attributes to tracing decorator #9297\n- Support auto_request_id in RequestIdPolicy #9163\n- Support fixed retry #6419\n- Support \"retry-after-ms\" in response header #9240\n\n### Bug fixes\n\n- Removed `__enter__` and `__exit__` from async context managers #9313\n\n## 1.1.1 (2019-12-03)\n\n### Bug fixes\n\n- Bearer token authorization requires HTTPS\n- Rewind the body position in retry #8307\n\n## 1.1.0 (2019-11-25)\n\n### Features\n\n- New RequestIdPolicy #8437\n- Enable logging policy in default pipeline #8053\n- Normalize transport timeout. #8000\n Now we have:\n * 'connection_timeout' - a single float in seconds for the connection timeout. Default 5min\n * 'read_timeout' - a single float in seconds for the read timeout. Default 5min\n\n### Bug fixes\n\n- RequestHistory: deepcopy fails if request contains a stream #7732\n- Retry: retry raises error if response does not have http_response #8629\n- Client kwargs are now passed to DistributedTracingPolicy correctly #8051\n- NetworkLoggingPolicy now logs correctly all requests in case of retry #8262\n\n## 1.0.0 (2019-10-29)\n\n### Features\n\n- Tracing: DistributedTracingPolicy now accepts kwargs network_span_namer to change network span name #7773\n- Tracing: Implementation of AbstractSpan can now use the mixin HttpSpanMixin to get HTTP span update automatically #7773\n- Tracing: AbstractSpan contract \"change_context\" introduced #7773\n- Introduce new policy HttpLoggingPolicy #7988\n\n### Bug fixes\n\n- Fix AsyncioRequestsTransport if input stream is an async generator #7743\n- Fix form-data with aiohttp transport #7749\n\n### Breaking changes\n\n- Tracing: AbstractSpan.set_current_span is longer supported. Use change_context instead. #7773\n- azure.core.pipeline.policies.ContentDecodePolicy.deserialize_from_text changed\n\n## 1.0.0b4 (2019-10-07)\n\n### Features\n\n- Tracing: network span context is available with the TRACING_CONTEXT in pipeline response #7252\n- Tracing: Span contract now has `kind`, `traceparent` and is a context manager #7252\n- SansIOHTTPPolicy methods can now be coroutines #7497\n- Add multipart/mixed support #7083:\n\n - HttpRequest now has a \"set_multipart_mixed\" method to set the parts of this request\n - HttpRequest now has a \"prepare_multipart_body\" method to build final body.\n - HttpResponse now has a \"parts\" method to return an iterator of parts\n - AsyncHttpResponse now has a \"parts\" methods to return an async iterator of parts\n - Note that multipart/mixed is a Python 3.x only feature\n\n### Bug fixes\n\n- Tracing: policy cannot fail the pipeline, even in the worst condition #7252\n- Tracing: policy pass correctly status message if exception #7252\n- Tracing: incorrect span if exception raised from decorated function #7133\n- Fixed urllib3 ConnectTimeoutError being raised by Requests during a socket timeout. Now this exception is caught and wrapped as a `ServiceRequestError` #7542\n\n### Breaking changes\n\n- Tracing: `azure.core.tracing.context` removed\n- Tracing: `azure.core.tracing.context.tracing_context.with_current_context` renamed to `azure.core.tracing.common.with_current_context` #7252\n- Tracing: `link` renamed `link_from_headers` and `link` takes now a string\n- Tracing: opencensus implementation has been moved to the package `azure-core-tracing-opencensus`\n- Some modules and classes that were importables from several different places have been removed:\n\n - `azure.core.HttpResponseError` is now only `azure.core.exceptions.HttpResponseError`\n - `azure.core.Configuration` is now only `azure.core.configuration.Configuration`\n - `azure.core.HttpRequest` is now only `azure.core.pipeline.transport.HttpRequest`\n - `azure.core.version` module has been removed. Use `azure.core.__version__` to get version number.\n - `azure.core.pipeline_client` has been removed. Import from `azure.core` instead.\n - `azure.core.pipeline_client_async` has been removed. Import from `azure.core` instead.\n - `azure.core.pipeline.base` has been removed. Import from `azure.core.pipeline` instead.\n - `azure.core.pipeline.base_async` has been removed. Import from `azure.core.pipeline` instead.\n - `azure.core.pipeline.policies.base` has been removed. Import from `azure.core.pipeline.policies` instead.\n - `azure.core.pipeline.policies.base_async` has been removed. Import from `azure.core.pipeline.policies` instead.\n - `azure.core.pipeline.policies.authentication` has been removed. Import from `azure.core.pipeline.policies` instead.\n - `azure.core.pipeline.policies.authentication_async` has been removed. Import from `azure.core.pipeline.policies` instead.\n - `azure.core.pipeline.policies.custom_hook` has been removed. Import from `azure.core.pipeline.policies` instead.\n - `azure.core.pipeline.policies.redirect` has been removed. Import from `azure.core.pipeline.policies` instead.\n - `azure.core.pipeline.policies.redirect_async` has been removed. Import from `azure.core.pipeline.policies` instead.\n - `azure.core.pipeline.policies.retry` has been removed. Import from `azure.core.pipeline.policies` instead.\n - `azure.core.pipeline.policies.retry_async` has been removed. Import from `azure.core.pipeline.policies` instead.\n - `azure.core.pipeline.policies.distributed_tracing` has been removed. Import from `azure.core.pipeline.policies` instead.\n - `azure.core.pipeline.policies.universal` has been removed. Import from `azure.core.pipeline.policies` instead.\n - `azure.core.tracing.abstract_span` has been removed. Import from `azure.core.tracing` instead.\n - `azure.core.pipeline.transport.base` has been removed. Import from `azure.core.pipeline.transport` instead.\n - `azure.core.pipeline.transport.base_async` has been removed. Import from `azure.core.pipeline.transport` instead.\n - `azure.core.pipeline.transport.requests_basic` has been removed. Import from `azure.core.pipeline.transport` instead.\n - `azure.core.pipeline.transport.requests_asyncio` has been removed. Import from `azure.core.pipeline.transport` instead.\n - `azure.core.pipeline.transport.requests_trio` has been removed. Import from `azure.core.pipeline.transport` instead.\n - `azure.core.pipeline.transport.aiohttp` has been removed. Import from `azure.core.pipeline.transport` instead.\n - `azure.core.polling.poller` has been removed. Import from `azure.core.polling` instead.\n - `azure.core.polling.async_poller` has been removed. Import from `azure.core.polling` instead.\n\n## 1.0.0b3 (2019-09-09)\n\n### Bug fixes\n\n- Fix aiohttp auto-headers #6992\n- Add tracing to policies module init #6951\n\n## 1.0.0b2 (2019-08-05)\n\n### Breaking changes\n\n- Transport classes don't take `config` parameter anymore (use kwargs instead) #6372\n- `azure.core.paging` has been completely refactored #6420\n- HttpResponse.content_type attribute is now a string (was a list) #6490\n- For `StreamDownloadGenerator` subclasses, `response` is now an `HttpResponse`, and not a transport response like `aiohttp.ClientResponse` or `requests.Response`. The transport response is available in `internal_response` attribute #6490\n\n### Bug fixes\n\n- aiohttp is not required to import async pipelines classes #6496\n- `AsyncioRequestsTransport.sleep` is now a coroutine as expected #6490\n- `RequestsTransport` is not tight to `ProxyPolicy` implementation details anymore #6372\n- `AiohttpTransport` does not raise on unexpected kwargs #6355\n\n### Features\n\n- New paging base classes that support `continuation_token` and `by_page()` #6420\n- Proxy support for `AiohttpTransport` #6372\n\n## 1.0.0b1 (2019-06-26)\n\n- Preview 1 release", - "release_date": "2025-04-03T23:51:03", + "description": "Microsoft Azure Core Library for Python\n# Azure Core shared client library for Python\n\nAzure core provides shared exceptions and modules for Python SDK client libraries.\nThese libraries follow the [Azure SDK Design Guidelines for Python](https://azure.github.io/azure-sdk/python/guidelines/index.html) .\n\nIf you are a client library developer, please reference [client library developer reference](https://github.com/Azure/azure-sdk-for-python/blob/main/sdk/core/azure-core/CLIENT_LIBRARY_DEVELOPER.md) for more information.\n\n[Source code](https://github.com/Azure/azure-sdk-for-python/blob/main/sdk/core/azure-core/)\n| [Package (Pypi)][package]\n| [Package (Conda)](https://anaconda.org/microsoft/azure-core/)\n| [API reference documentation](https://learn.microsoft.com/python/api/overview/azure/core-readme)\n\n## Getting started\n\nTypically, you will not need to install azure core;\nit will be installed when you install one of the client libraries using it.\nIn case you want to install it explicitly (to implement your own client library, for example),\nyou can find it [here](https://pypi.org/project/azure-core/).\n\n## Key concepts\n\n### Azure Core Library Exceptions\n\n#### AzureError\n\nAzureError is the base exception for all errors.\n\n```python\nclass AzureError(Exception):\n def __init__(self, message, *args, **kwargs):\n self.inner_exception = kwargs.get(\"error\")\n self.exc_type, self.exc_value, self.exc_traceback = sys.exc_info()\n self.exc_type = self.exc_type.__name__ if self.exc_type else type(self.inner_exception)\n self.exc_msg = \"{}, {}: {}\".format(message, self.exc_type, self.exc_value) # type: ignore\n self.message = str(message)\n self.continuation_token = kwargs.get(\"continuation_token\")\n super(AzureError, self).__init__(self.message, *args)\n```\n\n*message* is any message (str) to be associated with the exception.\n\n*args* are any additional args to be included with exception.\n\n*kwargs* are keyword arguments to include with the exception. Use the keyword *error* to pass in an internal exception and *continuation_token* for a token reference to continue an incomplete operation.\n\n**The following exceptions inherit from AzureError:**\n\n#### ServiceRequestError\n\nAn error occurred while attempt to make a request to the service. No request was sent.\n\n#### ServiceResponseError\n\nThe request was sent, but the client failed to understand the response.\nThe connection may have timed out. These errors can be retried for idempotent or safe operations.\n\n#### HttpResponseError\n\nA request was made, and a non-success status code was received from the service.\n\n```python\nclass HttpResponseError(AzureError):\n def __init__(self, message=None, response=None, **kwargs):\n self.reason = None\n self.response = response\n if response:\n self.reason = response.reason\n self.status_code = response.status_code\n self.error = self._parse_odata_body(ODataV4Format, response) # type: Optional[ODataV4Format]\n if self.error:\n message = str(self.error)\n else:\n message = message or \"Operation returned an invalid status '{}'\".format(\n self.reason\n )\n\n super(HttpResponseError, self).__init__(message=message, **kwargs)\n```\n\n*message* is the HTTP response error message (optional)\n\n*response* is the HTTP response (optional).\n\n*kwargs* are keyword arguments to include with the exception.\n\n**The following exceptions inherit from HttpResponseError:**\n\n#### DecodeError\n\nAn error raised during response de-serialization.\n\n#### IncompleteReadError\n\nAn error raised if peer closes the connection before we have received the complete message body.\n\n#### ResourceExistsError\n\nAn error response with status code 4xx. This will not be raised directly by the Azure core pipeline.\n\n#### ResourceNotFoundError\n\nAn error response, typically triggered by a 412 response (for update) or 404 (for get/post).\n\n#### ResourceModifiedError\n\nAn error response with status code 4xx, typically 412 Conflict. This will not be raised directly by the Azure core pipeline.\n\n#### ResourceNotModifiedError\n\nAn error response with status code 304. This will not be raised directly by the Azure core pipeline.\n\n#### ClientAuthenticationError\n\nAn error response with status code 4xx. This will not be raised directly by the Azure core pipeline.\n\n#### TooManyRedirectsError\n\nAn error raised when the maximum number of redirect attempts is reached. The maximum amount of redirects can be configured in the RedirectPolicy.\n\n```python\nclass TooManyRedirectsError(HttpResponseError):\n def __init__(self, history, *args, **kwargs):\n self.history = history\n message = \"Reached maximum redirect attempts.\"\n super(TooManyRedirectsError, self).__init__(message, *args, **kwargs)\n```\n\n*history* is used to document the requests/responses that resulted in redirected requests.\n\n*args* are any additional args to be included with exception.\n\n*kwargs* are keyword arguments to include with the exception.\n\n#### StreamConsumedError\n\nAn error thrown if you try to access the stream of `azure.core.rest.HttpResponse` or `azure.core.rest.AsyncHttpResponse` once\nthe response stream has been consumed.\n\n#### StreamClosedError\n\nAn error thrown if you try to access the stream of the `azure.core.rest.HttpResponse` or `azure.core.rest.AsyncHttpResponse` once\nthe response stream has been closed.\n\n#### ResponseNotReadError\n\nAn error thrown if you try to access the `content` of `azure.core.rest.HttpResponse` or `azure.core.rest.AsyncHttpResponse` before\nreading in the response's bytes first.\n\n### Configurations\n\nWhen calling the methods, some properties can be configured by passing in as kwargs arguments.\n\n| Parameters | Description |\n| --- | --- |\n| headers | The HTTP Request headers. |\n| request_id | The request id to be added into header. |\n| user_agent | If specified, this will be added in front of the user agent string. |\n| logging_enable| Use to enable per operation. Defaults to `False`. |\n| logger | If specified, it will be used to log information. |\n| response_encoding | The encoding to use if known for this service (will disable auto-detection). |\n| raw_request_hook | Callback function. Will be invoked on request. |\n| raw_response_hook | Callback function. Will be invoked on response. |\n| network_span_namer | A callable to customize the span name. |\n| tracing_attributes | Attributes to set on all created spans. |\n| permit_redirects | Whether the client allows redirects. Defaults to `True`. |\n| redirect_max | The maximum allowed redirects. Defaults to `30`. |\n| retry_total | Total number of retries to allow. Takes precedence over other counts. Default value is `10`. |\n| retry_connect | How many connection-related errors to retry on. These are errors raised before the request is sent to the remote server, which we assume has not triggered the server to process the request. Default value is `3`. |\n| retry_read | How many times to retry on read errors. These errors are raised after the request was sent to the server, so the request may have side-effects. Default value is `3`. |\n| retry_status | How many times to retry on bad status codes. Default value is `3`. |\n| retry_backoff_factor | A backoff factor to apply between attempts after the second try (most errors are resolved immediately by a second try without a delay). Retry policy will sleep for: `{backoff factor} * (2 ** ({number of total retries} - 1))` seconds. If the backoff_factor is 0.1, then the retry will sleep for [0.0s, 0.2s, 0.4s, ...] between retries. The default value is `0.8`. |\n| retry_backoff_max | The maximum back off time. Default value is `120` seconds (2 minutes). |\n| retry_mode | Fixed or exponential delay between attempts, default is `Exponential`. |\n| timeout | Timeout setting for the operation in seconds, default is `604800`s (7 days). |\n| connection_timeout | A single float in seconds for the connection timeout. Defaults to `300` seconds. |\n| read_timeout | A single float in seconds for the read timeout. Defaults to `300` seconds. |\n| connection_verify | SSL certificate verification. Enabled by default. Set to False to disable, alternatively can be set to the path to a CA_BUNDLE file or directory with certificates of trusted CAs. |\n| connection_cert | Client-side certificates. You can specify a local cert to use as client side certificate, as a single file (containing the private key and the certificate) or as a tuple of both files' paths. |\n| proxies | Dictionary mapping protocol or protocol and hostname to the URL of the proxy. |\n| cookies | Dict or CookieJar object to send with the `Request`. |\n| connection_data_block_size | The block size of data sent over the connection. Defaults to `4096` bytes. |\n\n### Async transport\n\nThe async transport is designed to be opt-in. [AioHttp](https://pypi.org/project/aiohttp/) is one of the supported implementations of async transport. It is not installed by default. You need to install it separately.\n\n### Shared modules\n\n#### MatchConditions\n\nMatchConditions is an enum to describe match conditions.\n\n```python\nclass MatchConditions(Enum):\n Unconditionally = 1 # Matches any condition\n IfNotModified = 2 # If the target object is not modified. Usually it maps to etag=\n IfModified = 3 # Only if the target object is modified. Usually it maps to etag!=\n IfPresent = 4 # If the target object exists. Usually it maps to etag='*'\n IfMissing = 5 # If the target object does not exist. Usually it maps to etag!='*'\n```\n\n#### CaseInsensitiveEnumMeta\n\nA metaclass to support case-insensitive enums.\n\n```python\nfrom enum import Enum\n\nfrom azure.core import CaseInsensitiveEnumMeta\n\nclass MyCustomEnum(str, Enum, metaclass=CaseInsensitiveEnumMeta):\n FOO = 'foo'\n BAR = 'bar'\n```\n\n#### Null Sentinel Value\n\nA falsy sentinel object which is supposed to be used to specify attributes\nwith no data. This gets serialized to `null` on the wire.\n\n```python\nfrom azure.core.serialization import NULL\n\nassert bool(NULL) is False\n\nfoo = Foo(\n attr=NULL\n)\n```\n\n## Contributing\n\nThis project welcomes contributions and suggestions. Most contributions require\nyou to agree to a Contributor License Agreement (CLA) declaring that you have\nthe right to, and actually do, grant us the rights to use your contribution.\nFor details, visit [https://cla.microsoft.com](https://cla.microsoft.com).\n\nWhen you submit a pull request, a CLA-bot will automatically determine whether\nyou need to provide a CLA and decorate the PR appropriately (e.g., label,\ncomment). Simply follow the instructions provided by the bot. You will only\nneed to do this once across all repos using our CLA.\n\nThis project has adopted the\n[Microsoft Open Source Code of Conduct](https://opensource.microsoft.com/codeofconduct/).\nFor more information, see the\n[Code of Conduct FAQ](https://opensource.microsoft.com/codeofconduct/faq/)\nor contact [opencode@microsoft.com](mailto:opencode@microsoft.com) with any\nadditional questions or comments.\n\n\n[package]: https://pypi.org/project/azure-core/\n\n\n# Release History\n\n## 1.34.0 (2025-05-01)\n\n### Features Added\n\n- Added a `set_span_error_status` method to the `OpenTelemetryTracer` class. This method allows users to set the status of a span to `ERROR` after it has been created. #40703\n\n### Other Changes\n\n- Python 3.8 is no longer supported. Please use Python version 3.9 or later.\n\n## 1.33.0 (2025-04-03)\n\n### Features Added\n\n- Added native OpenTelemetry tracing to Azure Core which enables users to use OpenTelemetry to trace Azure SDK operations without needing to install a plugin. #39563\n - To enable native OpenTelemetry tracing, users need to:\n 1. Have `opentelemetry-api` installed.\n 2. Ensure that `settings.tracing_implementation` is not set.\n 3. Ensure that `settings.tracing_enabled` is set to `True`.\n - If `setting.tracing_implementation` is set, the tracing plugin will be used instead of the native tracing.\n - If `settings.tracing_enabled` is set to `False`, tracing will be disabled.\n - The `OpenTelemetryTracer` class was added to the `azure.core.tracing.opentelemetry` module. This is a wrapper around the OpenTelemetry tracer that is used to create spans for Azure SDK operations.\n - Added a `get_tracer` method to the new `azure.core.instrumentation` module. This method returns an instance of the `OpenTelemetryTracer` class if OpenTelemetry is available.\n - A `TracingOptions` TypedDict class was added to define the options that SDK users can use to configure tracing per-operation. These options include the ability to enable or disable tracing and set additional attributes on spans.\n - Example usage: `client.method(tracing_options={\"enabled\": True, \"attributes\": {\"foo\": \"bar\"}})`\n - The `DistributedTracingPolicy` and `distributed_trace`/`distributed_trace_async` decorators now uses the OpenTelemetry tracer if it is available and native tracing is enabled.\n - SDK clients can define an `_instrumentation_config` class variable to configure the OpenTelemetry tracer used in method span creation. Possible configuration options are `library_name`, `library_version`, `schema_url`, and `attributes`.\n - `DistributedTracingPolicy` now accepts a `instrumentation_config` keyword argument to configure the OpenTelemetry tracer used in HTTP span creation.\n\n### Breaking Changes\n\n- Removed automatic tracing enablement for the OpenTelemetry plugin if `opentelemetry` was imported. To enable tracing with the plugin, please import `azure.core.settings.settings` and set `settings.tracing_implementation` to `\"opentelemetry\"`. #39563\n- In `DistributedTracingPolicy`, the default span name is now just the HTTP method (e.g., \"GET\", \"POST\") and no longer includes the URL path. This change was made to converge with the OpenTelemetry HTTP semantic conventions. The full URL is still included in the span attributes.\n- Renamed span attributes in `DistributedTracingPolicy`:\n - \"x-ms-client-request-id\" is now \"az.client_request_id\"\n - \"x-ms-request-id\" is now \"az.service_request_id\"\n\n### Bugs Fixed\n\n- Fixed an issue where the `traceparent` header was not being set correctly in the `DistributedTracingPolicy`. The `traceparent` header will now set based on the context of the HTTP client span. #40074\n\n### Other Changes\n\n- Added `opentelemetry-api` as an optional dependency for tracing. This can be installed with `pip install azure-core[tracing]`. #39563\n\n## 1.32.0 (2024-10-31)\n\n### Features Added\n\n- Added a default implementation to handle token challenges in `BearerTokenCredentialPolicy` and `AsyncBearerTokenCredentialPolicy`.\n\n### Bugs Fixed\n\n- Fixed an issue where the `tracing_attributes` keyword argument wasn't being handled at the request/method level. #38164\n\n### Other Changes\n\n- Log \"x-vss-e2eid\" and \"x-msedge-ref\" headers in `HttpLoggingPolicy`.\n\n## 1.31.0 (2024-09-12)\n\n### Features Added\n\n- Added azure.core.AzureClouds enum to represent the different Azure clouds.\n- Added two new credential protocol classes, `SupportsTokenInfo` and `AsyncSupportsTokenInfo`, to offer more extensibility in supporting various token acquisition scenarios. #36565\n - Each new protocol class defines a `get_token_info` method that returns an `AccessTokenInfo` object.\n- Added a new `TokenRequestOptions` class, which is a `TypedDict` with optional parameters, that can be used to define options for token requests through the `get_token_info` method. #36565\n- Added a new `AccessTokenInfo` class, which is returned by `get_token_info` implementations. This class contains the token, its expiration time, and optional additional information like when a token should be refreshed. #36565\n- `BearerTokenCredentialPolicy` and `AsyncBearerTokenCredentialPolicy` now first check if a credential has the `get_token_info` method defined. If so, the `get_token_info` method is used to acquire a token. Otherwise, the `get_token` method is used. #36565\n - These policies now also check the `refresh_on` attribute when determining if a new token request should be made.\n\n### Other Changes\n\n- The Azure Core OpenTelemetry tracing plugin will now be the preferred tracing plugin over the OpenCensus plugin. If both plugins are installed and `opentelemetry` is imported, then OpenTelemetry will be used to trace Azure SDK operations. #35050\n\n## 1.30.2 (2024-06-06)\n\n### Features Added\n\n- Tracing: `DistributedTracingPolicy` will now set an attribute, `http.request.resend_count`, on HTTP spans for resent requests to indicate the resend attempt number. #35069\n\n### Bugs Fixed\n\n- Raise correct exception if transport is used while already closed #35559\n\n### Other Changes\n\n- HTTP tracing spans will now include an `error.type` attribute if an error status code is returned. #34619\n- Minimum required Python version is now 3.8\n\n## 1.30.1 (2024-02-29)\n\n### Other Changes\n\n- Accept float for `retry_after` header. #34203\n\n## 1.30.0 (2024-02-01)\n\n### Features Added\n\n- Support tuple input for file values to `azure.core.rest.HttpRequest` #33948\n- Support tuple input to `files` with duplicate field names `azure.core.rest.HttpRequest` #34021\n\n## 1.29.7 (2024-01-18)\n\n### Other Changes\n\n- Removed dependency on `anyio`. #33282\n\n## 1.29.6 (2023-12-14)\n\n### Bugs Fixed\n\n- Adjusted `AsyncBearerTokenCredentialPolicy` to work properly with `trio` concurrency mechanisms. ([#33307](https://github.com/Azure/azure-sdk-for-python/pull/33307))\n\n### Other Changes\n\n- Added dependency on `anyio` >=3.0,<5.0\n- Bumped minimum dependency on `requests` to 2.21.0.\n\n## 1.29.5 (2023-10-19)\n\n### Bugs Fixed\n\n- Fixed an issue with `multipart/form-data` in the async transport where `data` was not getting encoded into the request body. #32473\n\n### Other Changes\n\n- Use ssl context from aiohttp by default.\n\n## 1.29.4 (2023-09-07)\n\n### Bugs Fixed\n\n- Fixed the issue that some urls trigger an infinite loop. #31346\n- Fixed issue where IndexError was raised if multipart responses did not match the number of requests. #31471\n- Fixed issue unbound variable exception if dict is invalid in CloudEvent.from_dict. #31835\n- Fixed issue asyncBearerTokenCredentialPolicy is not backward compatible with SansIOHTTPPolicy. #31836\n- Fixed issue mypy complains with new version of azure-core. #31564\n\n## 1.29.3 (2023-08-22)\n\n### Bugs Fixed\n\n- Typing fix: `message` cannot be `None` in `AzureError`. #31564\n\n## 1.29.2 (2023-08-14)\n\n### Bugs Fixed\n\n- Added a default implementation for `AsyncTokenCredential.__aexit__()` #31573\n\n### Other Changes\n\n- Bumped `typing-extensions` version to 4.6.0.\n\n## 1.29.1 (2023-08-09)\n\n### Bugs Fixed\n\n- Not pass `enabled_cae` unless it is explicitly enabled.\n\n## 1.29.0 (2023-08-03)\n\n### Features Added\n\n- A keyword argument `enable_cae` was added to the `get_token` method of the `TokenCredential` protocol. #31012\n- `BearerTokenCredentialPolicy` and `AsyncBearerTokenCredentialPolicy` now accept `enable_cae` keyword arguments in their constructors. This is used in determining if [Continuous Access Evaluation (CAE)](https://learn.microsoft.com/azure/active-directory/conditional-access/concept-continuous-access-evaluation) should be enabled for each `get_token` request. #31012\n\n## 1.28.0 (2023-07-06)\n\n### Features Added\n\n- Added header name parameter to `RequestIdPolicy`. #30772\n- Added `SensitiveHeaderCleanupPolicy` that cleans up sensitive headers if a redirect happens and the new destination is in another domain. #28349\n\n### Other Changes\n\n- Catch aiohttp errors and translate them into azure-core errors.\n\n## 1.27.1 (2023-06-13)\n\n### Bugs Fixed\n\n- Fix url building for some complex query parameters scenarios #30707\n\n## 1.27.0 (2023-06-01)\n\n### Features Added\n\n- Added support to use sync credentials in `AsyncBearerTokenCredentialPolicy`. #30381\n- Added \"prefix\" parameter to AzureKeyCredentialPolicy #29901\n\n### Bugs Fixed\n\n- Improve error message when providing the wrong credential type for AzureKeyCredential #30380\n\n## 1.26.4 (2023-04-06)\n\n### Features Added\n\n- Updated settings to include OpenTelemetry as a tracer provider. #29095\n\n### Other Changes\n\n- Improved typing\n\n## 1.26.3 (2023-02-02)\n\n### Bugs Fixed\n\n- Fixed deflate decompression for aiohttp #28483\n\n## 1.26.2 (2023-01-05)\n\n### Bugs Fixed\n\n- Fix 'ClientSession' object has no attribute 'auto_decompress' (thanks to @mghextreme for the contribution)\n\n### Other Changes\n\n- Add \"x-ms-error-code\" as secure header to log\n- Rename \"DEFAULT_HEADERS_WHITELIST\" to \"DEFAULT_HEADERS_ALLOWLIST\". Added a backward compatible alias.\n\n## 1.26.1 (2022-11-03)\n\n### Other Changes\n\n- Added example of RequestsTransport with custom session. (thanks to @inirudebwoy for the contribution) #26768\n- Added Python 3.11 support.\n\n## 1.26.0 (2022-10-06)\n\n### Other Changes\n\n- LRO polling will not wait anymore before doing the first status check #26376\n- Added extra dependency for [aio]. pip install azure-core[aio] installs aiohttp too.\n\n## 1.25.1 (2022-09-01)\n\n### Bugs Fixed\n\n- Added @runtime_checkable to `TokenCredential` protocol definitions #25187\n\n## 1.25.0 (2022-08-04)\n\nAzure-core is supported on Python 3.7 or later. For more details, please read our page on [Azure SDK for Python version support policy](https://github.com/Azure/azure-sdk-for-python/wiki/Azure-SDKs-Python-version-support-policy).\n\n### Features Added\n\n- Added `CaseInsensitiveDict` implementation in `azure.core.utils` removing dependency on `requests` and `aiohttp`\n\n## 1.24.2 (2022-06-30)\n\n### Bugs Fixed\n\n- Fixed the bug that azure-core could not be imported under Python 3.11.0b3 #24928\n- `ContentDecodePolicy` can now correctly deserialize more JSON bodies with different mime types #22410\n\n## 1.24.1 (2022-06-01)\n\n### Bugs Fixed\n\n- Declare method level span as INTERNAL by default #24492\n- Fixed type hints for `azure.core.paging.ItemPaged` #24548\n\n## 1.24.0 (2022-05-06)\n\n### Features Added\n\n- Add `SerializationError` and `DeserializationError` in `azure.core.exceptions` for errors raised during serialization / deserialization #24312\n\n## 1.23.1 (2022-03-31)\n\n### Bugs Fixed\n\n- Allow stream inputs to the `content` kwarg of `azure.core.rest.HttpRequest` from objects with a `read` method #23578\n\n## 1.23.0 (2022-03-03)\n\n### Features Added\n\n- Improve intellisense type hinting for service client methods. #22891\n\n- Add a case insensitive dict `case_insensitive_dict` in `azure.core.utils`. #23206\n\n### Bugs Fixed\n\n- Use \"\\n\" rather than \"/n\" for new line in log. #23261\n\n### Other Changes\n\n- Log \"WWW-Authenticate\" header in `HttpLoggingPolicy` #22990\n- Added dependency on `typing-extensions` >= 4.0.1\n\n## 1.22.1 (2022-02-09)\n\n### Bugs Fixed\n\n- Limiting `final-state-via` scope to POST until consuming SDKs has been fixed to use this option properly on PUT. #22989\n\n## 1.22.0 (2022-02-03)\n_[**This version is deprecated.**]_\n\n### Features Added\n\n- Add support for `final-state-via` LRO option in core. #22713\n\n### Bugs Fixed\n\n- Add response body to string representation of `HttpResponseError` if we're not able to parse out information #22302\n- Raise `AttributeError` when calling azure.core.pipeline.transport.\\_\\_bases__ #22469\n\n### Other Changes\n\n- Python 2.7 is no longer supported. Please use Python version 3.6 or later.\n\n## 1.21.1 (2021-12-06)\n\n### Other Changes\n\n- Revert change in str method #22023\n\n## 1.21.0 (2021-12-02)\n\n### Breaking Changes\n\n- Sync stream downloading now raises `azure.core.exceptions.DecodeError` rather than `requests.exceptions.ContentDecodingError`\n\n### Bugs Fixed\n\n- Add response body to string representation of `HttpResponseError` if we're not able to parse out information #21800\n\n## 1.20.1 (2021-11-08)\n\n### Bugs Fixed\n\n- Correctly set response's content to decompressed body when users are using aiohttp transport with decompression headers #21620\n\n## 1.20.0 (2021-11-04)\n\n### Features Added\n\n- GA `send_request` onto the `azure.core.PipelineClient` and `azure.core.AsyncPipelineClient`. This method takes in\nrequests and sends them through our pipelines.\n- GA `azure.core.rest`. `azure.core.rest` is our new public simple HTTP library in `azure.core` that users will use to create requests, and consume responses.\n- GA errors `StreamConsumedError`, `StreamClosedError`, and `ResponseNotReadError` to `azure.core.exceptions`. These errors\nare thrown if you mishandle streamed responses from the `azure.core.rest` module\n- add kwargs to the methods for `iter_raw` and `iter_bytes` #21529\n- no longer raise JSON errors if users pass in file descriptors of JSON to the `json` kwarg in `HttpRequest` #21504\n- Added new error type `IncompleteReadError` which is raised if peer closes the connection before we have received the complete message body.\n\n### Breaking Changes\n\n- SansIOHTTPPolicy.on_exception returns None instead of bool.\n\n### Bugs Fixed\n\n- The `Content-Length` header in a http response is strictly checked against the actual number of bytes in the body,\n rather than silently truncating data in case the underlying tcp connection is closed prematurely.\n (thanks to @jochen-ott-by for the contribution) #20412\n- UnboundLocalError when SansIOHTTPPolicy handles an exception #15222\n- Add default content type header of `text/plain` and content length header for users who pass unicode strings to the `content` kwarg of `HttpRequest` in 2.7 #21550\n\n## 1.19.1 (2021-11-01)\n\n### Bugs Fixed\n\n- respect text encoding specified in argument (thanks to @ryohji for the contribution) #20796\n- Fix \"coroutine x.read() was never awaited\" warning from `ContentDecodePolicy` #21318\n- fix type check for `data` input to `azure.core.rest` for python 2.7 users #21341\n- use `charset_normalizer` if `chardet` is not installed to migrate aiohttp 3.8.0 changes.\n\n### Other Changes\n\n- Refactor AzureJSONEncoder (thanks to @Codejune for the contribution) #21028\n\n## 1.19.0 (2021-09-30)\n\n### Breaking Changes in the Provisional `azure.core.rest` package\n\n- `azure.core.rest.HttpResponse` and `azure.core.rest.AsyncHttpResponse` are now abstract base classes. They should not be initialized directly, instead\nyour transport responses should inherit from them and implement them.\n- The properties of the `azure.core.rest` responses are now all read-only\n\n- HttpLoggingPolicy integrates logs into one record #19925\n\n## 1.18.0 (2021-09-02)\n\n### Features Added\n\n- `azure.core.serialization.AzureJSONEncoder` (introduced in 1.17.0) serializes `datetime.datetime` objects in ISO 8601 format, conforming to RFC 3339's specification. #20190\n- We now use `azure.core.serialization.AzureJSONEncoder` to serialize `json` input to `azure.core.rest.HttpRequest`.\n\n### Breaking Changes in the Provisional `azure.core.rest` package\n\n- The `text` property on `azure.core.rest.HttpResponse` and `azure.core.rest.AsyncHttpResponse` has changed to a method, which also takes\nan `encoding` parameter.\n- Removed `iter_text` and `iter_lines` from `azure.core.rest.HttpResponse` and `azure.core.rest.AsyncHttpResponse`\n\n### Bugs Fixed\n\n- The behaviour of the headers returned in `azure.core.rest` responses now aligns across sync and async. Items can now be checked case-insensitively and without raising an error for format.\n\n## 1.17.0 (2021-08-05)\n\n### Features Added\n\n- Cut hard dependency on requests library\n- Added a `from_json` method which now accepts storage QueueMessage, eventhub's EventData or ServiceBusMessage or simply json bytes to return a `CloudEvent`\n\n### Fixed\n\n- Not override \"x-ms-client-request-id\" if it already exists in the header. #17757\n\n### Breaking Changes in the Provisional `azure.core.rest` package\n\n- `azure.core.rest` will not try to guess the `charset` anymore if it was impossible to extract it from `HttpResponse` analysis. This removes our dependency on `charset`.\n\n## 1.16.0 (2021-07-01)\n\n### Features Added\n\n- Add new ***provisional*** methods `send_request` onto the `azure.core.PipelineClient` and `azure.core.AsyncPipelineClient`. This method takes in\nrequests and sends them through our pipelines.\n- Add new ***provisional*** module `azure.core.rest`. `azure.core.rest` is our new public simple HTTP library in `azure.core` that users will use to create requests, and consume responses.\n- Add new ***provisional*** errors `StreamConsumedError`, `StreamClosedError`, and `ResponseNotReadError` to `azure.core.exceptions`. These errors\nare thrown if you mishandle streamed responses from the provisional `azure.core.rest` module\n\n### Fixed\n\n- Improved error message in the `from_dict` method of `CloudEvent` when a wrong schema is sent.\n\n## 1.15.0 (2021-06-04)\n\n### New Features\n\n- Added `BearerTokenCredentialPolicy.on_challenge` and `.authorize_request` to allow subclasses to optionally handle authentication challenges\n\n### Bug Fixes\n\n- Retry policies don't sleep after operations time out\n- The `from_dict` methhod in the `CloudEvent` can now convert a datetime string to datetime object when microsecond exceeds the python limitation\n\n## 1.14.0 (2021-05-13)\n\n### New Features\n\n- Added `azure.core.credentials.AzureNamedKeyCredential` credential #17548.\n- Added `decompress` parameter for `stream_download` method. If it is set to `False`, will not do decompression upon the stream. #17920\n\n## 1.13.0 (2021-04-02)\n\nAzure core requires Python 2.7 or Python 3.6+ since this release.\n\n### New Features\n\n- Added `azure.core.utils.parse_connection_string` function to parse connection strings across SDKs, with common validation and support for case insensitive keys.\n- Supported adding custom policies #16519\n- Added `~azure.core.tracing.Link` that should be used while passing `Links` to `AbstractSpan`.\n- `AbstractSpan` constructor can now take in additional keyword only args.\n\n### Bug fixes\n\n- Make NetworkTraceLoggingPolicy show the auth token in plain text. #14191\n- Fixed RetryPolicy overriding default connection timeout with an extreme value #17481\n\n## 1.12.0 (2021-03-08)\n\nThis version will be the last version to officially support Python 3.5, future versions will require Python 2.7 or Python 3.6+.\n\n### Features\n\n- Added `azure.core.messaging.CloudEvent` model that follows the cloud event spec.\n- Added `azure.core.serialization.NULL` sentinel value\n- Improve `repr`s for `HttpRequest` and `HttpResponse`s #16972\n\n### Bug Fixes\n\n- Disable retry in stream downloading. (thanks to @jochen-ott-by @hoffmann for the contribution) #16723\n\n## 1.11.0 (2021-02-08)\n\n### Features\n\n- Added `CaseInsensitiveEnumMeta` class for case-insensitive enums. #16316\n- Add `raise_for_status` method onto `HttpResponse`. Calling `response.raise_for_status()` on a response with an error code\nwill raise an `HttpResponseError`. Calling it on a good response will do nothing #16399\n\n### Bug Fixes\n\n- Update conn.conn_kw rather than overriding it when setting block size. (thanks for @jiasli for the contribution) #16587\n\n## 1.10.0 (2021-01-11)\n\n### Features\n\n- Added `AzureSasCredential` and its respective policy. #15946\n\n## 1.9.0 (2020-11-09)\n\n### Features\n\n- Add a `continuation_token` attribute to the base `AzureError` exception, and set this value for errors raised\n during paged or long-running operations.\n\n### Bug Fixes\n\n- Set retry_interval to 1 second instead of 1000 seconds (thanks **vbarbaresi** for contributing) #14357\n\n\n## 1.8.2 (2020-10-05)\n\n### Bug Fixes\n\n- Fixed bug to allow polling in the case of parameterized endpoints with relative polling urls #14097\n\n\n## 1.8.1 (2020-09-08)\n\n### Bug fixes\n\n- SAS credential replicated \"/\" fix #13159\n\n## 1.8.0 (2020-08-10)\n\n### Features\n\n- Support params as list for exploding parameters #12410\n\n\n## 1.7.0 (2020-07-06)\n\n### Bug fixes\n\n- `AzureKeyCredentialPolicy` will now accept (and ignore) passed in kwargs #11963\n- Better error messages if passed endpoint is incorrect #12106\n- Do not JSON encore a string if content type is \"text\" #12137\n\n### Features\n\n- Added `http_logging_policy` property on the `Configuration` object, allowing users to individually\nset the http logging policy of the config #12218\n\n## 1.6.0 (2020-06-03)\n\n### Bug fixes\n\n- Fixed deadlocks in AsyncBearerTokenCredentialPolicy #11543\n- Fix AttributeException in StreamDownloadGenerator #11462\n\n### Features\n\n- Added support for changesets as part of multipart message support #10485\n- Add AsyncLROPoller in azure.core.polling #10801\n- Add get_continuation_token/from_continuation_token/polling_method methods in pollers (sync and async) #10801\n- HttpResponse and PipelineContext objects are now pickable #10801\n\n## 1.5.0 (2020-05-04)\n\n### Features\n\n- Support \"x-ms-retry-after-ms\" in response header #10743\n- `link` and `link_from_headers` now accepts attributes #10765\n\n### Bug fixes\n\n- Not retry if the status code is less than 400 #10778\n- \"x-ms-request-id\" is not considered safe header for logging #10967\n\n## 1.4.0 (2020-04-06)\n\n### Features\n\n- Support a default error type in map_error #9773\n- Added `AzureKeyCredential` and its respective policy. #10509\n- Added `azure.core.polling.base_polling` module with a \"Microsoft One API\" polling implementation #10090\n Also contains the async version in `azure.core.polling.async_base_polling`\n- Support kwarg `enforce_https` to disable HTTPS check on authentication #9821\n- Support additional kwargs in `HttpRequest.set_multipart_mixed` that will be passed into pipeline context.\n\n## 1.3.0 (2020-03-09)\n\n### Bug fixes\n\n- Appended RequestIdPolicy to the default pipeline #9841\n- Rewind the body position in async_retry #10117\n\n### Features\n\n- Add raw_request_hook support in custom_hook_policy #9958\n- Add timeout support in retry_policy #10011\n- Add OdataV4 error format auto-parsing in all exceptions ('error' attribute) #9738\n\n## 1.2.2 (2020-02-10)\n\n### Bug fixes\n\n- Fixed a bug that sends None as request_id #9545\n- Enable mypy for customers #9572\n- Handle TypeError in deep copy #9620\n- Fix text/plain content-type in decoder #9589\n\n## 1.2.1 (2020-01-14)\n\n### Bug fixes\n\n- Fixed a regression in 1.2.0 that was incompatible with azure-keyvault-* 4.0.0\n[#9462](https://github.com/Azure/azure-sdk-for-python/issues/9462)\n\n\n## 1.2.0 (2020-01-14)\n\n### Features\n\n- Add user_agent & sdk_moniker kwargs in UserAgentPolicy init #9355\n- Support OPTIONS HTTP verb #9322\n- Add tracing_attributes to tracing decorator #9297\n- Support auto_request_id in RequestIdPolicy #9163\n- Support fixed retry #6419\n- Support \"retry-after-ms\" in response header #9240\n\n### Bug fixes\n\n- Removed `__enter__` and `__exit__` from async context managers #9313\n\n## 1.1.1 (2019-12-03)\n\n### Bug fixes\n\n- Bearer token authorization requires HTTPS\n- Rewind the body position in retry #8307\n\n## 1.1.0 (2019-11-25)\n\n### Features\n\n- New RequestIdPolicy #8437\n- Enable logging policy in default pipeline #8053\n- Normalize transport timeout. #8000\n Now we have:\n * 'connection_timeout' - a single float in seconds for the connection timeout. Default 5min\n * 'read_timeout' - a single float in seconds for the read timeout. Default 5min\n\n### Bug fixes\n\n- RequestHistory: deepcopy fails if request contains a stream #7732\n- Retry: retry raises error if response does not have http_response #8629\n- Client kwargs are now passed to DistributedTracingPolicy correctly #8051\n- NetworkLoggingPolicy now logs correctly all requests in case of retry #8262\n\n## 1.0.0 (2019-10-29)\n\n### Features\n\n- Tracing: DistributedTracingPolicy now accepts kwargs network_span_namer to change network span name #7773\n- Tracing: Implementation of AbstractSpan can now use the mixin HttpSpanMixin to get HTTP span update automatically #7773\n- Tracing: AbstractSpan contract \"change_context\" introduced #7773\n- Introduce new policy HttpLoggingPolicy #7988\n\n### Bug fixes\n\n- Fix AsyncioRequestsTransport if input stream is an async generator #7743\n- Fix form-data with aiohttp transport #7749\n\n### Breaking changes\n\n- Tracing: AbstractSpan.set_current_span is longer supported. Use change_context instead. #7773\n- azure.core.pipeline.policies.ContentDecodePolicy.deserialize_from_text changed\n\n## 1.0.0b4 (2019-10-07)\n\n### Features\n\n- Tracing: network span context is available with the TRACING_CONTEXT in pipeline response #7252\n- Tracing: Span contract now has `kind`, `traceparent` and is a context manager #7252\n- SansIOHTTPPolicy methods can now be coroutines #7497\n- Add multipart/mixed support #7083:\n\n - HttpRequest now has a \"set_multipart_mixed\" method to set the parts of this request\n - HttpRequest now has a \"prepare_multipart_body\" method to build final body.\n - HttpResponse now has a \"parts\" method to return an iterator of parts\n - AsyncHttpResponse now has a \"parts\" methods to return an async iterator of parts\n - Note that multipart/mixed is a Python 3.x only feature\n\n### Bug fixes\n\n- Tracing: policy cannot fail the pipeline, even in the worst condition #7252\n- Tracing: policy pass correctly status message if exception #7252\n- Tracing: incorrect span if exception raised from decorated function #7133\n- Fixed urllib3 ConnectTimeoutError being raised by Requests during a socket timeout. Now this exception is caught and wrapped as a `ServiceRequestError` #7542\n\n### Breaking changes\n\n- Tracing: `azure.core.tracing.context` removed\n- Tracing: `azure.core.tracing.context.tracing_context.with_current_context` renamed to `azure.core.tracing.common.with_current_context` #7252\n- Tracing: `link` renamed `link_from_headers` and `link` takes now a string\n- Tracing: opencensus implementation has been moved to the package `azure-core-tracing-opencensus`\n- Some modules and classes that were importables from several different places have been removed:\n\n - `azure.core.HttpResponseError` is now only `azure.core.exceptions.HttpResponseError`\n - `azure.core.Configuration` is now only `azure.core.configuration.Configuration`\n - `azure.core.HttpRequest` is now only `azure.core.pipeline.transport.HttpRequest`\n - `azure.core.version` module has been removed. Use `azure.core.__version__` to get version number.\n - `azure.core.pipeline_client` has been removed. Import from `azure.core` instead.\n - `azure.core.pipeline_client_async` has been removed. Import from `azure.core` instead.\n - `azure.core.pipeline.base` has been removed. Import from `azure.core.pipeline` instead.\n - `azure.core.pipeline.base_async` has been removed. Import from `azure.core.pipeline` instead.\n - `azure.core.pipeline.policies.base` has been removed. Import from `azure.core.pipeline.policies` instead.\n - `azure.core.pipeline.policies.base_async` has been removed. Import from `azure.core.pipeline.policies` instead.\n - `azure.core.pipeline.policies.authentication` has been removed. Import from `azure.core.pipeline.policies` instead.\n - `azure.core.pipeline.policies.authentication_async` has been removed. Import from `azure.core.pipeline.policies` instead.\n - `azure.core.pipeline.policies.custom_hook` has been removed. Import from `azure.core.pipeline.policies` instead.\n - `azure.core.pipeline.policies.redirect` has been removed. Import from `azure.core.pipeline.policies` instead.\n - `azure.core.pipeline.policies.redirect_async` has been removed. Import from `azure.core.pipeline.policies` instead.\n - `azure.core.pipeline.policies.retry` has been removed. Import from `azure.core.pipeline.policies` instead.\n - `azure.core.pipeline.policies.retry_async` has been removed. Import from `azure.core.pipeline.policies` instead.\n - `azure.core.pipeline.policies.distributed_tracing` has been removed. Import from `azure.core.pipeline.policies` instead.\n - `azure.core.pipeline.policies.universal` has been removed. Import from `azure.core.pipeline.policies` instead.\n - `azure.core.tracing.abstract_span` has been removed. Import from `azure.core.tracing` instead.\n - `azure.core.pipeline.transport.base` has been removed. Import from `azure.core.pipeline.transport` instead.\n - `azure.core.pipeline.transport.base_async` has been removed. Import from `azure.core.pipeline.transport` instead.\n - `azure.core.pipeline.transport.requests_basic` has been removed. Import from `azure.core.pipeline.transport` instead.\n - `azure.core.pipeline.transport.requests_asyncio` has been removed. Import from `azure.core.pipeline.transport` instead.\n - `azure.core.pipeline.transport.requests_trio` has been removed. Import from `azure.core.pipeline.transport` instead.\n - `azure.core.pipeline.transport.aiohttp` has been removed. Import from `azure.core.pipeline.transport` instead.\n - `azure.core.polling.poller` has been removed. Import from `azure.core.polling` instead.\n - `azure.core.polling.async_poller` has been removed. Import from `azure.core.polling` instead.\n\n## 1.0.0b3 (2019-09-09)\n\n### Bug fixes\n\n- Fix aiohttp auto-headers #6992\n- Add tracing to policies module init #6951\n\n## 1.0.0b2 (2019-08-05)\n\n### Breaking changes\n\n- Transport classes don't take `config` parameter anymore (use kwargs instead) #6372\n- `azure.core.paging` has been completely refactored #6420\n- HttpResponse.content_type attribute is now a string (was a list) #6490\n- For `StreamDownloadGenerator` subclasses, `response` is now an `HttpResponse`, and not a transport response like `aiohttp.ClientResponse` or `requests.Response`. The transport response is available in `internal_response` attribute #6490\n\n### Bug fixes\n\n- aiohttp is not required to import async pipelines classes #6496\n- `AsyncioRequestsTransport.sleep` is now a coroutine as expected #6490\n- `RequestsTransport` is not tight to `ProxyPolicy` implementation details anymore #6372\n- `AiohttpTransport` does not raise on unexpected kwargs #6355\n\n### Features\n\n- New paging base classes that support `continuation_token` and `by_page()` #6420\n- Proxy support for `AiohttpTransport` #6372\n\n## 1.0.0b1 (2019-06-26)\n\n- Preview 1 release", + "release_date": "2025-05-01T23:17:29", "parties": [ { "type": "person", @@ -152,15 +152,14 @@ "Programming Language :: Python :: 3.11", "Programming Language :: Python :: 3.12", "Programming Language :: Python :: 3.13", - "Programming Language :: Python :: 3.8", "Programming Language :: Python :: 3.9" ], "homepage_url": "https://github.com/Azure/azure-sdk-for-python/tree/main/sdk/core/azure-core", - "download_url": "https://files.pythonhosted.org/packages/07/b7/76b7e144aa53bd206bf1ce34fa75350472c3f69bf30e5c8c18bc9881035d/azure_core-1.33.0-py3-none-any.whl", - "size": 207071, + "download_url": "https://files.pythonhosted.org/packages/84/9e/5c87b49f65bb16571599bc789857d0ded2f53014d3392bc88a5d1f3ad779/azure_core-1.34.0-py3-none-any.whl", + "size": 207409, "sha1": null, - "md5": "ddd17415a5573ca4c435d7db0ff5631f", - "sha256": "9b5b6d0223a1d38c37500e6971118c1e0f13f54951e6893968b38910bc9cda8f", + "md5": "b03f983d92f026edffb1a5c50ee5a64b", + "sha256": "0615d3b756beccdb6624d1c0ae97284f38b78fb59a2a9839bf927c66fbbdddd6", "sha512": null, "bug_tracking_url": null, "code_view_url": null, @@ -180,9 +179,9 @@ "dependencies": [], "repository_homepage_url": null, "repository_download_url": null, - "api_data_url": "https://pypi.org/pypi/azure-core/1.33.0/json", + "api_data_url": "https://pypi.org/pypi/azure-core/1.34.0/json", "datasource_id": null, - "purl": "pkg:pypi/azure-core@1.33.0" + "purl": "pkg:pypi/azure-core@1.34.0" }, { "type": "pypi", @@ -441,12 +440,12 @@ "type": "pypi", "namespace": null, "name": "charset-normalizer", - "version": "3.4.1", + "version": "3.4.2", "qualifiers": {}, "subpath": null, "primary_language": "Python", - "description": "The Real First Universal Charset Detector. Open, modern and actively maintained alternative to Chardet.\n

Charset Detection, for Everyone \ud83d\udc4b

\n\n

\n The Real First Universal Charset Detector
\n \n \n \n \n \"Download\n \n \n \n \n

\n

\n Featured Packages
\n \n \"Static\n \n \n \"Static\n \n

\n

\n In other language (unofficial port - by the community)
\n \n \"Static\n \n

\n\n> A library that helps you read text from an unknown charset encoding.
Motivated by `chardet`,\n> I'm trying to resolve the issue by taking a new approach.\n> All IANA character set names for which the Python core library provides codecs are supported.\n\n

\n >>>>> \ud83d\udc49 Try Me Online Now, Then Adopt Me \ud83d\udc48 <<<<<\n

\n\nThis project offers you an alternative to **Universal Charset Encoding Detector**, also known as **Chardet**.\n\n| Feature | [Chardet](https://github.com/chardet/chardet) | Charset Normalizer | [cChardet](https://github.com/PyYoshi/cChardet) |\n|--------------------------------------------------|:---------------------------------------------:|:--------------------------------------------------------------------------------------------------:|:-----------------------------------------------:|\n| `Fast` | \u274c | \u2705 | \u2705 |\n| `Universal**` | \u274c | \u2705 | \u274c |\n| `Reliable` **without** distinguishable standards | \u274c | \u2705 | \u2705 |\n| `Reliable` **with** distinguishable standards | \u2705 | \u2705 | \u2705 |\n| `License` | LGPL-2.1
_restrictive_ | MIT | MPL-1.1
_restrictive_ |\n| `Native Python` | \u2705 | \u2705 | \u274c |\n| `Detect spoken language` | \u274c | \u2705 | N/A |\n| `UnicodeDecodeError Safety` | \u274c | \u2705 | \u274c |\n| `Whl Size (min)` | 193.6 kB | 42 kB | ~200 kB |\n| `Supported Encoding` | 33 | \ud83c\udf89 [99](https://charset-normalizer.readthedocs.io/en/latest/user/support.html#supported-encodings) | 40 |\n\n

\n\"Reading\"Cat\n

\n\n*\\*\\* : They are clearly using specific code for a specific encoding even if covering most of used one*
\n\n## \u26a1 Performance\n\nThis package offer better performance than its counterpart Chardet. Here are some numbers.\n\n| Package | Accuracy | Mean per file (ms) | File per sec (est) |\n|-----------------------------------------------|:--------:|:------------------:|:------------------:|\n| [chardet](https://github.com/chardet/chardet) | 86 % | 63 ms | 16 file/sec |\n| charset-normalizer | **98 %** | **10 ms** | 100 file/sec |\n\n| Package | 99th percentile | 95th percentile | 50th percentile |\n|-----------------------------------------------|:---------------:|:---------------:|:---------------:|\n| [chardet](https://github.com/chardet/chardet) | 265 ms | 71 ms | 7 ms |\n| charset-normalizer | 100 ms | 50 ms | 5 ms |\n\n_updated as of december 2024 using CPython 3.12_\n\nChardet's performance on larger file (1MB+) are very poor. Expect huge difference on large payload.\n\n> Stats are generated using 400+ files using default parameters. More details on used files, see GHA workflows.\n> And yes, these results might change at any time. The dataset can be updated to include more files.\n> The actual delays heavily depends on your CPU capabilities. The factors should remain the same.\n> Keep in mind that the stats are generous and that Chardet accuracy vs our is measured using Chardet initial capability\n> (e.g. Supported Encoding) Challenge-them if you want.\n\n## \u2728 Installation\n\nUsing pip:\n\n```sh\npip install charset-normalizer -U\n```\n\n## \ud83d\ude80 Basic Usage\n\n### CLI\nThis package comes with a CLI.\n\n```\nusage: normalizer [-h] [-v] [-a] [-n] [-m] [-r] [-f] [-t THRESHOLD]\n file [file ...]\n\nThe Real First Universal Charset Detector. Discover originating encoding used\non text file. Normalize text to unicode.\n\npositional arguments:\n files File(s) to be analysed\n\noptional arguments:\n -h, --help show this help message and exit\n -v, --verbose Display complementary information about file if any.\n Stdout will contain logs about the detection process.\n -a, --with-alternative\n Output complementary possibilities if any. Top-level\n JSON WILL be a list.\n -n, --normalize Permit to normalize input file. If not set, program\n does not write anything.\n -m, --minimal Only output the charset detected to STDOUT. Disabling\n JSON output.\n -r, --replace Replace file when trying to normalize it instead of\n creating a new one.\n -f, --force Replace file without asking if you are sure, use this\n flag with caution.\n -t THRESHOLD, --threshold THRESHOLD\n Define a custom maximum amount of chaos allowed in\n decoded content. 0. <= chaos <= 1.\n --version Show version information and exit.\n```\n\n```bash\nnormalizer ./data/sample.1.fr.srt\n```\n\nor\n\n```bash\npython -m charset_normalizer ./data/sample.1.fr.srt\n```\n\n\ud83c\udf89 Since version 1.4.0 the CLI produce easily usable stdout result in JSON format.\n\n```json\n{\n \"path\": \"/home/default/projects/charset_normalizer/data/sample.1.fr.srt\",\n \"encoding\": \"cp1252\",\n \"encoding_aliases\": [\n \"1252\",\n \"windows_1252\"\n ],\n \"alternative_encodings\": [\n \"cp1254\",\n \"cp1256\",\n \"cp1258\",\n \"iso8859_14\",\n \"iso8859_15\",\n \"iso8859_16\",\n \"iso8859_3\",\n \"iso8859_9\",\n \"latin_1\",\n \"mbcs\"\n ],\n \"language\": \"French\",\n \"alphabets\": [\n \"Basic Latin\",\n \"Latin-1 Supplement\"\n ],\n \"has_sig_or_bom\": false,\n \"chaos\": 0.149,\n \"coherence\": 97.152,\n \"unicode_path\": null,\n \"is_preferred\": true\n}\n```\n\n### Python\n*Just print out normalized text*\n```python\nfrom charset_normalizer import from_path\n\nresults = from_path('./my_subtitle.srt')\n\nprint(str(results.best()))\n```\n\n*Upgrade your code without effort*\n```python\nfrom charset_normalizer import detect\n```\n\nThe above code will behave the same as **chardet**. We ensure that we offer the best (reasonable) BC result possible.\n\nSee the docs for advanced usage : [readthedocs.io](https://charset-normalizer.readthedocs.io/en/latest/)\n\n## \ud83d\ude07 Why\n\nWhen I started using Chardet, I noticed that it was not suited to my expectations, and I wanted to propose a\nreliable alternative using a completely different method. Also! I never back down on a good challenge!\n\nI **don't care** about the **originating charset** encoding, because **two different tables** can\nproduce **two identical rendered string.**\nWhat I want is to get readable text, the best I can.\n\nIn a way, **I'm brute forcing text decoding.** How cool is that ? \ud83d\ude0e\n\nDon't confuse package **ftfy** with charset-normalizer or chardet. ftfy goal is to repair Unicode string whereas charset-normalizer to convert raw file in unknown encoding to unicode.\n\n## \ud83c\udf70 How\n\n - Discard all charset encoding table that could not fit the binary content.\n - Measure noise, or the mess once opened (by chunks) with a corresponding charset encoding.\n - Extract matches with the lowest mess detected.\n - Additionally, we measure coherence / probe for a language.\n\n**Wait a minute**, what is noise/mess and coherence according to **YOU ?**\n\n*Noise :* I opened hundred of text files, **written by humans**, with the wrong encoding table. **I observed**, then\n**I established** some ground rules about **what is obvious** when **it seems like** a mess (aka. defining noise in rendered text).\n I know that my interpretation of what is noise is probably incomplete, feel free to contribute in order to\n improve or rewrite it.\n\n*Coherence :* For each language there is on earth, we have computed ranked letter appearance occurrences (the best we can). So I thought\nthat intel is worth something here. So I use those records against decoded text to check if I can detect intelligent design.\n\n## \u26a1 Known limitations\n\n - Language detection is unreliable when text contains two or more languages sharing identical letters. (eg. HTML (english tags) + Turkish content (Sharing Latin characters))\n - Every charset detector heavily depends on sufficient content. In common cases, do not bother run detection on very tiny content.\n\n## \u26a0\ufe0f About Python EOLs\n\n**If you are running:**\n\n- Python >=2.7,<3.5: Unsupported\n- Python 3.5: charset-normalizer < 2.1\n- Python 3.6: charset-normalizer < 3.1\n- Python 3.7: charset-normalizer < 4.0\n\nUpgrade your Python interpreter as soon as possible.\n\n## \ud83d\udc64 Contributing\n\nContributions, issues and feature requests are very much welcome.
\nFeel free to check [issues page](https://github.com/ousret/charset_normalizer/issues) if you want to contribute.\n\n## \ud83d\udcdd License\n\nCopyright \u00a9 [Ahmed TAHRI @Ousret](https://github.com/Ousret).
\nThis project is [MIT](https://github.com/Ousret/charset_normalizer/blob/master/LICENSE) licensed.\n\nCharacters frequencies used in this project \u00a9 2012 [Denny Vrande\u010di\u0107](http://simia.net/letters/)\n\n## \ud83d\udcbc For Enterprise\n\nProfessional support for charset-normalizer is available as part of the [Tidelift\nSubscription][1]. Tidelift gives software development teams a single source for\npurchasing and maintaining their software, with professional grade assurances\nfrom the experts who know it best, while seamlessly integrating with existing\ntools.\n\n[1]: https://tidelift.com/subscription/pkg/pypi-charset-normalizer?utm_source=pypi-charset-normalizer&utm_medium=readme\n\n[![OpenSSF Best Practices](https://www.bestpractices.dev/projects/7297/badge)](https://www.bestpractices.dev/projects/7297)\n\n# Changelog\nAll notable changes to charset-normalizer will be documented in this file. This project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html).\nThe format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/).\n\n## [3.4.1](https://github.com/Ousret/charset_normalizer/compare/3.4.0...3.4.1) (2024-12-24)\n\n### Changed\n- Project metadata are now stored using `pyproject.toml` instead of `setup.cfg` using setuptools as the build backend.\n- Enforce annotation delayed loading for a simpler and consistent types in the project.\n- Optional mypyc compilation upgraded to version 1.14 for Python >= 3.8\n\n### Added\n- pre-commit configuration.\n- noxfile.\n\n### Removed\n- `build-requirements.txt` as per using `pyproject.toml` native build configuration.\n- `bin/integration.py` and `bin/serve.py` in favor of downstream integration test (see noxfile).\n- `setup.cfg` in favor of `pyproject.toml` metadata configuration.\n- Unused `utils.range_scan` function.\n\n### Fixed\n- Converting content to Unicode bytes may insert `utf_8` instead of preferred `utf-8`. (#572)\n- Deprecation warning \"'count' is passed as positional argument\" when converting to Unicode bytes on Python 3.13+\n\n## [3.4.0](https://github.com/Ousret/charset_normalizer/compare/3.3.2...3.4.0) (2024-10-08)\n\n### Added\n- Argument `--no-preemptive` in the CLI to prevent the detector to search for hints.\n- Support for Python 3.13 (#512)\n\n### Fixed\n- Relax the TypeError exception thrown when trying to compare a CharsetMatch with anything else than a CharsetMatch.\n- Improved the general reliability of the detector based on user feedbacks. (#520) (#509) (#498) (#407) (#537)\n- Declared charset in content (preemptive detection) not changed when converting to utf-8 bytes. (#381)\n\n## [3.3.2](https://github.com/Ousret/charset_normalizer/compare/3.3.1...3.3.2) (2023-10-31)\n\n### Fixed\n- Unintentional memory usage regression when using large payload that match several encoding (#376)\n- Regression on some detection case showcased in the documentation (#371)\n\n### Added\n- Noise (md) probe that identify malformed arabic representation due to the presence of letters in isolated form (credit to my wife)\n\n## [3.3.1](https://github.com/Ousret/charset_normalizer/compare/3.3.0...3.3.1) (2023-10-22)\n\n### Changed\n- Optional mypyc compilation upgraded to version 1.6.1 for Python >= 3.8\n- Improved the general detection reliability based on reports from the community\n\n## [3.3.0](https://github.com/Ousret/charset_normalizer/compare/3.2.0...3.3.0) (2023-09-30)\n\n### Added\n- Allow to execute the CLI (e.g. normalizer) through `python -m charset_normalizer.cli` or `python -m charset_normalizer`\n- Support for 9 forgotten encoding that are supported by Python but unlisted in `encoding.aliases` as they have no alias (#323)\n\n### Removed\n- (internal) Redundant utils.is_ascii function and unused function is_private_use_only\n- (internal) charset_normalizer.assets is moved inside charset_normalizer.constant\n\n### Changed\n- (internal) Unicode code blocks in constants are updated using the latest v15.0.0 definition to improve detection\n- Optional mypyc compilation upgraded to version 1.5.1 for Python >= 3.8\n\n### Fixed\n- Unable to properly sort CharsetMatch when both chaos/noise and coherence were close due to an unreachable condition in \\_\\_lt\\_\\_ (#350)\n\n## [3.2.0](https://github.com/Ousret/charset_normalizer/compare/3.1.0...3.2.0) (2023-06-07)\n\n### Changed\n- Typehint for function `from_path` no longer enforce `PathLike` as its first argument\n- Minor improvement over the global detection reliability\n\n### Added\n- Introduce function `is_binary` that relies on main capabilities, and optimized to detect binaries\n- Propagate `enable_fallback` argument throughout `from_bytes`, `from_path`, and `from_fp` that allow a deeper control over the detection (default True)\n- Explicit support for Python 3.12\n\n### Fixed\n- Edge case detection failure where a file would contain 'very-long' camel cased word (Issue #289)\n\n## [3.1.0](https://github.com/Ousret/charset_normalizer/compare/3.0.1...3.1.0) (2023-03-06)\n\n### Added\n- Argument `should_rename_legacy` for legacy function `detect` and disregard any new arguments without errors (PR #262)\n\n### Removed\n- Support for Python 3.6 (PR #260)\n\n### Changed\n- Optional speedup provided by mypy/c 1.0.1\n\n## [3.0.1](https://github.com/Ousret/charset_normalizer/compare/3.0.0...3.0.1) (2022-11-18)\n\n### Fixed\n- Multi-bytes cutter/chunk generator did not always cut correctly (PR #233)\n\n### Changed\n- Speedup provided by mypy/c 0.990 on Python >= 3.7\n\n## [3.0.0](https://github.com/Ousret/charset_normalizer/compare/2.1.1...3.0.0) (2022-10-20)\n\n### Added\n- Extend the capability of explain=True when cp_isolation contains at most two entries (min one), will log in details of the Mess-detector results\n- Support for alternative language frequency set in charset_normalizer.assets.FREQUENCIES\n- Add parameter `language_threshold` in `from_bytes`, `from_path` and `from_fp` to adjust the minimum expected coherence ratio\n- `normalizer --version` now specify if current version provide extra speedup (meaning mypyc compilation whl)\n\n### Changed\n- Build with static metadata using 'build' frontend\n- Make the language detection stricter\n- Optional: Module `md.py` can be compiled using Mypyc to provide an extra speedup up to 4x faster than v2.1\n\n### Fixed\n- CLI with opt --normalize fail when using full path for files\n- TooManyAccentuatedPlugin induce false positive on the mess detection when too few alpha character have been fed to it\n- Sphinx warnings when generating the documentation\n\n### Removed\n- Coherence detector no longer return 'Simple English' instead return 'English'\n- Coherence detector no longer return 'Classical Chinese' instead return 'Chinese'\n- Breaking: Method `first()` and `best()` from CharsetMatch\n- UTF-7 will no longer appear as \"detected\" without a recognized SIG/mark (is unreliable/conflict with ASCII)\n- Breaking: Class aliases CharsetDetector, CharsetDoctor, CharsetNormalizerMatch and CharsetNormalizerMatches\n- Breaking: Top-level function `normalize`\n- Breaking: Properties `chaos_secondary_pass`, `coherence_non_latin` and `w_counter` from CharsetMatch\n- Support for the backport `unicodedata2`\n\n## [3.0.0rc1](https://github.com/Ousret/charset_normalizer/compare/3.0.0b2...3.0.0rc1) (2022-10-18)\n\n### Added\n- Extend the capability of explain=True when cp_isolation contains at most two entries (min one), will log in details of the Mess-detector results\n- Support for alternative language frequency set in charset_normalizer.assets.FREQUENCIES\n- Add parameter `language_threshold` in `from_bytes`, `from_path` and `from_fp` to adjust the minimum expected coherence ratio\n\n### Changed\n- Build with static metadata using 'build' frontend\n- Make the language detection stricter\n\n### Fixed\n- CLI with opt --normalize fail when using full path for files\n- TooManyAccentuatedPlugin induce false positive on the mess detection when too few alpha character have been fed to it\n\n### Removed\n- Coherence detector no longer return 'Simple English' instead return 'English'\n- Coherence detector no longer return 'Classical Chinese' instead return 'Chinese'\n\n## [3.0.0b2](https://github.com/Ousret/charset_normalizer/compare/3.0.0b1...3.0.0b2) (2022-08-21)\n\n### Added\n- `normalizer --version` now specify if current version provide extra speedup (meaning mypyc compilation whl)\n\n### Removed\n- Breaking: Method `first()` and `best()` from CharsetMatch\n- UTF-7 will no longer appear as \"detected\" without a recognized SIG/mark (is unreliable/conflict with ASCII)\n\n### Fixed\n- Sphinx warnings when generating the documentation\n\n## [3.0.0b1](https://github.com/Ousret/charset_normalizer/compare/2.1.0...3.0.0b1) (2022-08-15)\n\n### Changed\n- Optional: Module `md.py` can be compiled using Mypyc to provide an extra speedup up to 4x faster than v2.1\n\n### Removed\n- Breaking: Class aliases CharsetDetector, CharsetDoctor, CharsetNormalizerMatch and CharsetNormalizerMatches\n- Breaking: Top-level function `normalize`\n- Breaking: Properties `chaos_secondary_pass`, `coherence_non_latin` and `w_counter` from CharsetMatch\n- Support for the backport `unicodedata2`\n\n## [2.1.1](https://github.com/Ousret/charset_normalizer/compare/2.1.0...2.1.1) (2022-08-19)\n\n### Deprecated\n- Function `normalize` scheduled for removal in 3.0\n\n### Changed\n- Removed useless call to decode in fn is_unprintable (#206)\n\n### Fixed\n- Third-party library (i18n xgettext) crashing not recognizing utf_8 (PEP 263) with underscore from [@aleksandernovikov](https://github.com/aleksandernovikov) (#204)\n\n## [2.1.0](https://github.com/Ousret/charset_normalizer/compare/2.0.12...2.1.0) (2022-06-19)\n\n### Added\n- Output the Unicode table version when running the CLI with `--version` (PR #194)\n\n### Changed\n- Re-use decoded buffer for single byte character sets from [@nijel](https://github.com/nijel) (PR #175)\n- Fixing some performance bottlenecks from [@deedy5](https://github.com/deedy5) (PR #183)\n\n### Fixed\n- Workaround potential bug in cpython with Zero Width No-Break Space located in Arabic Presentation Forms-B, Unicode 1.1 not acknowledged as space (PR #175)\n- CLI default threshold aligned with the API threshold from [@oleksandr-kuzmenko](https://github.com/oleksandr-kuzmenko) (PR #181)\n\n### Removed\n- Support for Python 3.5 (PR #192)\n\n### Deprecated\n- Use of backport unicodedata from `unicodedata2` as Python is quickly catching up, scheduled for removal in 3.0 (PR #194)\n\n## [2.0.12](https://github.com/Ousret/charset_normalizer/compare/2.0.11...2.0.12) (2022-02-12)\n\n### Fixed\n- ASCII miss-detection on rare cases (PR #170)\n\n## [2.0.11](https://github.com/Ousret/charset_normalizer/compare/2.0.10...2.0.11) (2022-01-30)\n\n### Added\n- Explicit support for Python 3.11 (PR #164)\n\n### Changed\n- The logging behavior have been completely reviewed, now using only TRACE and DEBUG levels (PR #163 #165)\n\n## [2.0.10](https://github.com/Ousret/charset_normalizer/compare/2.0.9...2.0.10) (2022-01-04)\n\n### Fixed\n- Fallback match entries might lead to UnicodeDecodeError for large bytes sequence (PR #154)\n\n### Changed\n- Skipping the language-detection (CD) on ASCII (PR #155)\n\n## [2.0.9](https://github.com/Ousret/charset_normalizer/compare/2.0.8...2.0.9) (2021-12-03)\n\n### Changed\n- Moderating the logging impact (since 2.0.8) for specific environments (PR #147)\n\n### Fixed\n- Wrong logging level applied when setting kwarg `explain` to True (PR #146)\n\n## [2.0.8](https://github.com/Ousret/charset_normalizer/compare/2.0.7...2.0.8) (2021-11-24)\n### Changed\n- Improvement over Vietnamese detection (PR #126)\n- MD improvement on trailing data and long foreign (non-pure latin) data (PR #124)\n- Efficiency improvements in cd/alphabet_languages from [@adbar](https://github.com/adbar) (PR #122)\n- call sum() without an intermediary list following PEP 289 recommendations from [@adbar](https://github.com/adbar) (PR #129)\n- Code style as refactored by Sourcery-AI (PR #131)\n- Minor adjustment on the MD around european words (PR #133)\n- Remove and replace SRTs from assets / tests (PR #139)\n- Initialize the library logger with a `NullHandler` by default from [@nmaynes](https://github.com/nmaynes) (PR #135)\n- Setting kwarg `explain` to True will add provisionally (bounded to function lifespan) a specific stream handler (PR #135)\n\n### Fixed\n- Fix large (misleading) sequence giving UnicodeDecodeError (PR #137)\n- Avoid using too insignificant chunk (PR #137)\n\n### Added\n- Add and expose function `set_logging_handler` to configure a specific StreamHandler from [@nmaynes](https://github.com/nmaynes) (PR #135)\n- Add `CHANGELOG.md` entries, format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/) (PR #141)\n\n## [2.0.7](https://github.com/Ousret/charset_normalizer/compare/2.0.6...2.0.7) (2021-10-11)\n### Added\n- Add support for Kazakh (Cyrillic) language detection (PR #109)\n\n### Changed\n- Further, improve inferring the language from a given single-byte code page (PR #112)\n- Vainly trying to leverage PEP263 when PEP3120 is not supported (PR #116)\n- Refactoring for potential performance improvements in loops from [@adbar](https://github.com/adbar) (PR #113)\n- Various detection improvement (MD+CD) (PR #117)\n\n### Removed\n- Remove redundant logging entry about detected language(s) (PR #115)\n\n### Fixed\n- Fix a minor inconsistency between Python 3.5 and other versions regarding language detection (PR #117 #102)\n\n## [2.0.6](https://github.com/Ousret/charset_normalizer/compare/2.0.5...2.0.6) (2021-09-18)\n### Fixed\n- Unforeseen regression with the loss of the backward-compatibility with some older minor of Python 3.5.x (PR #100)\n- Fix CLI crash when using --minimal output in certain cases (PR #103)\n\n### Changed\n- Minor improvement to the detection efficiency (less than 1%) (PR #106 #101)\n\n## [2.0.5](https://github.com/Ousret/charset_normalizer/compare/2.0.4...2.0.5) (2021-09-14)\n### Changed\n- The project now comply with: flake8, mypy, isort and black to ensure a better overall quality (PR #81)\n- The BC-support with v1.x was improved, the old staticmethods are restored (PR #82)\n- The Unicode detection is slightly improved (PR #93)\n- Add syntax sugar \\_\\_bool\\_\\_ for results CharsetMatches list-container (PR #91)\n\n### Removed\n- The project no longer raise warning on tiny content given for detection, will be simply logged as warning instead (PR #92)\n\n### Fixed\n- In some rare case, the chunks extractor could cut in the middle of a multi-byte character and could mislead the mess detection (PR #95)\n- Some rare 'space' characters could trip up the UnprintablePlugin/Mess detection (PR #96)\n- The MANIFEST.in was not exhaustive (PR #78)\n\n## [2.0.4](https://github.com/Ousret/charset_normalizer/compare/2.0.3...2.0.4) (2021-07-30)\n### Fixed\n- The CLI no longer raise an unexpected exception when no encoding has been found (PR #70)\n- Fix accessing the 'alphabets' property when the payload contains surrogate characters (PR #68)\n- The logger could mislead (explain=True) on detected languages and the impact of one MBCS match (PR #72)\n- Submatch factoring could be wrong in rare edge cases (PR #72)\n- Multiple files given to the CLI were ignored when publishing results to STDOUT. (After the first path) (PR #72)\n- Fix line endings from CRLF to LF for certain project files (PR #67)\n\n### Changed\n- Adjust the MD to lower the sensitivity, thus improving the global detection reliability (PR #69 #76)\n- Allow fallback on specified encoding if any (PR #71)\n\n## [2.0.3](https://github.com/Ousret/charset_normalizer/compare/2.0.2...2.0.3) (2021-07-16)\n### Changed\n- Part of the detection mechanism has been improved to be less sensitive, resulting in more accurate detection results. Especially ASCII. (PR #63)\n- According to the community wishes, the detection will fall back on ASCII or UTF-8 in a last-resort case. (PR #64)\n\n## [2.0.2](https://github.com/Ousret/charset_normalizer/compare/2.0.1...2.0.2) (2021-07-15)\n### Fixed\n- Empty/Too small JSON payload miss-detection fixed. Report from [@tseaver](https://github.com/tseaver) (PR #59)\n\n### Changed\n- Don't inject unicodedata2 into sys.modules from [@akx](https://github.com/akx) (PR #57)\n\n## [2.0.1](https://github.com/Ousret/charset_normalizer/compare/2.0.0...2.0.1) (2021-07-13)\n### Fixed\n- Make it work where there isn't a filesystem available, dropping assets frequencies.json. Report from [@sethmlarson](https://github.com/sethmlarson). (PR #55)\n- Using explain=False permanently disable the verbose output in the current runtime (PR #47)\n- One log entry (language target preemptive) was not show in logs when using explain=True (PR #47)\n- Fix undesired exception (ValueError) on getitem of instance CharsetMatches (PR #52)\n\n### Changed\n- Public function normalize default args values were not aligned with from_bytes (PR #53)\n\n### Added\n- You may now use charset aliases in cp_isolation and cp_exclusion arguments (PR #47)\n\n## [2.0.0](https://github.com/Ousret/charset_normalizer/compare/1.4.1...2.0.0) (2021-07-02)\n### Changed\n- 4x to 5 times faster than the previous 1.4.0 release. At least 2x faster than Chardet.\n- Accent has been made on UTF-8 detection, should perform rather instantaneous.\n- The backward compatibility with Chardet has been greatly improved. The legacy detect function returns an identical charset name whenever possible.\n- The detection mechanism has been slightly improved, now Turkish content is detected correctly (most of the time)\n- The program has been rewritten to ease the readability and maintainability. (+Using static typing)+\n- utf_7 detection has been reinstated.\n\n### Removed\n- This package no longer require anything when used with Python 3.5 (Dropped cached_property)\n- Removed support for these languages: Catalan, Esperanto, Kazakh, Baque, Volap\u00fck, Azeri, Galician, Nynorsk, Macedonian, and Serbocroatian.\n- The exception hook on UnicodeDecodeError has been removed.\n\n### Deprecated\n- Methods coherence_non_latin, w_counter, chaos_secondary_pass of the class CharsetMatch are now deprecated and scheduled for removal in v3.0\n\n### Fixed\n- The CLI output used the relative path of the file(s). Should be absolute.\n\n## [1.4.1](https://github.com/Ousret/charset_normalizer/compare/1.4.0...1.4.1) (2021-05-28)\n### Fixed\n- Logger configuration/usage no longer conflict with others (PR #44)\n\n## [1.4.0](https://github.com/Ousret/charset_normalizer/compare/1.3.9...1.4.0) (2021-05-21)\n### Removed\n- Using standard logging instead of using the package loguru.\n- Dropping nose test framework in favor of the maintained pytest.\n- Choose to not use dragonmapper package to help with gibberish Chinese/CJK text.\n- Require cached_property only for Python 3.5 due to constraint. Dropping for every other interpreter version.\n- Stop support for UTF-7 that does not contain a SIG.\n- Dropping PrettyTable, replaced with pure JSON output in CLI.\n\n### Fixed\n- BOM marker in a CharsetNormalizerMatch instance could be False in rare cases even if obviously present. Due to the sub-match factoring process.\n- Not searching properly for the BOM when trying utf32/16 parent codec.\n\n### Changed\n- Improving the package final size by compressing frequencies.json.\n- Huge improvement over the larges payload.\n\n### Added\n- CLI now produces JSON consumable output.\n- Return ASCII if given sequences fit. Given reasonable confidence.\n\n## [1.3.9](https://github.com/Ousret/charset_normalizer/compare/1.3.8...1.3.9) (2021-05-13)\n\n### Fixed\n- In some very rare cases, you may end up getting encode/decode errors due to a bad bytes payload (PR #40)\n\n## [1.3.8](https://github.com/Ousret/charset_normalizer/compare/1.3.7...1.3.8) (2021-05-12)\n\n### Fixed\n- Empty given payload for detection may cause an exception if trying to access the `alphabets` property. (PR #39)\n\n## [1.3.7](https://github.com/Ousret/charset_normalizer/compare/1.3.6...1.3.7) (2021-05-12)\n\n### Fixed\n- The legacy detect function should return UTF-8-SIG if sig is present in the payload. (PR #38)\n\n## [1.3.6](https://github.com/Ousret/charset_normalizer/compare/1.3.5...1.3.6) (2021-02-09)\n\n### Changed\n- Amend the previous release to allow prettytable 2.0 (PR #35)\n\n## [1.3.5](https://github.com/Ousret/charset_normalizer/compare/1.3.4...1.3.5) (2021-02-08)\n\n### Fixed\n- Fix error while using the package with a python pre-release interpreter (PR #33)\n\n### Changed\n- Dependencies refactoring, constraints revised.\n\n### Added\n- Add python 3.9 and 3.10 to the supported interpreters\n\nMIT License\n\nCopyright (c) 2025 TAHRI Ahmed R.\n\nPermission is hereby granted, free of charge, to any person obtaining a copy\nof this software and associated documentation files (the \"Software\"), to deal\nin the Software without restriction, including without limitation the rights\nto use, copy, modify, merge, publish, distribute, sublicense, and/or sell\ncopies of the Software, and to permit persons to whom the Software is\nfurnished to do so, subject to the following conditions:\n\nThe above copyright notice and this permission notice shall be included in all\ncopies or substantial portions of the Software.\n\nTHE SOFTWARE IS PROVIDED \"AS IS\", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR\nIMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,\nFITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE\nAUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER\nLIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,\nOUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE\nSOFTWARE.", - "release_date": "2024-12-24T18:09:54", + "description": "The Real First Universal Charset Detector. Open, modern and actively maintained alternative to Chardet.\n

Charset Detection, for Everyone \ud83d\udc4b

\n\n

\n The Real First Universal Charset Detector
\n \n \n \n \n \"Download\n \n \n \n \n

\n

\n Featured Packages
\n \n \"Static\n \n \n \"Static\n \n

\n

\n In other language (unofficial port - by the community)
\n \n \"Static\n \n

\n\n> A library that helps you read text from an unknown charset encoding.
Motivated by `chardet`,\n> I'm trying to resolve the issue by taking a new approach.\n> All IANA character set names for which the Python core library provides codecs are supported.\n\n

\n >>>>> \ud83d\udc49 Try Me Online Now, Then Adopt Me \ud83d\udc48 <<<<<\n

\n\nThis project offers you an alternative to **Universal Charset Encoding Detector**, also known as **Chardet**.\n\n| Feature | [Chardet](https://github.com/chardet/chardet) | Charset Normalizer | [cChardet](https://github.com/PyYoshi/cChardet) |\n|--------------------------------------------------|:---------------------------------------------:|:--------------------------------------------------------------------------------------------------:|:-----------------------------------------------:|\n| `Fast` | \u274c | \u2705 | \u2705 |\n| `Universal**` | \u274c | \u2705 | \u274c |\n| `Reliable` **without** distinguishable standards | \u274c | \u2705 | \u2705 |\n| `Reliable` **with** distinguishable standards | \u2705 | \u2705 | \u2705 |\n| `License` | LGPL-2.1
_restrictive_ | MIT | MPL-1.1
_restrictive_ |\n| `Native Python` | \u2705 | \u2705 | \u274c |\n| `Detect spoken language` | \u274c | \u2705 | N/A |\n| `UnicodeDecodeError Safety` | \u274c | \u2705 | \u274c |\n| `Whl Size (min)` | 193.6 kB | 42 kB | ~200 kB |\n| `Supported Encoding` | 33 | \ud83c\udf89 [99](https://charset-normalizer.readthedocs.io/en/latest/user/support.html#supported-encodings) | 40 |\n\n

\n\"Reading\"Cat\n

\n\n*\\*\\* : They are clearly using specific code for a specific encoding even if covering most of used one*
\n\n## \u26a1 Performance\n\nThis package offer better performance than its counterpart Chardet. Here are some numbers.\n\n| Package | Accuracy | Mean per file (ms) | File per sec (est) |\n|-----------------------------------------------|:--------:|:------------------:|:------------------:|\n| [chardet](https://github.com/chardet/chardet) | 86 % | 63 ms | 16 file/sec |\n| charset-normalizer | **98 %** | **10 ms** | 100 file/sec |\n\n| Package | 99th percentile | 95th percentile | 50th percentile |\n|-----------------------------------------------|:---------------:|:---------------:|:---------------:|\n| [chardet](https://github.com/chardet/chardet) | 265 ms | 71 ms | 7 ms |\n| charset-normalizer | 100 ms | 50 ms | 5 ms |\n\n_updated as of december 2024 using CPython 3.12_\n\nChardet's performance on larger file (1MB+) are very poor. Expect huge difference on large payload.\n\n> Stats are generated using 400+ files using default parameters. More details on used files, see GHA workflows.\n> And yes, these results might change at any time. The dataset can be updated to include more files.\n> The actual delays heavily depends on your CPU capabilities. The factors should remain the same.\n> Keep in mind that the stats are generous and that Chardet accuracy vs our is measured using Chardet initial capability\n> (e.g. Supported Encoding) Challenge-them if you want.\n\n## \u2728 Installation\n\nUsing pip:\n\n```sh\npip install charset-normalizer -U\n```\n\n## \ud83d\ude80 Basic Usage\n\n### CLI\nThis package comes with a CLI.\n\n```\nusage: normalizer [-h] [-v] [-a] [-n] [-m] [-r] [-f] [-t THRESHOLD]\n file [file ...]\n\nThe Real First Universal Charset Detector. Discover originating encoding used\non text file. Normalize text to unicode.\n\npositional arguments:\n files File(s) to be analysed\n\noptional arguments:\n -h, --help show this help message and exit\n -v, --verbose Display complementary information about file if any.\n Stdout will contain logs about the detection process.\n -a, --with-alternative\n Output complementary possibilities if any. Top-level\n JSON WILL be a list.\n -n, --normalize Permit to normalize input file. If not set, program\n does not write anything.\n -m, --minimal Only output the charset detected to STDOUT. Disabling\n JSON output.\n -r, --replace Replace file when trying to normalize it instead of\n creating a new one.\n -f, --force Replace file without asking if you are sure, use this\n flag with caution.\n -t THRESHOLD, --threshold THRESHOLD\n Define a custom maximum amount of chaos allowed in\n decoded content. 0. <= chaos <= 1.\n --version Show version information and exit.\n```\n\n```bash\nnormalizer ./data/sample.1.fr.srt\n```\n\nor\n\n```bash\npython -m charset_normalizer ./data/sample.1.fr.srt\n```\n\n\ud83c\udf89 Since version 1.4.0 the CLI produce easily usable stdout result in JSON format.\n\n```json\n{\n \"path\": \"/home/default/projects/charset_normalizer/data/sample.1.fr.srt\",\n \"encoding\": \"cp1252\",\n \"encoding_aliases\": [\n \"1252\",\n \"windows_1252\"\n ],\n \"alternative_encodings\": [\n \"cp1254\",\n \"cp1256\",\n \"cp1258\",\n \"iso8859_14\",\n \"iso8859_15\",\n \"iso8859_16\",\n \"iso8859_3\",\n \"iso8859_9\",\n \"latin_1\",\n \"mbcs\"\n ],\n \"language\": \"French\",\n \"alphabets\": [\n \"Basic Latin\",\n \"Latin-1 Supplement\"\n ],\n \"has_sig_or_bom\": false,\n \"chaos\": 0.149,\n \"coherence\": 97.152,\n \"unicode_path\": null,\n \"is_preferred\": true\n}\n```\n\n### Python\n*Just print out normalized text*\n```python\nfrom charset_normalizer import from_path\n\nresults = from_path('./my_subtitle.srt')\n\nprint(str(results.best()))\n```\n\n*Upgrade your code without effort*\n```python\nfrom charset_normalizer import detect\n```\n\nThe above code will behave the same as **chardet**. We ensure that we offer the best (reasonable) BC result possible.\n\nSee the docs for advanced usage : [readthedocs.io](https://charset-normalizer.readthedocs.io/en/latest/)\n\n## \ud83d\ude07 Why\n\nWhen I started using Chardet, I noticed that it was not suited to my expectations, and I wanted to propose a\nreliable alternative using a completely different method. Also! I never back down on a good challenge!\n\nI **don't care** about the **originating charset** encoding, because **two different tables** can\nproduce **two identical rendered string.**\nWhat I want is to get readable text, the best I can.\n\nIn a way, **I'm brute forcing text decoding.** How cool is that ? \ud83d\ude0e\n\nDon't confuse package **ftfy** with charset-normalizer or chardet. ftfy goal is to repair Unicode string whereas charset-normalizer to convert raw file in unknown encoding to unicode.\n\n## \ud83c\udf70 How\n\n - Discard all charset encoding table that could not fit the binary content.\n - Measure noise, or the mess once opened (by chunks) with a corresponding charset encoding.\n - Extract matches with the lowest mess detected.\n - Additionally, we measure coherence / probe for a language.\n\n**Wait a minute**, what is noise/mess and coherence according to **YOU ?**\n\n*Noise :* I opened hundred of text files, **written by humans**, with the wrong encoding table. **I observed**, then\n**I established** some ground rules about **what is obvious** when **it seems like** a mess (aka. defining noise in rendered text).\n I know that my interpretation of what is noise is probably incomplete, feel free to contribute in order to\n improve or rewrite it.\n\n*Coherence :* For each language there is on earth, we have computed ranked letter appearance occurrences (the best we can). So I thought\nthat intel is worth something here. So I use those records against decoded text to check if I can detect intelligent design.\n\n## \u26a1 Known limitations\n\n - Language detection is unreliable when text contains two or more languages sharing identical letters. (eg. HTML (english tags) + Turkish content (Sharing Latin characters))\n - Every charset detector heavily depends on sufficient content. In common cases, do not bother run detection on very tiny content.\n\n## \u26a0\ufe0f About Python EOLs\n\n**If you are running:**\n\n- Python >=2.7,<3.5: Unsupported\n- Python 3.5: charset-normalizer < 2.1\n- Python 3.6: charset-normalizer < 3.1\n- Python 3.7: charset-normalizer < 4.0\n\nUpgrade your Python interpreter as soon as possible.\n\n## \ud83d\udc64 Contributing\n\nContributions, issues and feature requests are very much welcome.
\nFeel free to check [issues page](https://github.com/ousret/charset_normalizer/issues) if you want to contribute.\n\n## \ud83d\udcdd License\n\nCopyright \u00a9 [Ahmed TAHRI @Ousret](https://github.com/Ousret).
\nThis project is [MIT](https://github.com/Ousret/charset_normalizer/blob/master/LICENSE) licensed.\n\nCharacters frequencies used in this project \u00a9 2012 [Denny Vrande\u010di\u0107](http://simia.net/letters/)\n\n## \ud83d\udcbc For Enterprise\n\nProfessional support for charset-normalizer is available as part of the [Tidelift\nSubscription][1]. Tidelift gives software development teams a single source for\npurchasing and maintaining their software, with professional grade assurances\nfrom the experts who know it best, while seamlessly integrating with existing\ntools.\n\n[1]: https://tidelift.com/subscription/pkg/pypi-charset-normalizer?utm_source=pypi-charset-normalizer&utm_medium=readme\n\n[![OpenSSF Best Practices](https://www.bestpractices.dev/projects/7297/badge)](https://www.bestpractices.dev/projects/7297)\n\n# Changelog\nAll notable changes to charset-normalizer will be documented in this file. This project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html).\nThe format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/).\n\n## [3.4.2](https://github.com/Ousret/charset_normalizer/compare/3.4.1...3.4.2) (2025-05-02)\n\n### Fixed\n- Addressed the DeprecationWarning in our CLI regarding `argparse.FileType` by backporting the target class into the package. (#591)\n- Improved the overall reliability of the detector with CJK Ideographs. (#605) (#587)\n\n### Changed\n- Optional mypyc compilation upgraded to version 1.15 for Python >= 3.8\n\n## [3.4.1](https://github.com/Ousret/charset_normalizer/compare/3.4.0...3.4.1) (2024-12-24)\n\n### Changed\n- Project metadata are now stored using `pyproject.toml` instead of `setup.cfg` using setuptools as the build backend.\n- Enforce annotation delayed loading for a simpler and consistent types in the project.\n- Optional mypyc compilation upgraded to version 1.14 for Python >= 3.8\n\n### Added\n- pre-commit configuration.\n- noxfile.\n\n### Removed\n- `build-requirements.txt` as per using `pyproject.toml` native build configuration.\n- `bin/integration.py` and `bin/serve.py` in favor of downstream integration test (see noxfile).\n- `setup.cfg` in favor of `pyproject.toml` metadata configuration.\n- Unused `utils.range_scan` function.\n\n### Fixed\n- Converting content to Unicode bytes may insert `utf_8` instead of preferred `utf-8`. (#572)\n- Deprecation warning \"'count' is passed as positional argument\" when converting to Unicode bytes on Python 3.13+\n\n## [3.4.0](https://github.com/Ousret/charset_normalizer/compare/3.3.2...3.4.0) (2024-10-08)\n\n### Added\n- Argument `--no-preemptive` in the CLI to prevent the detector to search for hints.\n- Support for Python 3.13 (#512)\n\n### Fixed\n- Relax the TypeError exception thrown when trying to compare a CharsetMatch with anything else than a CharsetMatch.\n- Improved the general reliability of the detector based on user feedbacks. (#520) (#509) (#498) (#407) (#537)\n- Declared charset in content (preemptive detection) not changed when converting to utf-8 bytes. (#381)\n\n## [3.3.2](https://github.com/Ousret/charset_normalizer/compare/3.3.1...3.3.2) (2023-10-31)\n\n### Fixed\n- Unintentional memory usage regression when using large payload that match several encoding (#376)\n- Regression on some detection case showcased in the documentation (#371)\n\n### Added\n- Noise (md) probe that identify malformed arabic representation due to the presence of letters in isolated form (credit to my wife)\n\n## [3.3.1](https://github.com/Ousret/charset_normalizer/compare/3.3.0...3.3.1) (2023-10-22)\n\n### Changed\n- Optional mypyc compilation upgraded to version 1.6.1 for Python >= 3.8\n- Improved the general detection reliability based on reports from the community\n\n## [3.3.0](https://github.com/Ousret/charset_normalizer/compare/3.2.0...3.3.0) (2023-09-30)\n\n### Added\n- Allow to execute the CLI (e.g. normalizer) through `python -m charset_normalizer.cli` or `python -m charset_normalizer`\n- Support for 9 forgotten encoding that are supported by Python but unlisted in `encoding.aliases` as they have no alias (#323)\n\n### Removed\n- (internal) Redundant utils.is_ascii function and unused function is_private_use_only\n- (internal) charset_normalizer.assets is moved inside charset_normalizer.constant\n\n### Changed\n- (internal) Unicode code blocks in constants are updated using the latest v15.0.0 definition to improve detection\n- Optional mypyc compilation upgraded to version 1.5.1 for Python >= 3.8\n\n### Fixed\n- Unable to properly sort CharsetMatch when both chaos/noise and coherence were close due to an unreachable condition in \\_\\_lt\\_\\_ (#350)\n\n## [3.2.0](https://github.com/Ousret/charset_normalizer/compare/3.1.0...3.2.0) (2023-06-07)\n\n### Changed\n- Typehint for function `from_path` no longer enforce `PathLike` as its first argument\n- Minor improvement over the global detection reliability\n\n### Added\n- Introduce function `is_binary` that relies on main capabilities, and optimized to detect binaries\n- Propagate `enable_fallback` argument throughout `from_bytes`, `from_path`, and `from_fp` that allow a deeper control over the detection (default True)\n- Explicit support for Python 3.12\n\n### Fixed\n- Edge case detection failure where a file would contain 'very-long' camel cased word (Issue #289)\n\n## [3.1.0](https://github.com/Ousret/charset_normalizer/compare/3.0.1...3.1.0) (2023-03-06)\n\n### Added\n- Argument `should_rename_legacy` for legacy function `detect` and disregard any new arguments without errors (PR #262)\n\n### Removed\n- Support for Python 3.6 (PR #260)\n\n### Changed\n- Optional speedup provided by mypy/c 1.0.1\n\n## [3.0.1](https://github.com/Ousret/charset_normalizer/compare/3.0.0...3.0.1) (2022-11-18)\n\n### Fixed\n- Multi-bytes cutter/chunk generator did not always cut correctly (PR #233)\n\n### Changed\n- Speedup provided by mypy/c 0.990 on Python >= 3.7\n\n## [3.0.0](https://github.com/Ousret/charset_normalizer/compare/2.1.1...3.0.0) (2022-10-20)\n\n### Added\n- Extend the capability of explain=True when cp_isolation contains at most two entries (min one), will log in details of the Mess-detector results\n- Support for alternative language frequency set in charset_normalizer.assets.FREQUENCIES\n- Add parameter `language_threshold` in `from_bytes`, `from_path` and `from_fp` to adjust the minimum expected coherence ratio\n- `normalizer --version` now specify if current version provide extra speedup (meaning mypyc compilation whl)\n\n### Changed\n- Build with static metadata using 'build' frontend\n- Make the language detection stricter\n- Optional: Module `md.py` can be compiled using Mypyc to provide an extra speedup up to 4x faster than v2.1\n\n### Fixed\n- CLI with opt --normalize fail when using full path for files\n- TooManyAccentuatedPlugin induce false positive on the mess detection when too few alpha character have been fed to it\n- Sphinx warnings when generating the documentation\n\n### Removed\n- Coherence detector no longer return 'Simple English' instead return 'English'\n- Coherence detector no longer return 'Classical Chinese' instead return 'Chinese'\n- Breaking: Method `first()` and `best()` from CharsetMatch\n- UTF-7 will no longer appear as \"detected\" without a recognized SIG/mark (is unreliable/conflict with ASCII)\n- Breaking: Class aliases CharsetDetector, CharsetDoctor, CharsetNormalizerMatch and CharsetNormalizerMatches\n- Breaking: Top-level function `normalize`\n- Breaking: Properties `chaos_secondary_pass`, `coherence_non_latin` and `w_counter` from CharsetMatch\n- Support for the backport `unicodedata2`\n\n## [3.0.0rc1](https://github.com/Ousret/charset_normalizer/compare/3.0.0b2...3.0.0rc1) (2022-10-18)\n\n### Added\n- Extend the capability of explain=True when cp_isolation contains at most two entries (min one), will log in details of the Mess-detector results\n- Support for alternative language frequency set in charset_normalizer.assets.FREQUENCIES\n- Add parameter `language_threshold` in `from_bytes`, `from_path` and `from_fp` to adjust the minimum expected coherence ratio\n\n### Changed\n- Build with static metadata using 'build' frontend\n- Make the language detection stricter\n\n### Fixed\n- CLI with opt --normalize fail when using full path for files\n- TooManyAccentuatedPlugin induce false positive on the mess detection when too few alpha character have been fed to it\n\n### Removed\n- Coherence detector no longer return 'Simple English' instead return 'English'\n- Coherence detector no longer return 'Classical Chinese' instead return 'Chinese'\n\n## [3.0.0b2](https://github.com/Ousret/charset_normalizer/compare/3.0.0b1...3.0.0b2) (2022-08-21)\n\n### Added\n- `normalizer --version` now specify if current version provide extra speedup (meaning mypyc compilation whl)\n\n### Removed\n- Breaking: Method `first()` and `best()` from CharsetMatch\n- UTF-7 will no longer appear as \"detected\" without a recognized SIG/mark (is unreliable/conflict with ASCII)\n\n### Fixed\n- Sphinx warnings when generating the documentation\n\n## [3.0.0b1](https://github.com/Ousret/charset_normalizer/compare/2.1.0...3.0.0b1) (2022-08-15)\n\n### Changed\n- Optional: Module `md.py` can be compiled using Mypyc to provide an extra speedup up to 4x faster than v2.1\n\n### Removed\n- Breaking: Class aliases CharsetDetector, CharsetDoctor, CharsetNormalizerMatch and CharsetNormalizerMatches\n- Breaking: Top-level function `normalize`\n- Breaking: Properties `chaos_secondary_pass`, `coherence_non_latin` and `w_counter` from CharsetMatch\n- Support for the backport `unicodedata2`\n\n## [2.1.1](https://github.com/Ousret/charset_normalizer/compare/2.1.0...2.1.1) (2022-08-19)\n\n### Deprecated\n- Function `normalize` scheduled for removal in 3.0\n\n### Changed\n- Removed useless call to decode in fn is_unprintable (#206)\n\n### Fixed\n- Third-party library (i18n xgettext) crashing not recognizing utf_8 (PEP 263) with underscore from [@aleksandernovikov](https://github.com/aleksandernovikov) (#204)\n\n## [2.1.0](https://github.com/Ousret/charset_normalizer/compare/2.0.12...2.1.0) (2022-06-19)\n\n### Added\n- Output the Unicode table version when running the CLI with `--version` (PR #194)\n\n### Changed\n- Re-use decoded buffer for single byte character sets from [@nijel](https://github.com/nijel) (PR #175)\n- Fixing some performance bottlenecks from [@deedy5](https://github.com/deedy5) (PR #183)\n\n### Fixed\n- Workaround potential bug in cpython with Zero Width No-Break Space located in Arabic Presentation Forms-B, Unicode 1.1 not acknowledged as space (PR #175)\n- CLI default threshold aligned with the API threshold from [@oleksandr-kuzmenko](https://github.com/oleksandr-kuzmenko) (PR #181)\n\n### Removed\n- Support for Python 3.5 (PR #192)\n\n### Deprecated\n- Use of backport unicodedata from `unicodedata2` as Python is quickly catching up, scheduled for removal in 3.0 (PR #194)\n\n## [2.0.12](https://github.com/Ousret/charset_normalizer/compare/2.0.11...2.0.12) (2022-02-12)\n\n### Fixed\n- ASCII miss-detection on rare cases (PR #170)\n\n## [2.0.11](https://github.com/Ousret/charset_normalizer/compare/2.0.10...2.0.11) (2022-01-30)\n\n### Added\n- Explicit support for Python 3.11 (PR #164)\n\n### Changed\n- The logging behavior have been completely reviewed, now using only TRACE and DEBUG levels (PR #163 #165)\n\n## [2.0.10](https://github.com/Ousret/charset_normalizer/compare/2.0.9...2.0.10) (2022-01-04)\n\n### Fixed\n- Fallback match entries might lead to UnicodeDecodeError for large bytes sequence (PR #154)\n\n### Changed\n- Skipping the language-detection (CD) on ASCII (PR #155)\n\n## [2.0.9](https://github.com/Ousret/charset_normalizer/compare/2.0.8...2.0.9) (2021-12-03)\n\n### Changed\n- Moderating the logging impact (since 2.0.8) for specific environments (PR #147)\n\n### Fixed\n- Wrong logging level applied when setting kwarg `explain` to True (PR #146)\n\n## [2.0.8](https://github.com/Ousret/charset_normalizer/compare/2.0.7...2.0.8) (2021-11-24)\n### Changed\n- Improvement over Vietnamese detection (PR #126)\n- MD improvement on trailing data and long foreign (non-pure latin) data (PR #124)\n- Efficiency improvements in cd/alphabet_languages from [@adbar](https://github.com/adbar) (PR #122)\n- call sum() without an intermediary list following PEP 289 recommendations from [@adbar](https://github.com/adbar) (PR #129)\n- Code style as refactored by Sourcery-AI (PR #131)\n- Minor adjustment on the MD around european words (PR #133)\n- Remove and replace SRTs from assets / tests (PR #139)\n- Initialize the library logger with a `NullHandler` by default from [@nmaynes](https://github.com/nmaynes) (PR #135)\n- Setting kwarg `explain` to True will add provisionally (bounded to function lifespan) a specific stream handler (PR #135)\n\n### Fixed\n- Fix large (misleading) sequence giving UnicodeDecodeError (PR #137)\n- Avoid using too insignificant chunk (PR #137)\n\n### Added\n- Add and expose function `set_logging_handler` to configure a specific StreamHandler from [@nmaynes](https://github.com/nmaynes) (PR #135)\n- Add `CHANGELOG.md` entries, format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/) (PR #141)\n\n## [2.0.7](https://github.com/Ousret/charset_normalizer/compare/2.0.6...2.0.7) (2021-10-11)\n### Added\n- Add support for Kazakh (Cyrillic) language detection (PR #109)\n\n### Changed\n- Further, improve inferring the language from a given single-byte code page (PR #112)\n- Vainly trying to leverage PEP263 when PEP3120 is not supported (PR #116)\n- Refactoring for potential performance improvements in loops from [@adbar](https://github.com/adbar) (PR #113)\n- Various detection improvement (MD+CD) (PR #117)\n\n### Removed\n- Remove redundant logging entry about detected language(s) (PR #115)\n\n### Fixed\n- Fix a minor inconsistency between Python 3.5 and other versions regarding language detection (PR #117 #102)\n\n## [2.0.6](https://github.com/Ousret/charset_normalizer/compare/2.0.5...2.0.6) (2021-09-18)\n### Fixed\n- Unforeseen regression with the loss of the backward-compatibility with some older minor of Python 3.5.x (PR #100)\n- Fix CLI crash when using --minimal output in certain cases (PR #103)\n\n### Changed\n- Minor improvement to the detection efficiency (less than 1%) (PR #106 #101)\n\n## [2.0.5](https://github.com/Ousret/charset_normalizer/compare/2.0.4...2.0.5) (2021-09-14)\n### Changed\n- The project now comply with: flake8, mypy, isort and black to ensure a better overall quality (PR #81)\n- The BC-support with v1.x was improved, the old staticmethods are restored (PR #82)\n- The Unicode detection is slightly improved (PR #93)\n- Add syntax sugar \\_\\_bool\\_\\_ for results CharsetMatches list-container (PR #91)\n\n### Removed\n- The project no longer raise warning on tiny content given for detection, will be simply logged as warning instead (PR #92)\n\n### Fixed\n- In some rare case, the chunks extractor could cut in the middle of a multi-byte character and could mislead the mess detection (PR #95)\n- Some rare 'space' characters could trip up the UnprintablePlugin/Mess detection (PR #96)\n- The MANIFEST.in was not exhaustive (PR #78)\n\n## [2.0.4](https://github.com/Ousret/charset_normalizer/compare/2.0.3...2.0.4) (2021-07-30)\n### Fixed\n- The CLI no longer raise an unexpected exception when no encoding has been found (PR #70)\n- Fix accessing the 'alphabets' property when the payload contains surrogate characters (PR #68)\n- The logger could mislead (explain=True) on detected languages and the impact of one MBCS match (PR #72)\n- Submatch factoring could be wrong in rare edge cases (PR #72)\n- Multiple files given to the CLI were ignored when publishing results to STDOUT. (After the first path) (PR #72)\n- Fix line endings from CRLF to LF for certain project files (PR #67)\n\n### Changed\n- Adjust the MD to lower the sensitivity, thus improving the global detection reliability (PR #69 #76)\n- Allow fallback on specified encoding if any (PR #71)\n\n## [2.0.3](https://github.com/Ousret/charset_normalizer/compare/2.0.2...2.0.3) (2021-07-16)\n### Changed\n- Part of the detection mechanism has been improved to be less sensitive, resulting in more accurate detection results. Especially ASCII. (PR #63)\n- According to the community wishes, the detection will fall back on ASCII or UTF-8 in a last-resort case. (PR #64)\n\n## [2.0.2](https://github.com/Ousret/charset_normalizer/compare/2.0.1...2.0.2) (2021-07-15)\n### Fixed\n- Empty/Too small JSON payload miss-detection fixed. Report from [@tseaver](https://github.com/tseaver) (PR #59)\n\n### Changed\n- Don't inject unicodedata2 into sys.modules from [@akx](https://github.com/akx) (PR #57)\n\n## [2.0.1](https://github.com/Ousret/charset_normalizer/compare/2.0.0...2.0.1) (2021-07-13)\n### Fixed\n- Make it work where there isn't a filesystem available, dropping assets frequencies.json. Report from [@sethmlarson](https://github.com/sethmlarson). (PR #55)\n- Using explain=False permanently disable the verbose output in the current runtime (PR #47)\n- One log entry (language target preemptive) was not show in logs when using explain=True (PR #47)\n- Fix undesired exception (ValueError) on getitem of instance CharsetMatches (PR #52)\n\n### Changed\n- Public function normalize default args values were not aligned with from_bytes (PR #53)\n\n### Added\n- You may now use charset aliases in cp_isolation and cp_exclusion arguments (PR #47)\n\n## [2.0.0](https://github.com/Ousret/charset_normalizer/compare/1.4.1...2.0.0) (2021-07-02)\n### Changed\n- 4x to 5 times faster than the previous 1.4.0 release. At least 2x faster than Chardet.\n- Accent has been made on UTF-8 detection, should perform rather instantaneous.\n- The backward compatibility with Chardet has been greatly improved. The legacy detect function returns an identical charset name whenever possible.\n- The detection mechanism has been slightly improved, now Turkish content is detected correctly (most of the time)\n- The program has been rewritten to ease the readability and maintainability. (+Using static typing)+\n- utf_7 detection has been reinstated.\n\n### Removed\n- This package no longer require anything when used with Python 3.5 (Dropped cached_property)\n- Removed support for these languages: Catalan, Esperanto, Kazakh, Baque, Volap\u00fck, Azeri, Galician, Nynorsk, Macedonian, and Serbocroatian.\n- The exception hook on UnicodeDecodeError has been removed.\n\n### Deprecated\n- Methods coherence_non_latin, w_counter, chaos_secondary_pass of the class CharsetMatch are now deprecated and scheduled for removal in v3.0\n\n### Fixed\n- The CLI output used the relative path of the file(s). Should be absolute.\n\n## [1.4.1](https://github.com/Ousret/charset_normalizer/compare/1.4.0...1.4.1) (2021-05-28)\n### Fixed\n- Logger configuration/usage no longer conflict with others (PR #44)\n\n## [1.4.0](https://github.com/Ousret/charset_normalizer/compare/1.3.9...1.4.0) (2021-05-21)\n### Removed\n- Using standard logging instead of using the package loguru.\n- Dropping nose test framework in favor of the maintained pytest.\n- Choose to not use dragonmapper package to help with gibberish Chinese/CJK text.\n- Require cached_property only for Python 3.5 due to constraint. Dropping for every other interpreter version.\n- Stop support for UTF-7 that does not contain a SIG.\n- Dropping PrettyTable, replaced with pure JSON output in CLI.\n\n### Fixed\n- BOM marker in a CharsetNormalizerMatch instance could be False in rare cases even if obviously present. Due to the sub-match factoring process.\n- Not searching properly for the BOM when trying utf32/16 parent codec.\n\n### Changed\n- Improving the package final size by compressing frequencies.json.\n- Huge improvement over the larges payload.\n\n### Added\n- CLI now produces JSON consumable output.\n- Return ASCII if given sequences fit. Given reasonable confidence.\n\n## [1.3.9](https://github.com/Ousret/charset_normalizer/compare/1.3.8...1.3.9) (2021-05-13)\n\n### Fixed\n- In some very rare cases, you may end up getting encode/decode errors due to a bad bytes payload (PR #40)\n\n## [1.3.8](https://github.com/Ousret/charset_normalizer/compare/1.3.7...1.3.8) (2021-05-12)\n\n### Fixed\n- Empty given payload for detection may cause an exception if trying to access the `alphabets` property. (PR #39)\n\n## [1.3.7](https://github.com/Ousret/charset_normalizer/compare/1.3.6...1.3.7) (2021-05-12)\n\n### Fixed\n- The legacy detect function should return UTF-8-SIG if sig is present in the payload. (PR #38)\n\n## [1.3.6](https://github.com/Ousret/charset_normalizer/compare/1.3.5...1.3.6) (2021-02-09)\n\n### Changed\n- Amend the previous release to allow prettytable 2.0 (PR #35)\n\n## [1.3.5](https://github.com/Ousret/charset_normalizer/compare/1.3.4...1.3.5) (2021-02-08)\n\n### Fixed\n- Fix error while using the package with a python pre-release interpreter (PR #33)\n\n### Changed\n- Dependencies refactoring, constraints revised.\n\n### Added\n- Add python 3.9 and 3.10 to the supported interpreters\n\nMIT License\n\nCopyright (c) 2025 TAHRI Ahmed R.\n\nPermission is hereby granted, free of charge, to any person obtaining a copy\nof this software and associated documentation files (the \"Software\"), to deal\nin the Software without restriction, including without limitation the rights\nto use, copy, modify, merge, publish, distribute, sublicense, and/or sell\ncopies of the Software, and to permit persons to whom the Software is\nfurnished to do so, subject to the following conditions:\n\nThe above copyright notice and this permission notice shall be included in all\ncopies or substantial portions of the Software.\n\nTHE SOFTWARE IS PROVIDED \"AS IS\", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR\nIMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,\nFITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE\nAUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER\nLIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,\nOUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE\nSOFTWARE.", + "release_date": "2025-05-02T08:31:56", "parties": [ { "type": "person", @@ -492,11 +491,11 @@ "Typing :: Typed" ], "homepage_url": null, - "download_url": "https://files.pythonhosted.org/packages/93/62/5e89cdfe04584cb7f4d36003ffa2936681b03ecc0754f8e969c2becb7e24/charset_normalizer-3.4.1-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", - "size": 146123, + "download_url": "https://files.pythonhosted.org/packages/a8/2d/7a5b635aa65284bf3eab7653e8b4151ab420ecbae918d3e359d1947b4d61/charset_normalizer-3.4.2-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", + "size": 149471, "sha1": null, - "md5": "022fc9420b83eef0aa97064c4610d0e6", - "sha256": "b010a7a4fd316c3c484d482922d13044979e78d1861f0e0650423144c616a46a", + "md5": "8e85a262cbd63cdd5f2a3d1304587a45", + "sha256": "8075c35cd58273fee266c58c0c9b670947c19df5fb98e7b66710e04ad4e9ff86", "sha512": null, "bug_tracking_url": null, "code_view_url": "https://github.com/jawah/charset_normalizer", @@ -516,9 +515,9 @@ "dependencies": [], "repository_homepage_url": null, "repository_download_url": null, - "api_data_url": "https://pypi.org/pypi/charset-normalizer/3.4.1/json", + "api_data_url": "https://pypi.org/pypi/charset-normalizer/3.4.2/json", "datasource_id": null, - "purl": "pkg:pypi/charset-normalizer@3.4.1" + "purl": "pkg:pypi/charset-normalizer@3.4.2" }, { "type": "pypi", @@ -1318,7 +1317,7 @@ ], "resolved_dependencies_graph": [ { - "package": "pkg:pypi/azure-core@1.33.0", + "package": "pkg:pypi/azure-core@1.34.0", "dependencies": [ "pkg:pypi/requests@2.32.3", "pkg:pypi/six@1.17.0", @@ -1334,7 +1333,7 @@ { "package": "pkg:pypi/azure-storage-blob@12.25.1", "dependencies": [ - "pkg:pypi/azure-core@1.33.0", + "pkg:pypi/azure-core@1.34.0", "pkg:pypi/cryptography@44.0.2", "pkg:pypi/isodate@0.7.2", "pkg:pypi/typing-extensions@4.13.2" @@ -1351,7 +1350,7 @@ ] }, { - "package": "pkg:pypi/charset-normalizer@3.4.1", + "package": "pkg:pypi/charset-normalizer@3.4.2", "dependencies": [] }, { @@ -1375,7 +1374,7 @@ { "package": "pkg:pypi/msrest@0.7.1", "dependencies": [ - "pkg:pypi/azure-core@1.33.0", + "pkg:pypi/azure-core@1.34.0", "pkg:pypi/certifi@2025.4.26", "pkg:pypi/isodate@0.7.2", "pkg:pypi/requests-oauthlib@2.0.0", @@ -1401,7 +1400,7 @@ "package": "pkg:pypi/requests@2.32.3", "dependencies": [ "pkg:pypi/certifi@2025.4.26", - "pkg:pypi/charset-normalizer@3.4.1", + "pkg:pypi/charset-normalizer@3.4.2", "pkg:pypi/idna@3.10", "pkg:pypi/urllib3@2.4.0" ] diff --git a/tests/data/azure-devops.req-312-expected.json b/tests/data/azure-devops.req-312-expected.json index fac7965..fa29754 100644 --- a/tests/data/azure-devops.req-312-expected.json +++ b/tests/data/azure-devops.req-312-expected.json @@ -126,12 +126,12 @@ "type": "pypi", "namespace": null, "name": "azure-core", - "version": "1.33.0", + "version": "1.34.0", "qualifiers": {}, "subpath": null, "primary_language": "Python", - "description": "Microsoft Azure Core Library for Python\n# Azure Core shared client library for Python\n\nAzure core provides shared exceptions and modules for Python SDK client libraries.\nThese libraries follow the [Azure SDK Design Guidelines for Python](https://azure.github.io/azure-sdk/python/guidelines/index.html) .\n\nIf you are a client library developer, please reference [client library developer reference](https://github.com/Azure/azure-sdk-for-python/blob/main/sdk/core/azure-core/CLIENT_LIBRARY_DEVELOPER.md) for more information.\n\n[Source code](https://github.com/Azure/azure-sdk-for-python/blob/main/sdk/core/azure-core/)\n| [Package (Pypi)][package]\n| [Package (Conda)](https://anaconda.org/microsoft/azure-core/)\n| [API reference documentation](https://learn.microsoft.com/python/api/overview/azure/core-readme)\n\n## Getting started\n\nTypically, you will not need to install azure core;\nit will be installed when you install one of the client libraries using it.\nIn case you want to install it explicitly (to implement your own client library, for example),\nyou can find it [here](https://pypi.org/project/azure-core/).\n\n## Key concepts\n\n### Azure Core Library Exceptions\n\n#### AzureError\n\nAzureError is the base exception for all errors.\n\n```python\nclass AzureError(Exception):\n def __init__(self, message, *args, **kwargs):\n self.inner_exception = kwargs.get(\"error\")\n self.exc_type, self.exc_value, self.exc_traceback = sys.exc_info()\n self.exc_type = self.exc_type.__name__ if self.exc_type else type(self.inner_exception)\n self.exc_msg = \"{}, {}: {}\".format(message, self.exc_type, self.exc_value) # type: ignore\n self.message = str(message)\n self.continuation_token = kwargs.get(\"continuation_token\")\n super(AzureError, self).__init__(self.message, *args)\n```\n\n*message* is any message (str) to be associated with the exception.\n\n*args* are any additional args to be included with exception.\n\n*kwargs* are keyword arguments to include with the exception. Use the keyword *error* to pass in an internal exception and *continuation_token* for a token reference to continue an incomplete operation.\n\n**The following exceptions inherit from AzureError:**\n\n#### ServiceRequestError\n\nAn error occurred while attempt to make a request to the service. No request was sent.\n\n#### ServiceResponseError\n\nThe request was sent, but the client failed to understand the response.\nThe connection may have timed out. These errors can be retried for idempotent or safe operations.\n\n#### HttpResponseError\n\nA request was made, and a non-success status code was received from the service.\n\n```python\nclass HttpResponseError(AzureError):\n def __init__(self, message=None, response=None, **kwargs):\n self.reason = None\n self.response = response\n if response:\n self.reason = response.reason\n self.status_code = response.status_code\n self.error = self._parse_odata_body(ODataV4Format, response) # type: Optional[ODataV4Format]\n if self.error:\n message = str(self.error)\n else:\n message = message or \"Operation returned an invalid status '{}'\".format(\n self.reason\n )\n\n super(HttpResponseError, self).__init__(message=message, **kwargs)\n```\n\n*message* is the HTTP response error message (optional)\n\n*response* is the HTTP response (optional).\n\n*kwargs* are keyword arguments to include with the exception.\n\n**The following exceptions inherit from HttpResponseError:**\n\n#### DecodeError\n\nAn error raised during response de-serialization.\n\n#### IncompleteReadError\n\nAn error raised if peer closes the connection before we have received the complete message body.\n\n#### ResourceExistsError\n\nAn error response with status code 4xx. This will not be raised directly by the Azure core pipeline.\n\n#### ResourceNotFoundError\n\nAn error response, typically triggered by a 412 response (for update) or 404 (for get/post).\n\n#### ResourceModifiedError\n\nAn error response with status code 4xx, typically 412 Conflict. This will not be raised directly by the Azure core pipeline.\n\n#### ResourceNotModifiedError\n\nAn error response with status code 304. This will not be raised directly by the Azure core pipeline.\n\n#### ClientAuthenticationError\n\nAn error response with status code 4xx. This will not be raised directly by the Azure core pipeline.\n\n#### TooManyRedirectsError\n\nAn error raised when the maximum number of redirect attempts is reached. The maximum amount of redirects can be configured in the RedirectPolicy.\n\n```python\nclass TooManyRedirectsError(HttpResponseError):\n def __init__(self, history, *args, **kwargs):\n self.history = history\n message = \"Reached maximum redirect attempts.\"\n super(TooManyRedirectsError, self).__init__(message, *args, **kwargs)\n```\n\n*history* is used to document the requests/responses that resulted in redirected requests.\n\n*args* are any additional args to be included with exception.\n\n*kwargs* are keyword arguments to include with the exception.\n\n#### StreamConsumedError\n\nAn error thrown if you try to access the stream of `azure.core.rest.HttpResponse` or `azure.core.rest.AsyncHttpResponse` once\nthe response stream has been consumed.\n\n#### StreamClosedError\n\nAn error thrown if you try to access the stream of the `azure.core.rest.HttpResponse` or `azure.core.rest.AsyncHttpResponse` once\nthe response stream has been closed.\n\n#### ResponseNotReadError\n\nAn error thrown if you try to access the `content` of `azure.core.rest.HttpResponse` or `azure.core.rest.AsyncHttpResponse` before\nreading in the response's bytes first.\n\n### Configurations\n\nWhen calling the methods, some properties can be configured by passing in as kwargs arguments.\n\n| Parameters | Description |\n| --- | --- |\n| headers | The HTTP Request headers. |\n| request_id | The request id to be added into header. |\n| user_agent | If specified, this will be added in front of the user agent string. |\n| logging_enable| Use to enable per operation. Defaults to `False`. |\n| logger | If specified, it will be used to log information. |\n| response_encoding | The encoding to use if known for this service (will disable auto-detection). |\n| raw_request_hook | Callback function. Will be invoked on request. |\n| raw_response_hook | Callback function. Will be invoked on response. |\n| network_span_namer | A callable to customize the span name. |\n| tracing_attributes | Attributes to set on all created spans. |\n| permit_redirects | Whether the client allows redirects. Defaults to `True`. |\n| redirect_max | The maximum allowed redirects. Defaults to `30`. |\n| retry_total | Total number of retries to allow. Takes precedence over other counts. Default value is `10`. |\n| retry_connect | How many connection-related errors to retry on. These are errors raised before the request is sent to the remote server, which we assume has not triggered the server to process the request. Default value is `3`. |\n| retry_read | How many times to retry on read errors. These errors are raised after the request was sent to the server, so the request may have side-effects. Default value is `3`. |\n| retry_status | How many times to retry on bad status codes. Default value is `3`. |\n| retry_backoff_factor | A backoff factor to apply between attempts after the second try (most errors are resolved immediately by a second try without a delay). Retry policy will sleep for: `{backoff factor} * (2 ** ({number of total retries} - 1))` seconds. If the backoff_factor is 0.1, then the retry will sleep for [0.0s, 0.2s, 0.4s, ...] between retries. The default value is `0.8`. |\n| retry_backoff_max | The maximum back off time. Default value is `120` seconds (2 minutes). |\n| retry_mode | Fixed or exponential delay between attempts, default is `Exponential`. |\n| timeout | Timeout setting for the operation in seconds, default is `604800`s (7 days). |\n| connection_timeout | A single float in seconds for the connection timeout. Defaults to `300` seconds. |\n| read_timeout | A single float in seconds for the read timeout. Defaults to `300` seconds. |\n| connection_verify | SSL certificate verification. Enabled by default. Set to False to disable, alternatively can be set to the path to a CA_BUNDLE file or directory with certificates of trusted CAs. |\n| connection_cert | Client-side certificates. You can specify a local cert to use as client side certificate, as a single file (containing the private key and the certificate) or as a tuple of both files' paths. |\n| proxies | Dictionary mapping protocol or protocol and hostname to the URL of the proxy. |\n| cookies | Dict or CookieJar object to send with the `Request`. |\n| connection_data_block_size | The block size of data sent over the connection. Defaults to `4096` bytes. |\n\n### Async transport\n\nThe async transport is designed to be opt-in. [AioHttp](https://pypi.org/project/aiohttp/) is one of the supported implementations of async transport. It is not installed by default. You need to install it separately.\n\n### Shared modules\n\n#### MatchConditions\n\nMatchConditions is an enum to describe match conditions.\n\n```python\nclass MatchConditions(Enum):\n Unconditionally = 1 # Matches any condition\n IfNotModified = 2 # If the target object is not modified. Usually it maps to etag=\n IfModified = 3 # Only if the target object is modified. Usually it maps to etag!=\n IfPresent = 4 # If the target object exists. Usually it maps to etag='*'\n IfMissing = 5 # If the target object does not exist. Usually it maps to etag!='*'\n```\n\n#### CaseInsensitiveEnumMeta\n\nA metaclass to support case-insensitive enums.\n\n```python\nfrom enum import Enum\n\nfrom azure.core import CaseInsensitiveEnumMeta\n\nclass MyCustomEnum(str, Enum, metaclass=CaseInsensitiveEnumMeta):\n FOO = 'foo'\n BAR = 'bar'\n```\n\n#### Null Sentinel Value\n\nA falsy sentinel object which is supposed to be used to specify attributes\nwith no data. This gets serialized to `null` on the wire.\n\n```python\nfrom azure.core.serialization import NULL\n\nassert bool(NULL) is False\n\nfoo = Foo(\n attr=NULL\n)\n```\n\n## Contributing\n\nThis project welcomes contributions and suggestions. Most contributions require\nyou to agree to a Contributor License Agreement (CLA) declaring that you have\nthe right to, and actually do, grant us the rights to use your contribution.\nFor details, visit [https://cla.microsoft.com](https://cla.microsoft.com).\n\nWhen you submit a pull request, a CLA-bot will automatically determine whether\nyou need to provide a CLA and decorate the PR appropriately (e.g., label,\ncomment). Simply follow the instructions provided by the bot. You will only\nneed to do this once across all repos using our CLA.\n\nThis project has adopted the\n[Microsoft Open Source Code of Conduct](https://opensource.microsoft.com/codeofconduct/).\nFor more information, see the\n[Code of Conduct FAQ](https://opensource.microsoft.com/codeofconduct/faq/)\nor contact [opencode@microsoft.com](mailto:opencode@microsoft.com) with any\nadditional questions or comments.\n\n\n[package]: https://pypi.org/project/azure-core/\n\n\n# Release History\n\n## 1.33.0 (2025-04-03)\n\n### Features Added\n\n- Added native OpenTelemetry tracing to Azure Core which enables users to use OpenTelemetry to trace Azure SDK operations without needing to install a plugin. #39563\n - To enable native OpenTelemetry tracing, users need to:\n 1. Have `opentelemetry-api` installed.\n 2. Ensure that `settings.tracing_implementation` is not set.\n 3. Ensure that `settings.tracing_enabled` is set to `True`.\n - If `setting.tracing_implementation` is set, the tracing plugin will be used instead of the native tracing.\n - If `settings.tracing_enabled` is set to `False`, tracing will be disabled.\n - The `OpenTelemetryTracer` class was added to the `azure.core.tracing.opentelemetry` module. This is a wrapper around the OpenTelemetry tracer that is used to create spans for Azure SDK operations.\n - Added a `get_tracer` method to the new `azure.core.instrumentation` module. This method returns an instance of the `OpenTelemetryTracer` class if OpenTelemetry is available.\n - A `TracingOptions` TypedDict class was added to define the options that SDK users can use to configure tracing per-operation. These options include the ability to enable or disable tracing and set additional attributes on spans.\n - Example usage: `client.method(tracing_options={\"enabled\": True, \"attributes\": {\"foo\": \"bar\"}})`\n - The `DistributedTracingPolicy` and `distributed_trace`/`distributed_trace_async` decorators now uses the OpenTelemetry tracer if it is available and native tracing is enabled.\n - SDK clients can define an `_instrumentation_config` class variable to configure the OpenTelemetry tracer used in method span creation. Possible configuration options are `library_name`, `library_version`, `schema_url`, and `attributes`.\n - `DistributedTracingPolicy` now accepts a `instrumentation_config` keyword argument to configure the OpenTelemetry tracer used in HTTP span creation.\n\n### Breaking Changes\n\n- Removed automatic tracing enablement for the OpenTelemetry plugin if `opentelemetry` was imported. To enable tracing with the plugin, please import `azure.core.settings.settings` and set `settings.tracing_implementation` to `\"opentelemetry\"`. #39563\n- In `DistributedTracingPolicy`, the default span name is now just the HTTP method (e.g., \"GET\", \"POST\") and no longer includes the URL path. This change was made to converge with the OpenTelemetry HTTP semantic conventions. The full URL is still included in the span attributes.\n- Renamed span attributes in `DistributedTracingPolicy`:\n - \"x-ms-client-request-id\" is now \"az.client_request_id\"\n - \"x-ms-request-id\" is now \"az.service_request_id\"\n\n### Bugs Fixed\n\n- Fixed an issue where the `traceparent` header was not being set correctly in the `DistributedTracingPolicy`. The `traceparent` header will now set based on the context of the HTTP client span. #40074\n\n### Other Changes\n\n- Added `opentelemetry-api` as an optional dependency for tracing. This can be installed with `pip install azure-core[tracing]`. #39563\n\n## 1.32.0 (2024-10-31)\n\n### Features Added\n\n- Added a default implementation to handle token challenges in `BearerTokenCredentialPolicy` and `AsyncBearerTokenCredentialPolicy`.\n\n### Bugs Fixed\n\n- Fixed an issue where the `tracing_attributes` keyword argument wasn't being handled at the request/method level. #38164\n\n### Other Changes\n\n- Log \"x-vss-e2eid\" and \"x-msedge-ref\" headers in `HttpLoggingPolicy`.\n\n## 1.31.0 (2024-09-12)\n\n### Features Added\n\n- Added azure.core.AzureClouds enum to represent the different Azure clouds.\n- Added two new credential protocol classes, `SupportsTokenInfo` and `AsyncSupportsTokenInfo`, to offer more extensibility in supporting various token acquisition scenarios. #36565\n - Each new protocol class defines a `get_token_info` method that returns an `AccessTokenInfo` object.\n- Added a new `TokenRequestOptions` class, which is a `TypedDict` with optional parameters, that can be used to define options for token requests through the `get_token_info` method. #36565\n- Added a new `AccessTokenInfo` class, which is returned by `get_token_info` implementations. This class contains the token, its expiration time, and optional additional information like when a token should be refreshed. #36565\n- `BearerTokenCredentialPolicy` and `AsyncBearerTokenCredentialPolicy` now first check if a credential has the `get_token_info` method defined. If so, the `get_token_info` method is used to acquire a token. Otherwise, the `get_token` method is used. #36565\n - These policies now also check the `refresh_on` attribute when determining if a new token request should be made.\n\n### Other Changes\n\n- The Azure Core OpenTelemetry tracing plugin will now be the preferred tracing plugin over the OpenCensus plugin. If both plugins are installed and `opentelemetry` is imported, then OpenTelemetry will be used to trace Azure SDK operations. #35050\n\n## 1.30.2 (2024-06-06)\n\n### Features Added\n\n- Tracing: `DistributedTracingPolicy` will now set an attribute, `http.request.resend_count`, on HTTP spans for resent requests to indicate the resend attempt number. #35069\n\n### Bugs Fixed\n\n- Raise correct exception if transport is used while already closed #35559\n\n### Other Changes\n\n- HTTP tracing spans will now include an `error.type` attribute if an error status code is returned. #34619\n- Minimum required Python version is now 3.8\n\n## 1.30.1 (2024-02-29)\n\n### Other Changes\n\n- Accept float for `retry_after` header. #34203\n\n## 1.30.0 (2024-02-01)\n\n### Features Added\n\n- Support tuple input for file values to `azure.core.rest.HttpRequest` #33948\n- Support tuple input to `files` with duplicate field names `azure.core.rest.HttpRequest` #34021\n\n## 1.29.7 (2024-01-18)\n\n### Other Changes\n\n- Removed dependency on `anyio`. #33282\n\n## 1.29.6 (2023-12-14)\n\n### Bugs Fixed\n\n- Adjusted `AsyncBearerTokenCredentialPolicy` to work properly with `trio` concurrency mechanisms. ([#33307](https://github.com/Azure/azure-sdk-for-python/pull/33307))\n\n### Other Changes\n\n- Added dependency on `anyio` >=3.0,<5.0\n- Bumped minimum dependency on `requests` to 2.21.0.\n\n## 1.29.5 (2023-10-19)\n\n### Bugs Fixed\n\n- Fixed an issue with `multipart/form-data` in the async transport where `data` was not getting encoded into the request body. #32473\n\n### Other Changes\n\n- Use ssl context from aiohttp by default.\n\n## 1.29.4 (2023-09-07)\n\n### Bugs Fixed\n\n- Fixed the issue that some urls trigger an infinite loop. #31346\n- Fixed issue where IndexError was raised if multipart responses did not match the number of requests. #31471\n- Fixed issue unbound variable exception if dict is invalid in CloudEvent.from_dict. #31835\n- Fixed issue asyncBearerTokenCredentialPolicy is not backward compatible with SansIOHTTPPolicy. #31836\n- Fixed issue mypy complains with new version of azure-core. #31564\n\n## 1.29.3 (2023-08-22)\n\n### Bugs Fixed\n\n- Typing fix: `message` cannot be `None` in `AzureError`. #31564\n\n## 1.29.2 (2023-08-14)\n\n### Bugs Fixed\n\n- Added a default implementation for `AsyncTokenCredential.__aexit__()` #31573\n\n### Other Changes\n\n- Bumped `typing-extensions` version to 4.6.0.\n\n## 1.29.1 (2023-08-09)\n\n### Bugs Fixed\n\n- Not pass `enabled_cae` unless it is explicitly enabled.\n\n## 1.29.0 (2023-08-03)\n\n### Features Added\n\n- A keyword argument `enable_cae` was added to the `get_token` method of the `TokenCredential` protocol. #31012\n- `BearerTokenCredentialPolicy` and `AsyncBearerTokenCredentialPolicy` now accept `enable_cae` keyword arguments in their constructors. This is used in determining if [Continuous Access Evaluation (CAE)](https://learn.microsoft.com/azure/active-directory/conditional-access/concept-continuous-access-evaluation) should be enabled for each `get_token` request. #31012\n\n## 1.28.0 (2023-07-06)\n\n### Features Added\n\n- Added header name parameter to `RequestIdPolicy`. #30772\n- Added `SensitiveHeaderCleanupPolicy` that cleans up sensitive headers if a redirect happens and the new destination is in another domain. #28349\n\n### Other Changes\n\n- Catch aiohttp errors and translate them into azure-core errors.\n\n## 1.27.1 (2023-06-13)\n\n### Bugs Fixed\n\n- Fix url building for some complex query parameters scenarios #30707\n\n## 1.27.0 (2023-06-01)\n\n### Features Added\n\n- Added support to use sync credentials in `AsyncBearerTokenCredentialPolicy`. #30381\n- Added \"prefix\" parameter to AzureKeyCredentialPolicy #29901\n\n### Bugs Fixed\n\n- Improve error message when providing the wrong credential type for AzureKeyCredential #30380\n\n## 1.26.4 (2023-04-06)\n\n### Features Added\n\n- Updated settings to include OpenTelemetry as a tracer provider. #29095\n\n### Other Changes\n\n- Improved typing\n\n## 1.26.3 (2023-02-02)\n\n### Bugs Fixed\n\n- Fixed deflate decompression for aiohttp #28483\n\n## 1.26.2 (2023-01-05)\n\n### Bugs Fixed\n\n- Fix 'ClientSession' object has no attribute 'auto_decompress' (thanks to @mghextreme for the contribution)\n\n### Other Changes\n\n- Add \"x-ms-error-code\" as secure header to log\n- Rename \"DEFAULT_HEADERS_WHITELIST\" to \"DEFAULT_HEADERS_ALLOWLIST\". Added a backward compatible alias.\n\n## 1.26.1 (2022-11-03)\n\n### Other Changes\n\n- Added example of RequestsTransport with custom session. (thanks to @inirudebwoy for the contribution) #26768\n- Added Python 3.11 support.\n\n## 1.26.0 (2022-10-06)\n\n### Other Changes\n\n- LRO polling will not wait anymore before doing the first status check #26376\n- Added extra dependency for [aio]. pip install azure-core[aio] installs aiohttp too.\n\n## 1.25.1 (2022-09-01)\n\n### Bugs Fixed\n\n- Added @runtime_checkable to `TokenCredential` protocol definitions #25187\n\n## 1.25.0 (2022-08-04)\n\nAzure-core is supported on Python 3.7 or later. For more details, please read our page on [Azure SDK for Python version support policy](https://github.com/Azure/azure-sdk-for-python/wiki/Azure-SDKs-Python-version-support-policy).\n\n### Features Added\n\n- Added `CaseInsensitiveDict` implementation in `azure.core.utils` removing dependency on `requests` and `aiohttp`\n\n## 1.24.2 (2022-06-30)\n\n### Bugs Fixed\n\n- Fixed the bug that azure-core could not be imported under Python 3.11.0b3 #24928\n- `ContentDecodePolicy` can now correctly deserialize more JSON bodies with different mime types #22410\n\n## 1.24.1 (2022-06-01)\n\n### Bugs Fixed\n\n- Declare method level span as INTERNAL by default #24492\n- Fixed type hints for `azure.core.paging.ItemPaged` #24548\n\n## 1.24.0 (2022-05-06)\n\n### Features Added\n\n- Add `SerializationError` and `DeserializationError` in `azure.core.exceptions` for errors raised during serialization / deserialization #24312\n\n## 1.23.1 (2022-03-31)\n\n### Bugs Fixed\n\n- Allow stream inputs to the `content` kwarg of `azure.core.rest.HttpRequest` from objects with a `read` method #23578\n\n## 1.23.0 (2022-03-03)\n\n### Features Added\n\n- Improve intellisense type hinting for service client methods. #22891\n\n- Add a case insensitive dict `case_insensitive_dict` in `azure.core.utils`. #23206\n\n### Bugs Fixed\n\n- Use \"\\n\" rather than \"/n\" for new line in log. #23261\n\n### Other Changes\n\n- Log \"WWW-Authenticate\" header in `HttpLoggingPolicy` #22990\n- Added dependency on `typing-extensions` >= 4.0.1\n\n## 1.22.1 (2022-02-09)\n\n### Bugs Fixed\n\n- Limiting `final-state-via` scope to POST until consuming SDKs has been fixed to use this option properly on PUT. #22989\n\n## 1.22.0 (2022-02-03)\n_[**This version is deprecated.**]_\n\n### Features Added\n\n- Add support for `final-state-via` LRO option in core. #22713\n\n### Bugs Fixed\n\n- Add response body to string representation of `HttpResponseError` if we're not able to parse out information #22302\n- Raise `AttributeError` when calling azure.core.pipeline.transport.\\_\\_bases__ #22469\n\n### Other Changes\n\n- Python 2.7 is no longer supported. Please use Python version 3.6 or later.\n\n## 1.21.1 (2021-12-06)\n\n### Other Changes\n\n- Revert change in str method #22023\n\n## 1.21.0 (2021-12-02)\n\n### Breaking Changes\n\n- Sync stream downloading now raises `azure.core.exceptions.DecodeError` rather than `requests.exceptions.ContentDecodingError`\n\n### Bugs Fixed\n\n- Add response body to string representation of `HttpResponseError` if we're not able to parse out information #21800\n\n## 1.20.1 (2021-11-08)\n\n### Bugs Fixed\n\n- Correctly set response's content to decompressed body when users are using aiohttp transport with decompression headers #21620\n\n## 1.20.0 (2021-11-04)\n\n### Features Added\n\n- GA `send_request` onto the `azure.core.PipelineClient` and `azure.core.AsyncPipelineClient`. This method takes in\nrequests and sends them through our pipelines.\n- GA `azure.core.rest`. `azure.core.rest` is our new public simple HTTP library in `azure.core` that users will use to create requests, and consume responses.\n- GA errors `StreamConsumedError`, `StreamClosedError`, and `ResponseNotReadError` to `azure.core.exceptions`. These errors\nare thrown if you mishandle streamed responses from the `azure.core.rest` module\n- add kwargs to the methods for `iter_raw` and `iter_bytes` #21529\n- no longer raise JSON errors if users pass in file descriptors of JSON to the `json` kwarg in `HttpRequest` #21504\n- Added new error type `IncompleteReadError` which is raised if peer closes the connection before we have received the complete message body.\n\n### Breaking Changes\n\n- SansIOHTTPPolicy.on_exception returns None instead of bool.\n\n### Bugs Fixed\n\n- The `Content-Length` header in a http response is strictly checked against the actual number of bytes in the body,\n rather than silently truncating data in case the underlying tcp connection is closed prematurely.\n (thanks to @jochen-ott-by for the contribution) #20412\n- UnboundLocalError when SansIOHTTPPolicy handles an exception #15222\n- Add default content type header of `text/plain` and content length header for users who pass unicode strings to the `content` kwarg of `HttpRequest` in 2.7 #21550\n\n## 1.19.1 (2021-11-01)\n\n### Bugs Fixed\n\n- respect text encoding specified in argument (thanks to @ryohji for the contribution) #20796\n- Fix \"coroutine x.read() was never awaited\" warning from `ContentDecodePolicy` #21318\n- fix type check for `data` input to `azure.core.rest` for python 2.7 users #21341\n- use `charset_normalizer` if `chardet` is not installed to migrate aiohttp 3.8.0 changes.\n\n### Other Changes\n\n- Refactor AzureJSONEncoder (thanks to @Codejune for the contribution) #21028\n\n## 1.19.0 (2021-09-30)\n\n### Breaking Changes in the Provisional `azure.core.rest` package\n\n- `azure.core.rest.HttpResponse` and `azure.core.rest.AsyncHttpResponse` are now abstract base classes. They should not be initialized directly, instead\nyour transport responses should inherit from them and implement them.\n- The properties of the `azure.core.rest` responses are now all read-only\n\n- HttpLoggingPolicy integrates logs into one record #19925\n\n## 1.18.0 (2021-09-02)\n\n### Features Added\n\n- `azure.core.serialization.AzureJSONEncoder` (introduced in 1.17.0) serializes `datetime.datetime` objects in ISO 8601 format, conforming to RFC 3339's specification. #20190\n- We now use `azure.core.serialization.AzureJSONEncoder` to serialize `json` input to `azure.core.rest.HttpRequest`.\n\n### Breaking Changes in the Provisional `azure.core.rest` package\n\n- The `text` property on `azure.core.rest.HttpResponse` and `azure.core.rest.AsyncHttpResponse` has changed to a method, which also takes\nan `encoding` parameter.\n- Removed `iter_text` and `iter_lines` from `azure.core.rest.HttpResponse` and `azure.core.rest.AsyncHttpResponse`\n\n### Bugs Fixed\n\n- The behaviour of the headers returned in `azure.core.rest` responses now aligns across sync and async. Items can now be checked case-insensitively and without raising an error for format.\n\n## 1.17.0 (2021-08-05)\n\n### Features Added\n\n- Cut hard dependency on requests library\n- Added a `from_json` method which now accepts storage QueueMessage, eventhub's EventData or ServiceBusMessage or simply json bytes to return a `CloudEvent`\n\n### Fixed\n\n- Not override \"x-ms-client-request-id\" if it already exists in the header. #17757\n\n### Breaking Changes in the Provisional `azure.core.rest` package\n\n- `azure.core.rest` will not try to guess the `charset` anymore if it was impossible to extract it from `HttpResponse` analysis. This removes our dependency on `charset`.\n\n## 1.16.0 (2021-07-01)\n\n### Features Added\n\n- Add new ***provisional*** methods `send_request` onto the `azure.core.PipelineClient` and `azure.core.AsyncPipelineClient`. This method takes in\nrequests and sends them through our pipelines.\n- Add new ***provisional*** module `azure.core.rest`. `azure.core.rest` is our new public simple HTTP library in `azure.core` that users will use to create requests, and consume responses.\n- Add new ***provisional*** errors `StreamConsumedError`, `StreamClosedError`, and `ResponseNotReadError` to `azure.core.exceptions`. These errors\nare thrown if you mishandle streamed responses from the provisional `azure.core.rest` module\n\n### Fixed\n\n- Improved error message in the `from_dict` method of `CloudEvent` when a wrong schema is sent.\n\n## 1.15.0 (2021-06-04)\n\n### New Features\n\n- Added `BearerTokenCredentialPolicy.on_challenge` and `.authorize_request` to allow subclasses to optionally handle authentication challenges\n\n### Bug Fixes\n\n- Retry policies don't sleep after operations time out\n- The `from_dict` methhod in the `CloudEvent` can now convert a datetime string to datetime object when microsecond exceeds the python limitation\n\n## 1.14.0 (2021-05-13)\n\n### New Features\n\n- Added `azure.core.credentials.AzureNamedKeyCredential` credential #17548.\n- Added `decompress` parameter for `stream_download` method. If it is set to `False`, will not do decompression upon the stream. #17920\n\n## 1.13.0 (2021-04-02)\n\nAzure core requires Python 2.7 or Python 3.6+ since this release.\n\n### New Features\n\n- Added `azure.core.utils.parse_connection_string` function to parse connection strings across SDKs, with common validation and support for case insensitive keys.\n- Supported adding custom policies #16519\n- Added `~azure.core.tracing.Link` that should be used while passing `Links` to `AbstractSpan`.\n- `AbstractSpan` constructor can now take in additional keyword only args.\n\n### Bug fixes\n\n- Make NetworkTraceLoggingPolicy show the auth token in plain text. #14191\n- Fixed RetryPolicy overriding default connection timeout with an extreme value #17481\n\n## 1.12.0 (2021-03-08)\n\nThis version will be the last version to officially support Python 3.5, future versions will require Python 2.7 or Python 3.6+.\n\n### Features\n\n- Added `azure.core.messaging.CloudEvent` model that follows the cloud event spec.\n- Added `azure.core.serialization.NULL` sentinel value\n- Improve `repr`s for `HttpRequest` and `HttpResponse`s #16972\n\n### Bug Fixes\n\n- Disable retry in stream downloading. (thanks to @jochen-ott-by @hoffmann for the contribution) #16723\n\n## 1.11.0 (2021-02-08)\n\n### Features\n\n- Added `CaseInsensitiveEnumMeta` class for case-insensitive enums. #16316\n- Add `raise_for_status` method onto `HttpResponse`. Calling `response.raise_for_status()` on a response with an error code\nwill raise an `HttpResponseError`. Calling it on a good response will do nothing #16399\n\n### Bug Fixes\n\n- Update conn.conn_kw rather than overriding it when setting block size. (thanks for @jiasli for the contribution) #16587\n\n## 1.10.0 (2021-01-11)\n\n### Features\n\n- Added `AzureSasCredential` and its respective policy. #15946\n\n## 1.9.0 (2020-11-09)\n\n### Features\n\n- Add a `continuation_token` attribute to the base `AzureError` exception, and set this value for errors raised\n during paged or long-running operations.\n\n### Bug Fixes\n\n- Set retry_interval to 1 second instead of 1000 seconds (thanks **vbarbaresi** for contributing) #14357\n\n\n## 1.8.2 (2020-10-05)\n\n### Bug Fixes\n\n- Fixed bug to allow polling in the case of parameterized endpoints with relative polling urls #14097\n\n\n## 1.8.1 (2020-09-08)\n\n### Bug fixes\n\n- SAS credential replicated \"/\" fix #13159\n\n## 1.8.0 (2020-08-10)\n\n### Features\n\n- Support params as list for exploding parameters #12410\n\n\n## 1.7.0 (2020-07-06)\n\n### Bug fixes\n\n- `AzureKeyCredentialPolicy` will now accept (and ignore) passed in kwargs #11963\n- Better error messages if passed endpoint is incorrect #12106\n- Do not JSON encore a string if content type is \"text\" #12137\n\n### Features\n\n- Added `http_logging_policy` property on the `Configuration` object, allowing users to individually\nset the http logging policy of the config #12218\n\n## 1.6.0 (2020-06-03)\n\n### Bug fixes\n\n- Fixed deadlocks in AsyncBearerTokenCredentialPolicy #11543\n- Fix AttributeException in StreamDownloadGenerator #11462\n\n### Features\n\n- Added support for changesets as part of multipart message support #10485\n- Add AsyncLROPoller in azure.core.polling #10801\n- Add get_continuation_token/from_continuation_token/polling_method methods in pollers (sync and async) #10801\n- HttpResponse and PipelineContext objects are now pickable #10801\n\n## 1.5.0 (2020-05-04)\n\n### Features\n\n- Support \"x-ms-retry-after-ms\" in response header #10743\n- `link` and `link_from_headers` now accepts attributes #10765\n\n### Bug fixes\n\n- Not retry if the status code is less than 400 #10778\n- \"x-ms-request-id\" is not considered safe header for logging #10967\n\n## 1.4.0 (2020-04-06)\n\n### Features\n\n- Support a default error type in map_error #9773\n- Added `AzureKeyCredential` and its respective policy. #10509\n- Added `azure.core.polling.base_polling` module with a \"Microsoft One API\" polling implementation #10090\n Also contains the async version in `azure.core.polling.async_base_polling`\n- Support kwarg `enforce_https` to disable HTTPS check on authentication #9821\n- Support additional kwargs in `HttpRequest.set_multipart_mixed` that will be passed into pipeline context.\n\n## 1.3.0 (2020-03-09)\n\n### Bug fixes\n\n- Appended RequestIdPolicy to the default pipeline #9841\n- Rewind the body position in async_retry #10117\n\n### Features\n\n- Add raw_request_hook support in custom_hook_policy #9958\n- Add timeout support in retry_policy #10011\n- Add OdataV4 error format auto-parsing in all exceptions ('error' attribute) #9738\n\n## 1.2.2 (2020-02-10)\n\n### Bug fixes\n\n- Fixed a bug that sends None as request_id #9545\n- Enable mypy for customers #9572\n- Handle TypeError in deep copy #9620\n- Fix text/plain content-type in decoder #9589\n\n## 1.2.1 (2020-01-14)\n\n### Bug fixes\n\n- Fixed a regression in 1.2.0 that was incompatible with azure-keyvault-* 4.0.0\n[#9462](https://github.com/Azure/azure-sdk-for-python/issues/9462)\n\n\n## 1.2.0 (2020-01-14)\n\n### Features\n\n- Add user_agent & sdk_moniker kwargs in UserAgentPolicy init #9355\n- Support OPTIONS HTTP verb #9322\n- Add tracing_attributes to tracing decorator #9297\n- Support auto_request_id in RequestIdPolicy #9163\n- Support fixed retry #6419\n- Support \"retry-after-ms\" in response header #9240\n\n### Bug fixes\n\n- Removed `__enter__` and `__exit__` from async context managers #9313\n\n## 1.1.1 (2019-12-03)\n\n### Bug fixes\n\n- Bearer token authorization requires HTTPS\n- Rewind the body position in retry #8307\n\n## 1.1.0 (2019-11-25)\n\n### Features\n\n- New RequestIdPolicy #8437\n- Enable logging policy in default pipeline #8053\n- Normalize transport timeout. #8000\n Now we have:\n * 'connection_timeout' - a single float in seconds for the connection timeout. Default 5min\n * 'read_timeout' - a single float in seconds for the read timeout. Default 5min\n\n### Bug fixes\n\n- RequestHistory: deepcopy fails if request contains a stream #7732\n- Retry: retry raises error if response does not have http_response #8629\n- Client kwargs are now passed to DistributedTracingPolicy correctly #8051\n- NetworkLoggingPolicy now logs correctly all requests in case of retry #8262\n\n## 1.0.0 (2019-10-29)\n\n### Features\n\n- Tracing: DistributedTracingPolicy now accepts kwargs network_span_namer to change network span name #7773\n- Tracing: Implementation of AbstractSpan can now use the mixin HttpSpanMixin to get HTTP span update automatically #7773\n- Tracing: AbstractSpan contract \"change_context\" introduced #7773\n- Introduce new policy HttpLoggingPolicy #7988\n\n### Bug fixes\n\n- Fix AsyncioRequestsTransport if input stream is an async generator #7743\n- Fix form-data with aiohttp transport #7749\n\n### Breaking changes\n\n- Tracing: AbstractSpan.set_current_span is longer supported. Use change_context instead. #7773\n- azure.core.pipeline.policies.ContentDecodePolicy.deserialize_from_text changed\n\n## 1.0.0b4 (2019-10-07)\n\n### Features\n\n- Tracing: network span context is available with the TRACING_CONTEXT in pipeline response #7252\n- Tracing: Span contract now has `kind`, `traceparent` and is a context manager #7252\n- SansIOHTTPPolicy methods can now be coroutines #7497\n- Add multipart/mixed support #7083:\n\n - HttpRequest now has a \"set_multipart_mixed\" method to set the parts of this request\n - HttpRequest now has a \"prepare_multipart_body\" method to build final body.\n - HttpResponse now has a \"parts\" method to return an iterator of parts\n - AsyncHttpResponse now has a \"parts\" methods to return an async iterator of parts\n - Note that multipart/mixed is a Python 3.x only feature\n\n### Bug fixes\n\n- Tracing: policy cannot fail the pipeline, even in the worst condition #7252\n- Tracing: policy pass correctly status message if exception #7252\n- Tracing: incorrect span if exception raised from decorated function #7133\n- Fixed urllib3 ConnectTimeoutError being raised by Requests during a socket timeout. Now this exception is caught and wrapped as a `ServiceRequestError` #7542\n\n### Breaking changes\n\n- Tracing: `azure.core.tracing.context` removed\n- Tracing: `azure.core.tracing.context.tracing_context.with_current_context` renamed to `azure.core.tracing.common.with_current_context` #7252\n- Tracing: `link` renamed `link_from_headers` and `link` takes now a string\n- Tracing: opencensus implementation has been moved to the package `azure-core-tracing-opencensus`\n- Some modules and classes that were importables from several different places have been removed:\n\n - `azure.core.HttpResponseError` is now only `azure.core.exceptions.HttpResponseError`\n - `azure.core.Configuration` is now only `azure.core.configuration.Configuration`\n - `azure.core.HttpRequest` is now only `azure.core.pipeline.transport.HttpRequest`\n - `azure.core.version` module has been removed. Use `azure.core.__version__` to get version number.\n - `azure.core.pipeline_client` has been removed. Import from `azure.core` instead.\n - `azure.core.pipeline_client_async` has been removed. Import from `azure.core` instead.\n - `azure.core.pipeline.base` has been removed. Import from `azure.core.pipeline` instead.\n - `azure.core.pipeline.base_async` has been removed. Import from `azure.core.pipeline` instead.\n - `azure.core.pipeline.policies.base` has been removed. Import from `azure.core.pipeline.policies` instead.\n - `azure.core.pipeline.policies.base_async` has been removed. Import from `azure.core.pipeline.policies` instead.\n - `azure.core.pipeline.policies.authentication` has been removed. Import from `azure.core.pipeline.policies` instead.\n - `azure.core.pipeline.policies.authentication_async` has been removed. Import from `azure.core.pipeline.policies` instead.\n - `azure.core.pipeline.policies.custom_hook` has been removed. Import from `azure.core.pipeline.policies` instead.\n - `azure.core.pipeline.policies.redirect` has been removed. Import from `azure.core.pipeline.policies` instead.\n - `azure.core.pipeline.policies.redirect_async` has been removed. Import from `azure.core.pipeline.policies` instead.\n - `azure.core.pipeline.policies.retry` has been removed. Import from `azure.core.pipeline.policies` instead.\n - `azure.core.pipeline.policies.retry_async` has been removed. Import from `azure.core.pipeline.policies` instead.\n - `azure.core.pipeline.policies.distributed_tracing` has been removed. Import from `azure.core.pipeline.policies` instead.\n - `azure.core.pipeline.policies.universal` has been removed. Import from `azure.core.pipeline.policies` instead.\n - `azure.core.tracing.abstract_span` has been removed. Import from `azure.core.tracing` instead.\n - `azure.core.pipeline.transport.base` has been removed. Import from `azure.core.pipeline.transport` instead.\n - `azure.core.pipeline.transport.base_async` has been removed. Import from `azure.core.pipeline.transport` instead.\n - `azure.core.pipeline.transport.requests_basic` has been removed. Import from `azure.core.pipeline.transport` instead.\n - `azure.core.pipeline.transport.requests_asyncio` has been removed. Import from `azure.core.pipeline.transport` instead.\n - `azure.core.pipeline.transport.requests_trio` has been removed. Import from `azure.core.pipeline.transport` instead.\n - `azure.core.pipeline.transport.aiohttp` has been removed. Import from `azure.core.pipeline.transport` instead.\n - `azure.core.polling.poller` has been removed. Import from `azure.core.polling` instead.\n - `azure.core.polling.async_poller` has been removed. Import from `azure.core.polling` instead.\n\n## 1.0.0b3 (2019-09-09)\n\n### Bug fixes\n\n- Fix aiohttp auto-headers #6992\n- Add tracing to policies module init #6951\n\n## 1.0.0b2 (2019-08-05)\n\n### Breaking changes\n\n- Transport classes don't take `config` parameter anymore (use kwargs instead) #6372\n- `azure.core.paging` has been completely refactored #6420\n- HttpResponse.content_type attribute is now a string (was a list) #6490\n- For `StreamDownloadGenerator` subclasses, `response` is now an `HttpResponse`, and not a transport response like `aiohttp.ClientResponse` or `requests.Response`. The transport response is available in `internal_response` attribute #6490\n\n### Bug fixes\n\n- aiohttp is not required to import async pipelines classes #6496\n- `AsyncioRequestsTransport.sleep` is now a coroutine as expected #6490\n- `RequestsTransport` is not tight to `ProxyPolicy` implementation details anymore #6372\n- `AiohttpTransport` does not raise on unexpected kwargs #6355\n\n### Features\n\n- New paging base classes that support `continuation_token` and `by_page()` #6420\n- Proxy support for `AiohttpTransport` #6372\n\n## 1.0.0b1 (2019-06-26)\n\n- Preview 1 release", - "release_date": "2025-04-03T23:51:03", + "description": "Microsoft Azure Core Library for Python\n# Azure Core shared client library for Python\n\nAzure core provides shared exceptions and modules for Python SDK client libraries.\nThese libraries follow the [Azure SDK Design Guidelines for Python](https://azure.github.io/azure-sdk/python/guidelines/index.html) .\n\nIf you are a client library developer, please reference [client library developer reference](https://github.com/Azure/azure-sdk-for-python/blob/main/sdk/core/azure-core/CLIENT_LIBRARY_DEVELOPER.md) for more information.\n\n[Source code](https://github.com/Azure/azure-sdk-for-python/blob/main/sdk/core/azure-core/)\n| [Package (Pypi)][package]\n| [Package (Conda)](https://anaconda.org/microsoft/azure-core/)\n| [API reference documentation](https://learn.microsoft.com/python/api/overview/azure/core-readme)\n\n## Getting started\n\nTypically, you will not need to install azure core;\nit will be installed when you install one of the client libraries using it.\nIn case you want to install it explicitly (to implement your own client library, for example),\nyou can find it [here](https://pypi.org/project/azure-core/).\n\n## Key concepts\n\n### Azure Core Library Exceptions\n\n#### AzureError\n\nAzureError is the base exception for all errors.\n\n```python\nclass AzureError(Exception):\n def __init__(self, message, *args, **kwargs):\n self.inner_exception = kwargs.get(\"error\")\n self.exc_type, self.exc_value, self.exc_traceback = sys.exc_info()\n self.exc_type = self.exc_type.__name__ if self.exc_type else type(self.inner_exception)\n self.exc_msg = \"{}, {}: {}\".format(message, self.exc_type, self.exc_value) # type: ignore\n self.message = str(message)\n self.continuation_token = kwargs.get(\"continuation_token\")\n super(AzureError, self).__init__(self.message, *args)\n```\n\n*message* is any message (str) to be associated with the exception.\n\n*args* are any additional args to be included with exception.\n\n*kwargs* are keyword arguments to include with the exception. Use the keyword *error* to pass in an internal exception and *continuation_token* for a token reference to continue an incomplete operation.\n\n**The following exceptions inherit from AzureError:**\n\n#### ServiceRequestError\n\nAn error occurred while attempt to make a request to the service. No request was sent.\n\n#### ServiceResponseError\n\nThe request was sent, but the client failed to understand the response.\nThe connection may have timed out. These errors can be retried for idempotent or safe operations.\n\n#### HttpResponseError\n\nA request was made, and a non-success status code was received from the service.\n\n```python\nclass HttpResponseError(AzureError):\n def __init__(self, message=None, response=None, **kwargs):\n self.reason = None\n self.response = response\n if response:\n self.reason = response.reason\n self.status_code = response.status_code\n self.error = self._parse_odata_body(ODataV4Format, response) # type: Optional[ODataV4Format]\n if self.error:\n message = str(self.error)\n else:\n message = message or \"Operation returned an invalid status '{}'\".format(\n self.reason\n )\n\n super(HttpResponseError, self).__init__(message=message, **kwargs)\n```\n\n*message* is the HTTP response error message (optional)\n\n*response* is the HTTP response (optional).\n\n*kwargs* are keyword arguments to include with the exception.\n\n**The following exceptions inherit from HttpResponseError:**\n\n#### DecodeError\n\nAn error raised during response de-serialization.\n\n#### IncompleteReadError\n\nAn error raised if peer closes the connection before we have received the complete message body.\n\n#### ResourceExistsError\n\nAn error response with status code 4xx. This will not be raised directly by the Azure core pipeline.\n\n#### ResourceNotFoundError\n\nAn error response, typically triggered by a 412 response (for update) or 404 (for get/post).\n\n#### ResourceModifiedError\n\nAn error response with status code 4xx, typically 412 Conflict. This will not be raised directly by the Azure core pipeline.\n\n#### ResourceNotModifiedError\n\nAn error response with status code 304. This will not be raised directly by the Azure core pipeline.\n\n#### ClientAuthenticationError\n\nAn error response with status code 4xx. This will not be raised directly by the Azure core pipeline.\n\n#### TooManyRedirectsError\n\nAn error raised when the maximum number of redirect attempts is reached. The maximum amount of redirects can be configured in the RedirectPolicy.\n\n```python\nclass TooManyRedirectsError(HttpResponseError):\n def __init__(self, history, *args, **kwargs):\n self.history = history\n message = \"Reached maximum redirect attempts.\"\n super(TooManyRedirectsError, self).__init__(message, *args, **kwargs)\n```\n\n*history* is used to document the requests/responses that resulted in redirected requests.\n\n*args* are any additional args to be included with exception.\n\n*kwargs* are keyword arguments to include with the exception.\n\n#### StreamConsumedError\n\nAn error thrown if you try to access the stream of `azure.core.rest.HttpResponse` or `azure.core.rest.AsyncHttpResponse` once\nthe response stream has been consumed.\n\n#### StreamClosedError\n\nAn error thrown if you try to access the stream of the `azure.core.rest.HttpResponse` or `azure.core.rest.AsyncHttpResponse` once\nthe response stream has been closed.\n\n#### ResponseNotReadError\n\nAn error thrown if you try to access the `content` of `azure.core.rest.HttpResponse` or `azure.core.rest.AsyncHttpResponse` before\nreading in the response's bytes first.\n\n### Configurations\n\nWhen calling the methods, some properties can be configured by passing in as kwargs arguments.\n\n| Parameters | Description |\n| --- | --- |\n| headers | The HTTP Request headers. |\n| request_id | The request id to be added into header. |\n| user_agent | If specified, this will be added in front of the user agent string. |\n| logging_enable| Use to enable per operation. Defaults to `False`. |\n| logger | If specified, it will be used to log information. |\n| response_encoding | The encoding to use if known for this service (will disable auto-detection). |\n| raw_request_hook | Callback function. Will be invoked on request. |\n| raw_response_hook | Callback function. Will be invoked on response. |\n| network_span_namer | A callable to customize the span name. |\n| tracing_attributes | Attributes to set on all created spans. |\n| permit_redirects | Whether the client allows redirects. Defaults to `True`. |\n| redirect_max | The maximum allowed redirects. Defaults to `30`. |\n| retry_total | Total number of retries to allow. Takes precedence over other counts. Default value is `10`. |\n| retry_connect | How many connection-related errors to retry on. These are errors raised before the request is sent to the remote server, which we assume has not triggered the server to process the request. Default value is `3`. |\n| retry_read | How many times to retry on read errors. These errors are raised after the request was sent to the server, so the request may have side-effects. Default value is `3`. |\n| retry_status | How many times to retry on bad status codes. Default value is `3`. |\n| retry_backoff_factor | A backoff factor to apply between attempts after the second try (most errors are resolved immediately by a second try without a delay). Retry policy will sleep for: `{backoff factor} * (2 ** ({number of total retries} - 1))` seconds. If the backoff_factor is 0.1, then the retry will sleep for [0.0s, 0.2s, 0.4s, ...] between retries. The default value is `0.8`. |\n| retry_backoff_max | The maximum back off time. Default value is `120` seconds (2 minutes). |\n| retry_mode | Fixed or exponential delay between attempts, default is `Exponential`. |\n| timeout | Timeout setting for the operation in seconds, default is `604800`s (7 days). |\n| connection_timeout | A single float in seconds for the connection timeout. Defaults to `300` seconds. |\n| read_timeout | A single float in seconds for the read timeout. Defaults to `300` seconds. |\n| connection_verify | SSL certificate verification. Enabled by default. Set to False to disable, alternatively can be set to the path to a CA_BUNDLE file or directory with certificates of trusted CAs. |\n| connection_cert | Client-side certificates. You can specify a local cert to use as client side certificate, as a single file (containing the private key and the certificate) or as a tuple of both files' paths. |\n| proxies | Dictionary mapping protocol or protocol and hostname to the URL of the proxy. |\n| cookies | Dict or CookieJar object to send with the `Request`. |\n| connection_data_block_size | The block size of data sent over the connection. Defaults to `4096` bytes. |\n\n### Async transport\n\nThe async transport is designed to be opt-in. [AioHttp](https://pypi.org/project/aiohttp/) is one of the supported implementations of async transport. It is not installed by default. You need to install it separately.\n\n### Shared modules\n\n#### MatchConditions\n\nMatchConditions is an enum to describe match conditions.\n\n```python\nclass MatchConditions(Enum):\n Unconditionally = 1 # Matches any condition\n IfNotModified = 2 # If the target object is not modified. Usually it maps to etag=\n IfModified = 3 # Only if the target object is modified. Usually it maps to etag!=\n IfPresent = 4 # If the target object exists. Usually it maps to etag='*'\n IfMissing = 5 # If the target object does not exist. Usually it maps to etag!='*'\n```\n\n#### CaseInsensitiveEnumMeta\n\nA metaclass to support case-insensitive enums.\n\n```python\nfrom enum import Enum\n\nfrom azure.core import CaseInsensitiveEnumMeta\n\nclass MyCustomEnum(str, Enum, metaclass=CaseInsensitiveEnumMeta):\n FOO = 'foo'\n BAR = 'bar'\n```\n\n#### Null Sentinel Value\n\nA falsy sentinel object which is supposed to be used to specify attributes\nwith no data. This gets serialized to `null` on the wire.\n\n```python\nfrom azure.core.serialization import NULL\n\nassert bool(NULL) is False\n\nfoo = Foo(\n attr=NULL\n)\n```\n\n## Contributing\n\nThis project welcomes contributions and suggestions. Most contributions require\nyou to agree to a Contributor License Agreement (CLA) declaring that you have\nthe right to, and actually do, grant us the rights to use your contribution.\nFor details, visit [https://cla.microsoft.com](https://cla.microsoft.com).\n\nWhen you submit a pull request, a CLA-bot will automatically determine whether\nyou need to provide a CLA and decorate the PR appropriately (e.g., label,\ncomment). Simply follow the instructions provided by the bot. You will only\nneed to do this once across all repos using our CLA.\n\nThis project has adopted the\n[Microsoft Open Source Code of Conduct](https://opensource.microsoft.com/codeofconduct/).\nFor more information, see the\n[Code of Conduct FAQ](https://opensource.microsoft.com/codeofconduct/faq/)\nor contact [opencode@microsoft.com](mailto:opencode@microsoft.com) with any\nadditional questions or comments.\n\n\n[package]: https://pypi.org/project/azure-core/\n\n\n# Release History\n\n## 1.34.0 (2025-05-01)\n\n### Features Added\n\n- Added a `set_span_error_status` method to the `OpenTelemetryTracer` class. This method allows users to set the status of a span to `ERROR` after it has been created. #40703\n\n### Other Changes\n\n- Python 3.8 is no longer supported. Please use Python version 3.9 or later.\n\n## 1.33.0 (2025-04-03)\n\n### Features Added\n\n- Added native OpenTelemetry tracing to Azure Core which enables users to use OpenTelemetry to trace Azure SDK operations without needing to install a plugin. #39563\n - To enable native OpenTelemetry tracing, users need to:\n 1. Have `opentelemetry-api` installed.\n 2. Ensure that `settings.tracing_implementation` is not set.\n 3. Ensure that `settings.tracing_enabled` is set to `True`.\n - If `setting.tracing_implementation` is set, the tracing plugin will be used instead of the native tracing.\n - If `settings.tracing_enabled` is set to `False`, tracing will be disabled.\n - The `OpenTelemetryTracer` class was added to the `azure.core.tracing.opentelemetry` module. This is a wrapper around the OpenTelemetry tracer that is used to create spans for Azure SDK operations.\n - Added a `get_tracer` method to the new `azure.core.instrumentation` module. This method returns an instance of the `OpenTelemetryTracer` class if OpenTelemetry is available.\n - A `TracingOptions` TypedDict class was added to define the options that SDK users can use to configure tracing per-operation. These options include the ability to enable or disable tracing and set additional attributes on spans.\n - Example usage: `client.method(tracing_options={\"enabled\": True, \"attributes\": {\"foo\": \"bar\"}})`\n - The `DistributedTracingPolicy` and `distributed_trace`/`distributed_trace_async` decorators now uses the OpenTelemetry tracer if it is available and native tracing is enabled.\n - SDK clients can define an `_instrumentation_config` class variable to configure the OpenTelemetry tracer used in method span creation. Possible configuration options are `library_name`, `library_version`, `schema_url`, and `attributes`.\n - `DistributedTracingPolicy` now accepts a `instrumentation_config` keyword argument to configure the OpenTelemetry tracer used in HTTP span creation.\n\n### Breaking Changes\n\n- Removed automatic tracing enablement for the OpenTelemetry plugin if `opentelemetry` was imported. To enable tracing with the plugin, please import `azure.core.settings.settings` and set `settings.tracing_implementation` to `\"opentelemetry\"`. #39563\n- In `DistributedTracingPolicy`, the default span name is now just the HTTP method (e.g., \"GET\", \"POST\") and no longer includes the URL path. This change was made to converge with the OpenTelemetry HTTP semantic conventions. The full URL is still included in the span attributes.\n- Renamed span attributes in `DistributedTracingPolicy`:\n - \"x-ms-client-request-id\" is now \"az.client_request_id\"\n - \"x-ms-request-id\" is now \"az.service_request_id\"\n\n### Bugs Fixed\n\n- Fixed an issue where the `traceparent` header was not being set correctly in the `DistributedTracingPolicy`. The `traceparent` header will now set based on the context of the HTTP client span. #40074\n\n### Other Changes\n\n- Added `opentelemetry-api` as an optional dependency for tracing. This can be installed with `pip install azure-core[tracing]`. #39563\n\n## 1.32.0 (2024-10-31)\n\n### Features Added\n\n- Added a default implementation to handle token challenges in `BearerTokenCredentialPolicy` and `AsyncBearerTokenCredentialPolicy`.\n\n### Bugs Fixed\n\n- Fixed an issue where the `tracing_attributes` keyword argument wasn't being handled at the request/method level. #38164\n\n### Other Changes\n\n- Log \"x-vss-e2eid\" and \"x-msedge-ref\" headers in `HttpLoggingPolicy`.\n\n## 1.31.0 (2024-09-12)\n\n### Features Added\n\n- Added azure.core.AzureClouds enum to represent the different Azure clouds.\n- Added two new credential protocol classes, `SupportsTokenInfo` and `AsyncSupportsTokenInfo`, to offer more extensibility in supporting various token acquisition scenarios. #36565\n - Each new protocol class defines a `get_token_info` method that returns an `AccessTokenInfo` object.\n- Added a new `TokenRequestOptions` class, which is a `TypedDict` with optional parameters, that can be used to define options for token requests through the `get_token_info` method. #36565\n- Added a new `AccessTokenInfo` class, which is returned by `get_token_info` implementations. This class contains the token, its expiration time, and optional additional information like when a token should be refreshed. #36565\n- `BearerTokenCredentialPolicy` and `AsyncBearerTokenCredentialPolicy` now first check if a credential has the `get_token_info` method defined. If so, the `get_token_info` method is used to acquire a token. Otherwise, the `get_token` method is used. #36565\n - These policies now also check the `refresh_on` attribute when determining if a new token request should be made.\n\n### Other Changes\n\n- The Azure Core OpenTelemetry tracing plugin will now be the preferred tracing plugin over the OpenCensus plugin. If both plugins are installed and `opentelemetry` is imported, then OpenTelemetry will be used to trace Azure SDK operations. #35050\n\n## 1.30.2 (2024-06-06)\n\n### Features Added\n\n- Tracing: `DistributedTracingPolicy` will now set an attribute, `http.request.resend_count`, on HTTP spans for resent requests to indicate the resend attempt number. #35069\n\n### Bugs Fixed\n\n- Raise correct exception if transport is used while already closed #35559\n\n### Other Changes\n\n- HTTP tracing spans will now include an `error.type` attribute if an error status code is returned. #34619\n- Minimum required Python version is now 3.8\n\n## 1.30.1 (2024-02-29)\n\n### Other Changes\n\n- Accept float for `retry_after` header. #34203\n\n## 1.30.0 (2024-02-01)\n\n### Features Added\n\n- Support tuple input for file values to `azure.core.rest.HttpRequest` #33948\n- Support tuple input to `files` with duplicate field names `azure.core.rest.HttpRequest` #34021\n\n## 1.29.7 (2024-01-18)\n\n### Other Changes\n\n- Removed dependency on `anyio`. #33282\n\n## 1.29.6 (2023-12-14)\n\n### Bugs Fixed\n\n- Adjusted `AsyncBearerTokenCredentialPolicy` to work properly with `trio` concurrency mechanisms. ([#33307](https://github.com/Azure/azure-sdk-for-python/pull/33307))\n\n### Other Changes\n\n- Added dependency on `anyio` >=3.0,<5.0\n- Bumped minimum dependency on `requests` to 2.21.0.\n\n## 1.29.5 (2023-10-19)\n\n### Bugs Fixed\n\n- Fixed an issue with `multipart/form-data` in the async transport where `data` was not getting encoded into the request body. #32473\n\n### Other Changes\n\n- Use ssl context from aiohttp by default.\n\n## 1.29.4 (2023-09-07)\n\n### Bugs Fixed\n\n- Fixed the issue that some urls trigger an infinite loop. #31346\n- Fixed issue where IndexError was raised if multipart responses did not match the number of requests. #31471\n- Fixed issue unbound variable exception if dict is invalid in CloudEvent.from_dict. #31835\n- Fixed issue asyncBearerTokenCredentialPolicy is not backward compatible with SansIOHTTPPolicy. #31836\n- Fixed issue mypy complains with new version of azure-core. #31564\n\n## 1.29.3 (2023-08-22)\n\n### Bugs Fixed\n\n- Typing fix: `message` cannot be `None` in `AzureError`. #31564\n\n## 1.29.2 (2023-08-14)\n\n### Bugs Fixed\n\n- Added a default implementation for `AsyncTokenCredential.__aexit__()` #31573\n\n### Other Changes\n\n- Bumped `typing-extensions` version to 4.6.0.\n\n## 1.29.1 (2023-08-09)\n\n### Bugs Fixed\n\n- Not pass `enabled_cae` unless it is explicitly enabled.\n\n## 1.29.0 (2023-08-03)\n\n### Features Added\n\n- A keyword argument `enable_cae` was added to the `get_token` method of the `TokenCredential` protocol. #31012\n- `BearerTokenCredentialPolicy` and `AsyncBearerTokenCredentialPolicy` now accept `enable_cae` keyword arguments in their constructors. This is used in determining if [Continuous Access Evaluation (CAE)](https://learn.microsoft.com/azure/active-directory/conditional-access/concept-continuous-access-evaluation) should be enabled for each `get_token` request. #31012\n\n## 1.28.0 (2023-07-06)\n\n### Features Added\n\n- Added header name parameter to `RequestIdPolicy`. #30772\n- Added `SensitiveHeaderCleanupPolicy` that cleans up sensitive headers if a redirect happens and the new destination is in another domain. #28349\n\n### Other Changes\n\n- Catch aiohttp errors and translate them into azure-core errors.\n\n## 1.27.1 (2023-06-13)\n\n### Bugs Fixed\n\n- Fix url building for some complex query parameters scenarios #30707\n\n## 1.27.0 (2023-06-01)\n\n### Features Added\n\n- Added support to use sync credentials in `AsyncBearerTokenCredentialPolicy`. #30381\n- Added \"prefix\" parameter to AzureKeyCredentialPolicy #29901\n\n### Bugs Fixed\n\n- Improve error message when providing the wrong credential type for AzureKeyCredential #30380\n\n## 1.26.4 (2023-04-06)\n\n### Features Added\n\n- Updated settings to include OpenTelemetry as a tracer provider. #29095\n\n### Other Changes\n\n- Improved typing\n\n## 1.26.3 (2023-02-02)\n\n### Bugs Fixed\n\n- Fixed deflate decompression for aiohttp #28483\n\n## 1.26.2 (2023-01-05)\n\n### Bugs Fixed\n\n- Fix 'ClientSession' object has no attribute 'auto_decompress' (thanks to @mghextreme for the contribution)\n\n### Other Changes\n\n- Add \"x-ms-error-code\" as secure header to log\n- Rename \"DEFAULT_HEADERS_WHITELIST\" to \"DEFAULT_HEADERS_ALLOWLIST\". Added a backward compatible alias.\n\n## 1.26.1 (2022-11-03)\n\n### Other Changes\n\n- Added example of RequestsTransport with custom session. (thanks to @inirudebwoy for the contribution) #26768\n- Added Python 3.11 support.\n\n## 1.26.0 (2022-10-06)\n\n### Other Changes\n\n- LRO polling will not wait anymore before doing the first status check #26376\n- Added extra dependency for [aio]. pip install azure-core[aio] installs aiohttp too.\n\n## 1.25.1 (2022-09-01)\n\n### Bugs Fixed\n\n- Added @runtime_checkable to `TokenCredential` protocol definitions #25187\n\n## 1.25.0 (2022-08-04)\n\nAzure-core is supported on Python 3.7 or later. For more details, please read our page on [Azure SDK for Python version support policy](https://github.com/Azure/azure-sdk-for-python/wiki/Azure-SDKs-Python-version-support-policy).\n\n### Features Added\n\n- Added `CaseInsensitiveDict` implementation in `azure.core.utils` removing dependency on `requests` and `aiohttp`\n\n## 1.24.2 (2022-06-30)\n\n### Bugs Fixed\n\n- Fixed the bug that azure-core could not be imported under Python 3.11.0b3 #24928\n- `ContentDecodePolicy` can now correctly deserialize more JSON bodies with different mime types #22410\n\n## 1.24.1 (2022-06-01)\n\n### Bugs Fixed\n\n- Declare method level span as INTERNAL by default #24492\n- Fixed type hints for `azure.core.paging.ItemPaged` #24548\n\n## 1.24.0 (2022-05-06)\n\n### Features Added\n\n- Add `SerializationError` and `DeserializationError` in `azure.core.exceptions` for errors raised during serialization / deserialization #24312\n\n## 1.23.1 (2022-03-31)\n\n### Bugs Fixed\n\n- Allow stream inputs to the `content` kwarg of `azure.core.rest.HttpRequest` from objects with a `read` method #23578\n\n## 1.23.0 (2022-03-03)\n\n### Features Added\n\n- Improve intellisense type hinting for service client methods. #22891\n\n- Add a case insensitive dict `case_insensitive_dict` in `azure.core.utils`. #23206\n\n### Bugs Fixed\n\n- Use \"\\n\" rather than \"/n\" for new line in log. #23261\n\n### Other Changes\n\n- Log \"WWW-Authenticate\" header in `HttpLoggingPolicy` #22990\n- Added dependency on `typing-extensions` >= 4.0.1\n\n## 1.22.1 (2022-02-09)\n\n### Bugs Fixed\n\n- Limiting `final-state-via` scope to POST until consuming SDKs has been fixed to use this option properly on PUT. #22989\n\n## 1.22.0 (2022-02-03)\n_[**This version is deprecated.**]_\n\n### Features Added\n\n- Add support for `final-state-via` LRO option in core. #22713\n\n### Bugs Fixed\n\n- Add response body to string representation of `HttpResponseError` if we're not able to parse out information #22302\n- Raise `AttributeError` when calling azure.core.pipeline.transport.\\_\\_bases__ #22469\n\n### Other Changes\n\n- Python 2.7 is no longer supported. Please use Python version 3.6 or later.\n\n## 1.21.1 (2021-12-06)\n\n### Other Changes\n\n- Revert change in str method #22023\n\n## 1.21.0 (2021-12-02)\n\n### Breaking Changes\n\n- Sync stream downloading now raises `azure.core.exceptions.DecodeError` rather than `requests.exceptions.ContentDecodingError`\n\n### Bugs Fixed\n\n- Add response body to string representation of `HttpResponseError` if we're not able to parse out information #21800\n\n## 1.20.1 (2021-11-08)\n\n### Bugs Fixed\n\n- Correctly set response's content to decompressed body when users are using aiohttp transport with decompression headers #21620\n\n## 1.20.0 (2021-11-04)\n\n### Features Added\n\n- GA `send_request` onto the `azure.core.PipelineClient` and `azure.core.AsyncPipelineClient`. This method takes in\nrequests and sends them through our pipelines.\n- GA `azure.core.rest`. `azure.core.rest` is our new public simple HTTP library in `azure.core` that users will use to create requests, and consume responses.\n- GA errors `StreamConsumedError`, `StreamClosedError`, and `ResponseNotReadError` to `azure.core.exceptions`. These errors\nare thrown if you mishandle streamed responses from the `azure.core.rest` module\n- add kwargs to the methods for `iter_raw` and `iter_bytes` #21529\n- no longer raise JSON errors if users pass in file descriptors of JSON to the `json` kwarg in `HttpRequest` #21504\n- Added new error type `IncompleteReadError` which is raised if peer closes the connection before we have received the complete message body.\n\n### Breaking Changes\n\n- SansIOHTTPPolicy.on_exception returns None instead of bool.\n\n### Bugs Fixed\n\n- The `Content-Length` header in a http response is strictly checked against the actual number of bytes in the body,\n rather than silently truncating data in case the underlying tcp connection is closed prematurely.\n (thanks to @jochen-ott-by for the contribution) #20412\n- UnboundLocalError when SansIOHTTPPolicy handles an exception #15222\n- Add default content type header of `text/plain` and content length header for users who pass unicode strings to the `content` kwarg of `HttpRequest` in 2.7 #21550\n\n## 1.19.1 (2021-11-01)\n\n### Bugs Fixed\n\n- respect text encoding specified in argument (thanks to @ryohji for the contribution) #20796\n- Fix \"coroutine x.read() was never awaited\" warning from `ContentDecodePolicy` #21318\n- fix type check for `data` input to `azure.core.rest` for python 2.7 users #21341\n- use `charset_normalizer` if `chardet` is not installed to migrate aiohttp 3.8.0 changes.\n\n### Other Changes\n\n- Refactor AzureJSONEncoder (thanks to @Codejune for the contribution) #21028\n\n## 1.19.0 (2021-09-30)\n\n### Breaking Changes in the Provisional `azure.core.rest` package\n\n- `azure.core.rest.HttpResponse` and `azure.core.rest.AsyncHttpResponse` are now abstract base classes. They should not be initialized directly, instead\nyour transport responses should inherit from them and implement them.\n- The properties of the `azure.core.rest` responses are now all read-only\n\n- HttpLoggingPolicy integrates logs into one record #19925\n\n## 1.18.0 (2021-09-02)\n\n### Features Added\n\n- `azure.core.serialization.AzureJSONEncoder` (introduced in 1.17.0) serializes `datetime.datetime` objects in ISO 8601 format, conforming to RFC 3339's specification. #20190\n- We now use `azure.core.serialization.AzureJSONEncoder` to serialize `json` input to `azure.core.rest.HttpRequest`.\n\n### Breaking Changes in the Provisional `azure.core.rest` package\n\n- The `text` property on `azure.core.rest.HttpResponse` and `azure.core.rest.AsyncHttpResponse` has changed to a method, which also takes\nan `encoding` parameter.\n- Removed `iter_text` and `iter_lines` from `azure.core.rest.HttpResponse` and `azure.core.rest.AsyncHttpResponse`\n\n### Bugs Fixed\n\n- The behaviour of the headers returned in `azure.core.rest` responses now aligns across sync and async. Items can now be checked case-insensitively and without raising an error for format.\n\n## 1.17.0 (2021-08-05)\n\n### Features Added\n\n- Cut hard dependency on requests library\n- Added a `from_json` method which now accepts storage QueueMessage, eventhub's EventData or ServiceBusMessage or simply json bytes to return a `CloudEvent`\n\n### Fixed\n\n- Not override \"x-ms-client-request-id\" if it already exists in the header. #17757\n\n### Breaking Changes in the Provisional `azure.core.rest` package\n\n- `azure.core.rest` will not try to guess the `charset` anymore if it was impossible to extract it from `HttpResponse` analysis. This removes our dependency on `charset`.\n\n## 1.16.0 (2021-07-01)\n\n### Features Added\n\n- Add new ***provisional*** methods `send_request` onto the `azure.core.PipelineClient` and `azure.core.AsyncPipelineClient`. This method takes in\nrequests and sends them through our pipelines.\n- Add new ***provisional*** module `azure.core.rest`. `azure.core.rest` is our new public simple HTTP library in `azure.core` that users will use to create requests, and consume responses.\n- Add new ***provisional*** errors `StreamConsumedError`, `StreamClosedError`, and `ResponseNotReadError` to `azure.core.exceptions`. These errors\nare thrown if you mishandle streamed responses from the provisional `azure.core.rest` module\n\n### Fixed\n\n- Improved error message in the `from_dict` method of `CloudEvent` when a wrong schema is sent.\n\n## 1.15.0 (2021-06-04)\n\n### New Features\n\n- Added `BearerTokenCredentialPolicy.on_challenge` and `.authorize_request` to allow subclasses to optionally handle authentication challenges\n\n### Bug Fixes\n\n- Retry policies don't sleep after operations time out\n- The `from_dict` methhod in the `CloudEvent` can now convert a datetime string to datetime object when microsecond exceeds the python limitation\n\n## 1.14.0 (2021-05-13)\n\n### New Features\n\n- Added `azure.core.credentials.AzureNamedKeyCredential` credential #17548.\n- Added `decompress` parameter for `stream_download` method. If it is set to `False`, will not do decompression upon the stream. #17920\n\n## 1.13.0 (2021-04-02)\n\nAzure core requires Python 2.7 or Python 3.6+ since this release.\n\n### New Features\n\n- Added `azure.core.utils.parse_connection_string` function to parse connection strings across SDKs, with common validation and support for case insensitive keys.\n- Supported adding custom policies #16519\n- Added `~azure.core.tracing.Link` that should be used while passing `Links` to `AbstractSpan`.\n- `AbstractSpan` constructor can now take in additional keyword only args.\n\n### Bug fixes\n\n- Make NetworkTraceLoggingPolicy show the auth token in plain text. #14191\n- Fixed RetryPolicy overriding default connection timeout with an extreme value #17481\n\n## 1.12.0 (2021-03-08)\n\nThis version will be the last version to officially support Python 3.5, future versions will require Python 2.7 or Python 3.6+.\n\n### Features\n\n- Added `azure.core.messaging.CloudEvent` model that follows the cloud event spec.\n- Added `azure.core.serialization.NULL` sentinel value\n- Improve `repr`s for `HttpRequest` and `HttpResponse`s #16972\n\n### Bug Fixes\n\n- Disable retry in stream downloading. (thanks to @jochen-ott-by @hoffmann for the contribution) #16723\n\n## 1.11.0 (2021-02-08)\n\n### Features\n\n- Added `CaseInsensitiveEnumMeta` class for case-insensitive enums. #16316\n- Add `raise_for_status` method onto `HttpResponse`. Calling `response.raise_for_status()` on a response with an error code\nwill raise an `HttpResponseError`. Calling it on a good response will do nothing #16399\n\n### Bug Fixes\n\n- Update conn.conn_kw rather than overriding it when setting block size. (thanks for @jiasli for the contribution) #16587\n\n## 1.10.0 (2021-01-11)\n\n### Features\n\n- Added `AzureSasCredential` and its respective policy. #15946\n\n## 1.9.0 (2020-11-09)\n\n### Features\n\n- Add a `continuation_token` attribute to the base `AzureError` exception, and set this value for errors raised\n during paged or long-running operations.\n\n### Bug Fixes\n\n- Set retry_interval to 1 second instead of 1000 seconds (thanks **vbarbaresi** for contributing) #14357\n\n\n## 1.8.2 (2020-10-05)\n\n### Bug Fixes\n\n- Fixed bug to allow polling in the case of parameterized endpoints with relative polling urls #14097\n\n\n## 1.8.1 (2020-09-08)\n\n### Bug fixes\n\n- SAS credential replicated \"/\" fix #13159\n\n## 1.8.0 (2020-08-10)\n\n### Features\n\n- Support params as list for exploding parameters #12410\n\n\n## 1.7.0 (2020-07-06)\n\n### Bug fixes\n\n- `AzureKeyCredentialPolicy` will now accept (and ignore) passed in kwargs #11963\n- Better error messages if passed endpoint is incorrect #12106\n- Do not JSON encore a string if content type is \"text\" #12137\n\n### Features\n\n- Added `http_logging_policy` property on the `Configuration` object, allowing users to individually\nset the http logging policy of the config #12218\n\n## 1.6.0 (2020-06-03)\n\n### Bug fixes\n\n- Fixed deadlocks in AsyncBearerTokenCredentialPolicy #11543\n- Fix AttributeException in StreamDownloadGenerator #11462\n\n### Features\n\n- Added support for changesets as part of multipart message support #10485\n- Add AsyncLROPoller in azure.core.polling #10801\n- Add get_continuation_token/from_continuation_token/polling_method methods in pollers (sync and async) #10801\n- HttpResponse and PipelineContext objects are now pickable #10801\n\n## 1.5.0 (2020-05-04)\n\n### Features\n\n- Support \"x-ms-retry-after-ms\" in response header #10743\n- `link` and `link_from_headers` now accepts attributes #10765\n\n### Bug fixes\n\n- Not retry if the status code is less than 400 #10778\n- \"x-ms-request-id\" is not considered safe header for logging #10967\n\n## 1.4.0 (2020-04-06)\n\n### Features\n\n- Support a default error type in map_error #9773\n- Added `AzureKeyCredential` and its respective policy. #10509\n- Added `azure.core.polling.base_polling` module with a \"Microsoft One API\" polling implementation #10090\n Also contains the async version in `azure.core.polling.async_base_polling`\n- Support kwarg `enforce_https` to disable HTTPS check on authentication #9821\n- Support additional kwargs in `HttpRequest.set_multipart_mixed` that will be passed into pipeline context.\n\n## 1.3.0 (2020-03-09)\n\n### Bug fixes\n\n- Appended RequestIdPolicy to the default pipeline #9841\n- Rewind the body position in async_retry #10117\n\n### Features\n\n- Add raw_request_hook support in custom_hook_policy #9958\n- Add timeout support in retry_policy #10011\n- Add OdataV4 error format auto-parsing in all exceptions ('error' attribute) #9738\n\n## 1.2.2 (2020-02-10)\n\n### Bug fixes\n\n- Fixed a bug that sends None as request_id #9545\n- Enable mypy for customers #9572\n- Handle TypeError in deep copy #9620\n- Fix text/plain content-type in decoder #9589\n\n## 1.2.1 (2020-01-14)\n\n### Bug fixes\n\n- Fixed a regression in 1.2.0 that was incompatible with azure-keyvault-* 4.0.0\n[#9462](https://github.com/Azure/azure-sdk-for-python/issues/9462)\n\n\n## 1.2.0 (2020-01-14)\n\n### Features\n\n- Add user_agent & sdk_moniker kwargs in UserAgentPolicy init #9355\n- Support OPTIONS HTTP verb #9322\n- Add tracing_attributes to tracing decorator #9297\n- Support auto_request_id in RequestIdPolicy #9163\n- Support fixed retry #6419\n- Support \"retry-after-ms\" in response header #9240\n\n### Bug fixes\n\n- Removed `__enter__` and `__exit__` from async context managers #9313\n\n## 1.1.1 (2019-12-03)\n\n### Bug fixes\n\n- Bearer token authorization requires HTTPS\n- Rewind the body position in retry #8307\n\n## 1.1.0 (2019-11-25)\n\n### Features\n\n- New RequestIdPolicy #8437\n- Enable logging policy in default pipeline #8053\n- Normalize transport timeout. #8000\n Now we have:\n * 'connection_timeout' - a single float in seconds for the connection timeout. Default 5min\n * 'read_timeout' - a single float in seconds for the read timeout. Default 5min\n\n### Bug fixes\n\n- RequestHistory: deepcopy fails if request contains a stream #7732\n- Retry: retry raises error if response does not have http_response #8629\n- Client kwargs are now passed to DistributedTracingPolicy correctly #8051\n- NetworkLoggingPolicy now logs correctly all requests in case of retry #8262\n\n## 1.0.0 (2019-10-29)\n\n### Features\n\n- Tracing: DistributedTracingPolicy now accepts kwargs network_span_namer to change network span name #7773\n- Tracing: Implementation of AbstractSpan can now use the mixin HttpSpanMixin to get HTTP span update automatically #7773\n- Tracing: AbstractSpan contract \"change_context\" introduced #7773\n- Introduce new policy HttpLoggingPolicy #7988\n\n### Bug fixes\n\n- Fix AsyncioRequestsTransport if input stream is an async generator #7743\n- Fix form-data with aiohttp transport #7749\n\n### Breaking changes\n\n- Tracing: AbstractSpan.set_current_span is longer supported. Use change_context instead. #7773\n- azure.core.pipeline.policies.ContentDecodePolicy.deserialize_from_text changed\n\n## 1.0.0b4 (2019-10-07)\n\n### Features\n\n- Tracing: network span context is available with the TRACING_CONTEXT in pipeline response #7252\n- Tracing: Span contract now has `kind`, `traceparent` and is a context manager #7252\n- SansIOHTTPPolicy methods can now be coroutines #7497\n- Add multipart/mixed support #7083:\n\n - HttpRequest now has a \"set_multipart_mixed\" method to set the parts of this request\n - HttpRequest now has a \"prepare_multipart_body\" method to build final body.\n - HttpResponse now has a \"parts\" method to return an iterator of parts\n - AsyncHttpResponse now has a \"parts\" methods to return an async iterator of parts\n - Note that multipart/mixed is a Python 3.x only feature\n\n### Bug fixes\n\n- Tracing: policy cannot fail the pipeline, even in the worst condition #7252\n- Tracing: policy pass correctly status message if exception #7252\n- Tracing: incorrect span if exception raised from decorated function #7133\n- Fixed urllib3 ConnectTimeoutError being raised by Requests during a socket timeout. Now this exception is caught and wrapped as a `ServiceRequestError` #7542\n\n### Breaking changes\n\n- Tracing: `azure.core.tracing.context` removed\n- Tracing: `azure.core.tracing.context.tracing_context.with_current_context` renamed to `azure.core.tracing.common.with_current_context` #7252\n- Tracing: `link` renamed `link_from_headers` and `link` takes now a string\n- Tracing: opencensus implementation has been moved to the package `azure-core-tracing-opencensus`\n- Some modules and classes that were importables from several different places have been removed:\n\n - `azure.core.HttpResponseError` is now only `azure.core.exceptions.HttpResponseError`\n - `azure.core.Configuration` is now only `azure.core.configuration.Configuration`\n - `azure.core.HttpRequest` is now only `azure.core.pipeline.transport.HttpRequest`\n - `azure.core.version` module has been removed. Use `azure.core.__version__` to get version number.\n - `azure.core.pipeline_client` has been removed. Import from `azure.core` instead.\n - `azure.core.pipeline_client_async` has been removed. Import from `azure.core` instead.\n - `azure.core.pipeline.base` has been removed. Import from `azure.core.pipeline` instead.\n - `azure.core.pipeline.base_async` has been removed. Import from `azure.core.pipeline` instead.\n - `azure.core.pipeline.policies.base` has been removed. Import from `azure.core.pipeline.policies` instead.\n - `azure.core.pipeline.policies.base_async` has been removed. Import from `azure.core.pipeline.policies` instead.\n - `azure.core.pipeline.policies.authentication` has been removed. Import from `azure.core.pipeline.policies` instead.\n - `azure.core.pipeline.policies.authentication_async` has been removed. Import from `azure.core.pipeline.policies` instead.\n - `azure.core.pipeline.policies.custom_hook` has been removed. Import from `azure.core.pipeline.policies` instead.\n - `azure.core.pipeline.policies.redirect` has been removed. Import from `azure.core.pipeline.policies` instead.\n - `azure.core.pipeline.policies.redirect_async` has been removed. Import from `azure.core.pipeline.policies` instead.\n - `azure.core.pipeline.policies.retry` has been removed. Import from `azure.core.pipeline.policies` instead.\n - `azure.core.pipeline.policies.retry_async` has been removed. Import from `azure.core.pipeline.policies` instead.\n - `azure.core.pipeline.policies.distributed_tracing` has been removed. Import from `azure.core.pipeline.policies` instead.\n - `azure.core.pipeline.policies.universal` has been removed. Import from `azure.core.pipeline.policies` instead.\n - `azure.core.tracing.abstract_span` has been removed. Import from `azure.core.tracing` instead.\n - `azure.core.pipeline.transport.base` has been removed. Import from `azure.core.pipeline.transport` instead.\n - `azure.core.pipeline.transport.base_async` has been removed. Import from `azure.core.pipeline.transport` instead.\n - `azure.core.pipeline.transport.requests_basic` has been removed. Import from `azure.core.pipeline.transport` instead.\n - `azure.core.pipeline.transport.requests_asyncio` has been removed. Import from `azure.core.pipeline.transport` instead.\n - `azure.core.pipeline.transport.requests_trio` has been removed. Import from `azure.core.pipeline.transport` instead.\n - `azure.core.pipeline.transport.aiohttp` has been removed. Import from `azure.core.pipeline.transport` instead.\n - `azure.core.polling.poller` has been removed. Import from `azure.core.polling` instead.\n - `azure.core.polling.async_poller` has been removed. Import from `azure.core.polling` instead.\n\n## 1.0.0b3 (2019-09-09)\n\n### Bug fixes\n\n- Fix aiohttp auto-headers #6992\n- Add tracing to policies module init #6951\n\n## 1.0.0b2 (2019-08-05)\n\n### Breaking changes\n\n- Transport classes don't take `config` parameter anymore (use kwargs instead) #6372\n- `azure.core.paging` has been completely refactored #6420\n- HttpResponse.content_type attribute is now a string (was a list) #6490\n- For `StreamDownloadGenerator` subclasses, `response` is now an `HttpResponse`, and not a transport response like `aiohttp.ClientResponse` or `requests.Response`. The transport response is available in `internal_response` attribute #6490\n\n### Bug fixes\n\n- aiohttp is not required to import async pipelines classes #6496\n- `AsyncioRequestsTransport.sleep` is now a coroutine as expected #6490\n- `RequestsTransport` is not tight to `ProxyPolicy` implementation details anymore #6372\n- `AiohttpTransport` does not raise on unexpected kwargs #6355\n\n### Features\n\n- New paging base classes that support `continuation_token` and `by_page()` #6420\n- Proxy support for `AiohttpTransport` #6372\n\n## 1.0.0b1 (2019-06-26)\n\n- Preview 1 release", + "release_date": "2025-05-01T23:17:29", "parties": [ { "type": "person", @@ -152,15 +152,14 @@ "Programming Language :: Python :: 3.11", "Programming Language :: Python :: 3.12", "Programming Language :: Python :: 3.13", - "Programming Language :: Python :: 3.8", "Programming Language :: Python :: 3.9" ], "homepage_url": "https://github.com/Azure/azure-sdk-for-python/tree/main/sdk/core/azure-core", - "download_url": "https://files.pythonhosted.org/packages/07/b7/76b7e144aa53bd206bf1ce34fa75350472c3f69bf30e5c8c18bc9881035d/azure_core-1.33.0-py3-none-any.whl", - "size": 207071, + "download_url": "https://files.pythonhosted.org/packages/84/9e/5c87b49f65bb16571599bc789857d0ded2f53014d3392bc88a5d1f3ad779/azure_core-1.34.0-py3-none-any.whl", + "size": 207409, "sha1": null, - "md5": "ddd17415a5573ca4c435d7db0ff5631f", - "sha256": "9b5b6d0223a1d38c37500e6971118c1e0f13f54951e6893968b38910bc9cda8f", + "md5": "b03f983d92f026edffb1a5c50ee5a64b", + "sha256": "0615d3b756beccdb6624d1c0ae97284f38b78fb59a2a9839bf927c66fbbdddd6", "sha512": null, "bug_tracking_url": null, "code_view_url": null, @@ -180,9 +179,9 @@ "dependencies": [], "repository_homepage_url": null, "repository_download_url": null, - "api_data_url": "https://pypi.org/pypi/azure-core/1.33.0/json", + "api_data_url": "https://pypi.org/pypi/azure-core/1.34.0/json", "datasource_id": null, - "purl": "pkg:pypi/azure-core@1.33.0" + "purl": "pkg:pypi/azure-core@1.34.0" }, { "type": "pypi", @@ -441,12 +440,12 @@ "type": "pypi", "namespace": null, "name": "charset-normalizer", - "version": "3.4.1", + "version": "3.4.2", "qualifiers": {}, "subpath": null, "primary_language": "Python", - "description": "The Real First Universal Charset Detector. Open, modern and actively maintained alternative to Chardet.\n

Charset Detection, for Everyone \ud83d\udc4b

\n\n

\n The Real First Universal Charset Detector
\n \n \n \n \n \"Download\n \n \n \n \n

\n

\n Featured Packages
\n \n \"Static\n \n \n \"Static\n \n

\n

\n In other language (unofficial port - by the community)
\n \n \"Static\n \n

\n\n> A library that helps you read text from an unknown charset encoding.
Motivated by `chardet`,\n> I'm trying to resolve the issue by taking a new approach.\n> All IANA character set names for which the Python core library provides codecs are supported.\n\n

\n >>>>> \ud83d\udc49 Try Me Online Now, Then Adopt Me \ud83d\udc48 <<<<<\n

\n\nThis project offers you an alternative to **Universal Charset Encoding Detector**, also known as **Chardet**.\n\n| Feature | [Chardet](https://github.com/chardet/chardet) | Charset Normalizer | [cChardet](https://github.com/PyYoshi/cChardet) |\n|--------------------------------------------------|:---------------------------------------------:|:--------------------------------------------------------------------------------------------------:|:-----------------------------------------------:|\n| `Fast` | \u274c | \u2705 | \u2705 |\n| `Universal**` | \u274c | \u2705 | \u274c |\n| `Reliable` **without** distinguishable standards | \u274c | \u2705 | \u2705 |\n| `Reliable` **with** distinguishable standards | \u2705 | \u2705 | \u2705 |\n| `License` | LGPL-2.1
_restrictive_ | MIT | MPL-1.1
_restrictive_ |\n| `Native Python` | \u2705 | \u2705 | \u274c |\n| `Detect spoken language` | \u274c | \u2705 | N/A |\n| `UnicodeDecodeError Safety` | \u274c | \u2705 | \u274c |\n| `Whl Size (min)` | 193.6 kB | 42 kB | ~200 kB |\n| `Supported Encoding` | 33 | \ud83c\udf89 [99](https://charset-normalizer.readthedocs.io/en/latest/user/support.html#supported-encodings) | 40 |\n\n

\n\"Reading\"Cat\n

\n\n*\\*\\* : They are clearly using specific code for a specific encoding even if covering most of used one*
\n\n## \u26a1 Performance\n\nThis package offer better performance than its counterpart Chardet. Here are some numbers.\n\n| Package | Accuracy | Mean per file (ms) | File per sec (est) |\n|-----------------------------------------------|:--------:|:------------------:|:------------------:|\n| [chardet](https://github.com/chardet/chardet) | 86 % | 63 ms | 16 file/sec |\n| charset-normalizer | **98 %** | **10 ms** | 100 file/sec |\n\n| Package | 99th percentile | 95th percentile | 50th percentile |\n|-----------------------------------------------|:---------------:|:---------------:|:---------------:|\n| [chardet](https://github.com/chardet/chardet) | 265 ms | 71 ms | 7 ms |\n| charset-normalizer | 100 ms | 50 ms | 5 ms |\n\n_updated as of december 2024 using CPython 3.12_\n\nChardet's performance on larger file (1MB+) are very poor. Expect huge difference on large payload.\n\n> Stats are generated using 400+ files using default parameters. More details on used files, see GHA workflows.\n> And yes, these results might change at any time. The dataset can be updated to include more files.\n> The actual delays heavily depends on your CPU capabilities. The factors should remain the same.\n> Keep in mind that the stats are generous and that Chardet accuracy vs our is measured using Chardet initial capability\n> (e.g. Supported Encoding) Challenge-them if you want.\n\n## \u2728 Installation\n\nUsing pip:\n\n```sh\npip install charset-normalizer -U\n```\n\n## \ud83d\ude80 Basic Usage\n\n### CLI\nThis package comes with a CLI.\n\n```\nusage: normalizer [-h] [-v] [-a] [-n] [-m] [-r] [-f] [-t THRESHOLD]\n file [file ...]\n\nThe Real First Universal Charset Detector. Discover originating encoding used\non text file. Normalize text to unicode.\n\npositional arguments:\n files File(s) to be analysed\n\noptional arguments:\n -h, --help show this help message and exit\n -v, --verbose Display complementary information about file if any.\n Stdout will contain logs about the detection process.\n -a, --with-alternative\n Output complementary possibilities if any. Top-level\n JSON WILL be a list.\n -n, --normalize Permit to normalize input file. If not set, program\n does not write anything.\n -m, --minimal Only output the charset detected to STDOUT. Disabling\n JSON output.\n -r, --replace Replace file when trying to normalize it instead of\n creating a new one.\n -f, --force Replace file without asking if you are sure, use this\n flag with caution.\n -t THRESHOLD, --threshold THRESHOLD\n Define a custom maximum amount of chaos allowed in\n decoded content. 0. <= chaos <= 1.\n --version Show version information and exit.\n```\n\n```bash\nnormalizer ./data/sample.1.fr.srt\n```\n\nor\n\n```bash\npython -m charset_normalizer ./data/sample.1.fr.srt\n```\n\n\ud83c\udf89 Since version 1.4.0 the CLI produce easily usable stdout result in JSON format.\n\n```json\n{\n \"path\": \"/home/default/projects/charset_normalizer/data/sample.1.fr.srt\",\n \"encoding\": \"cp1252\",\n \"encoding_aliases\": [\n \"1252\",\n \"windows_1252\"\n ],\n \"alternative_encodings\": [\n \"cp1254\",\n \"cp1256\",\n \"cp1258\",\n \"iso8859_14\",\n \"iso8859_15\",\n \"iso8859_16\",\n \"iso8859_3\",\n \"iso8859_9\",\n \"latin_1\",\n \"mbcs\"\n ],\n \"language\": \"French\",\n \"alphabets\": [\n \"Basic Latin\",\n \"Latin-1 Supplement\"\n ],\n \"has_sig_or_bom\": false,\n \"chaos\": 0.149,\n \"coherence\": 97.152,\n \"unicode_path\": null,\n \"is_preferred\": true\n}\n```\n\n### Python\n*Just print out normalized text*\n```python\nfrom charset_normalizer import from_path\n\nresults = from_path('./my_subtitle.srt')\n\nprint(str(results.best()))\n```\n\n*Upgrade your code without effort*\n```python\nfrom charset_normalizer import detect\n```\n\nThe above code will behave the same as **chardet**. We ensure that we offer the best (reasonable) BC result possible.\n\nSee the docs for advanced usage : [readthedocs.io](https://charset-normalizer.readthedocs.io/en/latest/)\n\n## \ud83d\ude07 Why\n\nWhen I started using Chardet, I noticed that it was not suited to my expectations, and I wanted to propose a\nreliable alternative using a completely different method. Also! I never back down on a good challenge!\n\nI **don't care** about the **originating charset** encoding, because **two different tables** can\nproduce **two identical rendered string.**\nWhat I want is to get readable text, the best I can.\n\nIn a way, **I'm brute forcing text decoding.** How cool is that ? \ud83d\ude0e\n\nDon't confuse package **ftfy** with charset-normalizer or chardet. ftfy goal is to repair Unicode string whereas charset-normalizer to convert raw file in unknown encoding to unicode.\n\n## \ud83c\udf70 How\n\n - Discard all charset encoding table that could not fit the binary content.\n - Measure noise, or the mess once opened (by chunks) with a corresponding charset encoding.\n - Extract matches with the lowest mess detected.\n - Additionally, we measure coherence / probe for a language.\n\n**Wait a minute**, what is noise/mess and coherence according to **YOU ?**\n\n*Noise :* I opened hundred of text files, **written by humans**, with the wrong encoding table. **I observed**, then\n**I established** some ground rules about **what is obvious** when **it seems like** a mess (aka. defining noise in rendered text).\n I know that my interpretation of what is noise is probably incomplete, feel free to contribute in order to\n improve or rewrite it.\n\n*Coherence :* For each language there is on earth, we have computed ranked letter appearance occurrences (the best we can). So I thought\nthat intel is worth something here. So I use those records against decoded text to check if I can detect intelligent design.\n\n## \u26a1 Known limitations\n\n - Language detection is unreliable when text contains two or more languages sharing identical letters. (eg. HTML (english tags) + Turkish content (Sharing Latin characters))\n - Every charset detector heavily depends on sufficient content. In common cases, do not bother run detection on very tiny content.\n\n## \u26a0\ufe0f About Python EOLs\n\n**If you are running:**\n\n- Python >=2.7,<3.5: Unsupported\n- Python 3.5: charset-normalizer < 2.1\n- Python 3.6: charset-normalizer < 3.1\n- Python 3.7: charset-normalizer < 4.0\n\nUpgrade your Python interpreter as soon as possible.\n\n## \ud83d\udc64 Contributing\n\nContributions, issues and feature requests are very much welcome.
\nFeel free to check [issues page](https://github.com/ousret/charset_normalizer/issues) if you want to contribute.\n\n## \ud83d\udcdd License\n\nCopyright \u00a9 [Ahmed TAHRI @Ousret](https://github.com/Ousret).
\nThis project is [MIT](https://github.com/Ousret/charset_normalizer/blob/master/LICENSE) licensed.\n\nCharacters frequencies used in this project \u00a9 2012 [Denny Vrande\u010di\u0107](http://simia.net/letters/)\n\n## \ud83d\udcbc For Enterprise\n\nProfessional support for charset-normalizer is available as part of the [Tidelift\nSubscription][1]. Tidelift gives software development teams a single source for\npurchasing and maintaining their software, with professional grade assurances\nfrom the experts who know it best, while seamlessly integrating with existing\ntools.\n\n[1]: https://tidelift.com/subscription/pkg/pypi-charset-normalizer?utm_source=pypi-charset-normalizer&utm_medium=readme\n\n[![OpenSSF Best Practices](https://www.bestpractices.dev/projects/7297/badge)](https://www.bestpractices.dev/projects/7297)\n\n# Changelog\nAll notable changes to charset-normalizer will be documented in this file. This project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html).\nThe format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/).\n\n## [3.4.1](https://github.com/Ousret/charset_normalizer/compare/3.4.0...3.4.1) (2024-12-24)\n\n### Changed\n- Project metadata are now stored using `pyproject.toml` instead of `setup.cfg` using setuptools as the build backend.\n- Enforce annotation delayed loading for a simpler and consistent types in the project.\n- Optional mypyc compilation upgraded to version 1.14 for Python >= 3.8\n\n### Added\n- pre-commit configuration.\n- noxfile.\n\n### Removed\n- `build-requirements.txt` as per using `pyproject.toml` native build configuration.\n- `bin/integration.py` and `bin/serve.py` in favor of downstream integration test (see noxfile).\n- `setup.cfg` in favor of `pyproject.toml` metadata configuration.\n- Unused `utils.range_scan` function.\n\n### Fixed\n- Converting content to Unicode bytes may insert `utf_8` instead of preferred `utf-8`. (#572)\n- Deprecation warning \"'count' is passed as positional argument\" when converting to Unicode bytes on Python 3.13+\n\n## [3.4.0](https://github.com/Ousret/charset_normalizer/compare/3.3.2...3.4.0) (2024-10-08)\n\n### Added\n- Argument `--no-preemptive` in the CLI to prevent the detector to search for hints.\n- Support for Python 3.13 (#512)\n\n### Fixed\n- Relax the TypeError exception thrown when trying to compare a CharsetMatch with anything else than a CharsetMatch.\n- Improved the general reliability of the detector based on user feedbacks. (#520) (#509) (#498) (#407) (#537)\n- Declared charset in content (preemptive detection) not changed when converting to utf-8 bytes. (#381)\n\n## [3.3.2](https://github.com/Ousret/charset_normalizer/compare/3.3.1...3.3.2) (2023-10-31)\n\n### Fixed\n- Unintentional memory usage regression when using large payload that match several encoding (#376)\n- Regression on some detection case showcased in the documentation (#371)\n\n### Added\n- Noise (md) probe that identify malformed arabic representation due to the presence of letters in isolated form (credit to my wife)\n\n## [3.3.1](https://github.com/Ousret/charset_normalizer/compare/3.3.0...3.3.1) (2023-10-22)\n\n### Changed\n- Optional mypyc compilation upgraded to version 1.6.1 for Python >= 3.8\n- Improved the general detection reliability based on reports from the community\n\n## [3.3.0](https://github.com/Ousret/charset_normalizer/compare/3.2.0...3.3.0) (2023-09-30)\n\n### Added\n- Allow to execute the CLI (e.g. normalizer) through `python -m charset_normalizer.cli` or `python -m charset_normalizer`\n- Support for 9 forgotten encoding that are supported by Python but unlisted in `encoding.aliases` as they have no alias (#323)\n\n### Removed\n- (internal) Redundant utils.is_ascii function and unused function is_private_use_only\n- (internal) charset_normalizer.assets is moved inside charset_normalizer.constant\n\n### Changed\n- (internal) Unicode code blocks in constants are updated using the latest v15.0.0 definition to improve detection\n- Optional mypyc compilation upgraded to version 1.5.1 for Python >= 3.8\n\n### Fixed\n- Unable to properly sort CharsetMatch when both chaos/noise and coherence were close due to an unreachable condition in \\_\\_lt\\_\\_ (#350)\n\n## [3.2.0](https://github.com/Ousret/charset_normalizer/compare/3.1.0...3.2.0) (2023-06-07)\n\n### Changed\n- Typehint for function `from_path` no longer enforce `PathLike` as its first argument\n- Minor improvement over the global detection reliability\n\n### Added\n- Introduce function `is_binary` that relies on main capabilities, and optimized to detect binaries\n- Propagate `enable_fallback` argument throughout `from_bytes`, `from_path`, and `from_fp` that allow a deeper control over the detection (default True)\n- Explicit support for Python 3.12\n\n### Fixed\n- Edge case detection failure where a file would contain 'very-long' camel cased word (Issue #289)\n\n## [3.1.0](https://github.com/Ousret/charset_normalizer/compare/3.0.1...3.1.0) (2023-03-06)\n\n### Added\n- Argument `should_rename_legacy` for legacy function `detect` and disregard any new arguments without errors (PR #262)\n\n### Removed\n- Support for Python 3.6 (PR #260)\n\n### Changed\n- Optional speedup provided by mypy/c 1.0.1\n\n## [3.0.1](https://github.com/Ousret/charset_normalizer/compare/3.0.0...3.0.1) (2022-11-18)\n\n### Fixed\n- Multi-bytes cutter/chunk generator did not always cut correctly (PR #233)\n\n### Changed\n- Speedup provided by mypy/c 0.990 on Python >= 3.7\n\n## [3.0.0](https://github.com/Ousret/charset_normalizer/compare/2.1.1...3.0.0) (2022-10-20)\n\n### Added\n- Extend the capability of explain=True when cp_isolation contains at most two entries (min one), will log in details of the Mess-detector results\n- Support for alternative language frequency set in charset_normalizer.assets.FREQUENCIES\n- Add parameter `language_threshold` in `from_bytes`, `from_path` and `from_fp` to adjust the minimum expected coherence ratio\n- `normalizer --version` now specify if current version provide extra speedup (meaning mypyc compilation whl)\n\n### Changed\n- Build with static metadata using 'build' frontend\n- Make the language detection stricter\n- Optional: Module `md.py` can be compiled using Mypyc to provide an extra speedup up to 4x faster than v2.1\n\n### Fixed\n- CLI with opt --normalize fail when using full path for files\n- TooManyAccentuatedPlugin induce false positive on the mess detection when too few alpha character have been fed to it\n- Sphinx warnings when generating the documentation\n\n### Removed\n- Coherence detector no longer return 'Simple English' instead return 'English'\n- Coherence detector no longer return 'Classical Chinese' instead return 'Chinese'\n- Breaking: Method `first()` and `best()` from CharsetMatch\n- UTF-7 will no longer appear as \"detected\" without a recognized SIG/mark (is unreliable/conflict with ASCII)\n- Breaking: Class aliases CharsetDetector, CharsetDoctor, CharsetNormalizerMatch and CharsetNormalizerMatches\n- Breaking: Top-level function `normalize`\n- Breaking: Properties `chaos_secondary_pass`, `coherence_non_latin` and `w_counter` from CharsetMatch\n- Support for the backport `unicodedata2`\n\n## [3.0.0rc1](https://github.com/Ousret/charset_normalizer/compare/3.0.0b2...3.0.0rc1) (2022-10-18)\n\n### Added\n- Extend the capability of explain=True when cp_isolation contains at most two entries (min one), will log in details of the Mess-detector results\n- Support for alternative language frequency set in charset_normalizer.assets.FREQUENCIES\n- Add parameter `language_threshold` in `from_bytes`, `from_path` and `from_fp` to adjust the minimum expected coherence ratio\n\n### Changed\n- Build with static metadata using 'build' frontend\n- Make the language detection stricter\n\n### Fixed\n- CLI with opt --normalize fail when using full path for files\n- TooManyAccentuatedPlugin induce false positive on the mess detection when too few alpha character have been fed to it\n\n### Removed\n- Coherence detector no longer return 'Simple English' instead return 'English'\n- Coherence detector no longer return 'Classical Chinese' instead return 'Chinese'\n\n## [3.0.0b2](https://github.com/Ousret/charset_normalizer/compare/3.0.0b1...3.0.0b2) (2022-08-21)\n\n### Added\n- `normalizer --version` now specify if current version provide extra speedup (meaning mypyc compilation whl)\n\n### Removed\n- Breaking: Method `first()` and `best()` from CharsetMatch\n- UTF-7 will no longer appear as \"detected\" without a recognized SIG/mark (is unreliable/conflict with ASCII)\n\n### Fixed\n- Sphinx warnings when generating the documentation\n\n## [3.0.0b1](https://github.com/Ousret/charset_normalizer/compare/2.1.0...3.0.0b1) (2022-08-15)\n\n### Changed\n- Optional: Module `md.py` can be compiled using Mypyc to provide an extra speedup up to 4x faster than v2.1\n\n### Removed\n- Breaking: Class aliases CharsetDetector, CharsetDoctor, CharsetNormalizerMatch and CharsetNormalizerMatches\n- Breaking: Top-level function `normalize`\n- Breaking: Properties `chaos_secondary_pass`, `coherence_non_latin` and `w_counter` from CharsetMatch\n- Support for the backport `unicodedata2`\n\n## [2.1.1](https://github.com/Ousret/charset_normalizer/compare/2.1.0...2.1.1) (2022-08-19)\n\n### Deprecated\n- Function `normalize` scheduled for removal in 3.0\n\n### Changed\n- Removed useless call to decode in fn is_unprintable (#206)\n\n### Fixed\n- Third-party library (i18n xgettext) crashing not recognizing utf_8 (PEP 263) with underscore from [@aleksandernovikov](https://github.com/aleksandernovikov) (#204)\n\n## [2.1.0](https://github.com/Ousret/charset_normalizer/compare/2.0.12...2.1.0) (2022-06-19)\n\n### Added\n- Output the Unicode table version when running the CLI with `--version` (PR #194)\n\n### Changed\n- Re-use decoded buffer for single byte character sets from [@nijel](https://github.com/nijel) (PR #175)\n- Fixing some performance bottlenecks from [@deedy5](https://github.com/deedy5) (PR #183)\n\n### Fixed\n- Workaround potential bug in cpython with Zero Width No-Break Space located in Arabic Presentation Forms-B, Unicode 1.1 not acknowledged as space (PR #175)\n- CLI default threshold aligned with the API threshold from [@oleksandr-kuzmenko](https://github.com/oleksandr-kuzmenko) (PR #181)\n\n### Removed\n- Support for Python 3.5 (PR #192)\n\n### Deprecated\n- Use of backport unicodedata from `unicodedata2` as Python is quickly catching up, scheduled for removal in 3.0 (PR #194)\n\n## [2.0.12](https://github.com/Ousret/charset_normalizer/compare/2.0.11...2.0.12) (2022-02-12)\n\n### Fixed\n- ASCII miss-detection on rare cases (PR #170)\n\n## [2.0.11](https://github.com/Ousret/charset_normalizer/compare/2.0.10...2.0.11) (2022-01-30)\n\n### Added\n- Explicit support for Python 3.11 (PR #164)\n\n### Changed\n- The logging behavior have been completely reviewed, now using only TRACE and DEBUG levels (PR #163 #165)\n\n## [2.0.10](https://github.com/Ousret/charset_normalizer/compare/2.0.9...2.0.10) (2022-01-04)\n\n### Fixed\n- Fallback match entries might lead to UnicodeDecodeError for large bytes sequence (PR #154)\n\n### Changed\n- Skipping the language-detection (CD) on ASCII (PR #155)\n\n## [2.0.9](https://github.com/Ousret/charset_normalizer/compare/2.0.8...2.0.9) (2021-12-03)\n\n### Changed\n- Moderating the logging impact (since 2.0.8) for specific environments (PR #147)\n\n### Fixed\n- Wrong logging level applied when setting kwarg `explain` to True (PR #146)\n\n## [2.0.8](https://github.com/Ousret/charset_normalizer/compare/2.0.7...2.0.8) (2021-11-24)\n### Changed\n- Improvement over Vietnamese detection (PR #126)\n- MD improvement on trailing data and long foreign (non-pure latin) data (PR #124)\n- Efficiency improvements in cd/alphabet_languages from [@adbar](https://github.com/adbar) (PR #122)\n- call sum() without an intermediary list following PEP 289 recommendations from [@adbar](https://github.com/adbar) (PR #129)\n- Code style as refactored by Sourcery-AI (PR #131)\n- Minor adjustment on the MD around european words (PR #133)\n- Remove and replace SRTs from assets / tests (PR #139)\n- Initialize the library logger with a `NullHandler` by default from [@nmaynes](https://github.com/nmaynes) (PR #135)\n- Setting kwarg `explain` to True will add provisionally (bounded to function lifespan) a specific stream handler (PR #135)\n\n### Fixed\n- Fix large (misleading) sequence giving UnicodeDecodeError (PR #137)\n- Avoid using too insignificant chunk (PR #137)\n\n### Added\n- Add and expose function `set_logging_handler` to configure a specific StreamHandler from [@nmaynes](https://github.com/nmaynes) (PR #135)\n- Add `CHANGELOG.md` entries, format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/) (PR #141)\n\n## [2.0.7](https://github.com/Ousret/charset_normalizer/compare/2.0.6...2.0.7) (2021-10-11)\n### Added\n- Add support for Kazakh (Cyrillic) language detection (PR #109)\n\n### Changed\n- Further, improve inferring the language from a given single-byte code page (PR #112)\n- Vainly trying to leverage PEP263 when PEP3120 is not supported (PR #116)\n- Refactoring for potential performance improvements in loops from [@adbar](https://github.com/adbar) (PR #113)\n- Various detection improvement (MD+CD) (PR #117)\n\n### Removed\n- Remove redundant logging entry about detected language(s) (PR #115)\n\n### Fixed\n- Fix a minor inconsistency between Python 3.5 and other versions regarding language detection (PR #117 #102)\n\n## [2.0.6](https://github.com/Ousret/charset_normalizer/compare/2.0.5...2.0.6) (2021-09-18)\n### Fixed\n- Unforeseen regression with the loss of the backward-compatibility with some older minor of Python 3.5.x (PR #100)\n- Fix CLI crash when using --minimal output in certain cases (PR #103)\n\n### Changed\n- Minor improvement to the detection efficiency (less than 1%) (PR #106 #101)\n\n## [2.0.5](https://github.com/Ousret/charset_normalizer/compare/2.0.4...2.0.5) (2021-09-14)\n### Changed\n- The project now comply with: flake8, mypy, isort and black to ensure a better overall quality (PR #81)\n- The BC-support with v1.x was improved, the old staticmethods are restored (PR #82)\n- The Unicode detection is slightly improved (PR #93)\n- Add syntax sugar \\_\\_bool\\_\\_ for results CharsetMatches list-container (PR #91)\n\n### Removed\n- The project no longer raise warning on tiny content given for detection, will be simply logged as warning instead (PR #92)\n\n### Fixed\n- In some rare case, the chunks extractor could cut in the middle of a multi-byte character and could mislead the mess detection (PR #95)\n- Some rare 'space' characters could trip up the UnprintablePlugin/Mess detection (PR #96)\n- The MANIFEST.in was not exhaustive (PR #78)\n\n## [2.0.4](https://github.com/Ousret/charset_normalizer/compare/2.0.3...2.0.4) (2021-07-30)\n### Fixed\n- The CLI no longer raise an unexpected exception when no encoding has been found (PR #70)\n- Fix accessing the 'alphabets' property when the payload contains surrogate characters (PR #68)\n- The logger could mislead (explain=True) on detected languages and the impact of one MBCS match (PR #72)\n- Submatch factoring could be wrong in rare edge cases (PR #72)\n- Multiple files given to the CLI were ignored when publishing results to STDOUT. (After the first path) (PR #72)\n- Fix line endings from CRLF to LF for certain project files (PR #67)\n\n### Changed\n- Adjust the MD to lower the sensitivity, thus improving the global detection reliability (PR #69 #76)\n- Allow fallback on specified encoding if any (PR #71)\n\n## [2.0.3](https://github.com/Ousret/charset_normalizer/compare/2.0.2...2.0.3) (2021-07-16)\n### Changed\n- Part of the detection mechanism has been improved to be less sensitive, resulting in more accurate detection results. Especially ASCII. (PR #63)\n- According to the community wishes, the detection will fall back on ASCII or UTF-8 in a last-resort case. (PR #64)\n\n## [2.0.2](https://github.com/Ousret/charset_normalizer/compare/2.0.1...2.0.2) (2021-07-15)\n### Fixed\n- Empty/Too small JSON payload miss-detection fixed. Report from [@tseaver](https://github.com/tseaver) (PR #59)\n\n### Changed\n- Don't inject unicodedata2 into sys.modules from [@akx](https://github.com/akx) (PR #57)\n\n## [2.0.1](https://github.com/Ousret/charset_normalizer/compare/2.0.0...2.0.1) (2021-07-13)\n### Fixed\n- Make it work where there isn't a filesystem available, dropping assets frequencies.json. Report from [@sethmlarson](https://github.com/sethmlarson). (PR #55)\n- Using explain=False permanently disable the verbose output in the current runtime (PR #47)\n- One log entry (language target preemptive) was not show in logs when using explain=True (PR #47)\n- Fix undesired exception (ValueError) on getitem of instance CharsetMatches (PR #52)\n\n### Changed\n- Public function normalize default args values were not aligned with from_bytes (PR #53)\n\n### Added\n- You may now use charset aliases in cp_isolation and cp_exclusion arguments (PR #47)\n\n## [2.0.0](https://github.com/Ousret/charset_normalizer/compare/1.4.1...2.0.0) (2021-07-02)\n### Changed\n- 4x to 5 times faster than the previous 1.4.0 release. At least 2x faster than Chardet.\n- Accent has been made on UTF-8 detection, should perform rather instantaneous.\n- The backward compatibility with Chardet has been greatly improved. The legacy detect function returns an identical charset name whenever possible.\n- The detection mechanism has been slightly improved, now Turkish content is detected correctly (most of the time)\n- The program has been rewritten to ease the readability and maintainability. (+Using static typing)+\n- utf_7 detection has been reinstated.\n\n### Removed\n- This package no longer require anything when used with Python 3.5 (Dropped cached_property)\n- Removed support for these languages: Catalan, Esperanto, Kazakh, Baque, Volap\u00fck, Azeri, Galician, Nynorsk, Macedonian, and Serbocroatian.\n- The exception hook on UnicodeDecodeError has been removed.\n\n### Deprecated\n- Methods coherence_non_latin, w_counter, chaos_secondary_pass of the class CharsetMatch are now deprecated and scheduled for removal in v3.0\n\n### Fixed\n- The CLI output used the relative path of the file(s). Should be absolute.\n\n## [1.4.1](https://github.com/Ousret/charset_normalizer/compare/1.4.0...1.4.1) (2021-05-28)\n### Fixed\n- Logger configuration/usage no longer conflict with others (PR #44)\n\n## [1.4.0](https://github.com/Ousret/charset_normalizer/compare/1.3.9...1.4.0) (2021-05-21)\n### Removed\n- Using standard logging instead of using the package loguru.\n- Dropping nose test framework in favor of the maintained pytest.\n- Choose to not use dragonmapper package to help with gibberish Chinese/CJK text.\n- Require cached_property only for Python 3.5 due to constraint. Dropping for every other interpreter version.\n- Stop support for UTF-7 that does not contain a SIG.\n- Dropping PrettyTable, replaced with pure JSON output in CLI.\n\n### Fixed\n- BOM marker in a CharsetNormalizerMatch instance could be False in rare cases even if obviously present. Due to the sub-match factoring process.\n- Not searching properly for the BOM when trying utf32/16 parent codec.\n\n### Changed\n- Improving the package final size by compressing frequencies.json.\n- Huge improvement over the larges payload.\n\n### Added\n- CLI now produces JSON consumable output.\n- Return ASCII if given sequences fit. Given reasonable confidence.\n\n## [1.3.9](https://github.com/Ousret/charset_normalizer/compare/1.3.8...1.3.9) (2021-05-13)\n\n### Fixed\n- In some very rare cases, you may end up getting encode/decode errors due to a bad bytes payload (PR #40)\n\n## [1.3.8](https://github.com/Ousret/charset_normalizer/compare/1.3.7...1.3.8) (2021-05-12)\n\n### Fixed\n- Empty given payload for detection may cause an exception if trying to access the `alphabets` property. (PR #39)\n\n## [1.3.7](https://github.com/Ousret/charset_normalizer/compare/1.3.6...1.3.7) (2021-05-12)\n\n### Fixed\n- The legacy detect function should return UTF-8-SIG if sig is present in the payload. (PR #38)\n\n## [1.3.6](https://github.com/Ousret/charset_normalizer/compare/1.3.5...1.3.6) (2021-02-09)\n\n### Changed\n- Amend the previous release to allow prettytable 2.0 (PR #35)\n\n## [1.3.5](https://github.com/Ousret/charset_normalizer/compare/1.3.4...1.3.5) (2021-02-08)\n\n### Fixed\n- Fix error while using the package with a python pre-release interpreter (PR #33)\n\n### Changed\n- Dependencies refactoring, constraints revised.\n\n### Added\n- Add python 3.9 and 3.10 to the supported interpreters\n\nMIT License\n\nCopyright (c) 2025 TAHRI Ahmed R.\n\nPermission is hereby granted, free of charge, to any person obtaining a copy\nof this software and associated documentation files (the \"Software\"), to deal\nin the Software without restriction, including without limitation the rights\nto use, copy, modify, merge, publish, distribute, sublicense, and/or sell\ncopies of the Software, and to permit persons to whom the Software is\nfurnished to do so, subject to the following conditions:\n\nThe above copyright notice and this permission notice shall be included in all\ncopies or substantial portions of the Software.\n\nTHE SOFTWARE IS PROVIDED \"AS IS\", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR\nIMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,\nFITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE\nAUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER\nLIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,\nOUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE\nSOFTWARE.", - "release_date": "2024-12-24T18:10:50", + "description": "The Real First Universal Charset Detector. Open, modern and actively maintained alternative to Chardet.\n

Charset Detection, for Everyone \ud83d\udc4b

\n\n

\n The Real First Universal Charset Detector
\n \n \n \n \n \"Download\n \n \n \n \n

\n

\n Featured Packages
\n \n \"Static\n \n \n \"Static\n \n

\n

\n In other language (unofficial port - by the community)
\n \n \"Static\n \n

\n\n> A library that helps you read text from an unknown charset encoding.
Motivated by `chardet`,\n> I'm trying to resolve the issue by taking a new approach.\n> All IANA character set names for which the Python core library provides codecs are supported.\n\n

\n >>>>> \ud83d\udc49 Try Me Online Now, Then Adopt Me \ud83d\udc48 <<<<<\n

\n\nThis project offers you an alternative to **Universal Charset Encoding Detector**, also known as **Chardet**.\n\n| Feature | [Chardet](https://github.com/chardet/chardet) | Charset Normalizer | [cChardet](https://github.com/PyYoshi/cChardet) |\n|--------------------------------------------------|:---------------------------------------------:|:--------------------------------------------------------------------------------------------------:|:-----------------------------------------------:|\n| `Fast` | \u274c | \u2705 | \u2705 |\n| `Universal**` | \u274c | \u2705 | \u274c |\n| `Reliable` **without** distinguishable standards | \u274c | \u2705 | \u2705 |\n| `Reliable` **with** distinguishable standards | \u2705 | \u2705 | \u2705 |\n| `License` | LGPL-2.1
_restrictive_ | MIT | MPL-1.1
_restrictive_ |\n| `Native Python` | \u2705 | \u2705 | \u274c |\n| `Detect spoken language` | \u274c | \u2705 | N/A |\n| `UnicodeDecodeError Safety` | \u274c | \u2705 | \u274c |\n| `Whl Size (min)` | 193.6 kB | 42 kB | ~200 kB |\n| `Supported Encoding` | 33 | \ud83c\udf89 [99](https://charset-normalizer.readthedocs.io/en/latest/user/support.html#supported-encodings) | 40 |\n\n

\n\"Reading\"Cat\n

\n\n*\\*\\* : They are clearly using specific code for a specific encoding even if covering most of used one*
\n\n## \u26a1 Performance\n\nThis package offer better performance than its counterpart Chardet. Here are some numbers.\n\n| Package | Accuracy | Mean per file (ms) | File per sec (est) |\n|-----------------------------------------------|:--------:|:------------------:|:------------------:|\n| [chardet](https://github.com/chardet/chardet) | 86 % | 63 ms | 16 file/sec |\n| charset-normalizer | **98 %** | **10 ms** | 100 file/sec |\n\n| Package | 99th percentile | 95th percentile | 50th percentile |\n|-----------------------------------------------|:---------------:|:---------------:|:---------------:|\n| [chardet](https://github.com/chardet/chardet) | 265 ms | 71 ms | 7 ms |\n| charset-normalizer | 100 ms | 50 ms | 5 ms |\n\n_updated as of december 2024 using CPython 3.12_\n\nChardet's performance on larger file (1MB+) are very poor. Expect huge difference on large payload.\n\n> Stats are generated using 400+ files using default parameters. More details on used files, see GHA workflows.\n> And yes, these results might change at any time. The dataset can be updated to include more files.\n> The actual delays heavily depends on your CPU capabilities. The factors should remain the same.\n> Keep in mind that the stats are generous and that Chardet accuracy vs our is measured using Chardet initial capability\n> (e.g. Supported Encoding) Challenge-them if you want.\n\n## \u2728 Installation\n\nUsing pip:\n\n```sh\npip install charset-normalizer -U\n```\n\n## \ud83d\ude80 Basic Usage\n\n### CLI\nThis package comes with a CLI.\n\n```\nusage: normalizer [-h] [-v] [-a] [-n] [-m] [-r] [-f] [-t THRESHOLD]\n file [file ...]\n\nThe Real First Universal Charset Detector. Discover originating encoding used\non text file. Normalize text to unicode.\n\npositional arguments:\n files File(s) to be analysed\n\noptional arguments:\n -h, --help show this help message and exit\n -v, --verbose Display complementary information about file if any.\n Stdout will contain logs about the detection process.\n -a, --with-alternative\n Output complementary possibilities if any. Top-level\n JSON WILL be a list.\n -n, --normalize Permit to normalize input file. If not set, program\n does not write anything.\n -m, --minimal Only output the charset detected to STDOUT. Disabling\n JSON output.\n -r, --replace Replace file when trying to normalize it instead of\n creating a new one.\n -f, --force Replace file without asking if you are sure, use this\n flag with caution.\n -t THRESHOLD, --threshold THRESHOLD\n Define a custom maximum amount of chaos allowed in\n decoded content. 0. <= chaos <= 1.\n --version Show version information and exit.\n```\n\n```bash\nnormalizer ./data/sample.1.fr.srt\n```\n\nor\n\n```bash\npython -m charset_normalizer ./data/sample.1.fr.srt\n```\n\n\ud83c\udf89 Since version 1.4.0 the CLI produce easily usable stdout result in JSON format.\n\n```json\n{\n \"path\": \"/home/default/projects/charset_normalizer/data/sample.1.fr.srt\",\n \"encoding\": \"cp1252\",\n \"encoding_aliases\": [\n \"1252\",\n \"windows_1252\"\n ],\n \"alternative_encodings\": [\n \"cp1254\",\n \"cp1256\",\n \"cp1258\",\n \"iso8859_14\",\n \"iso8859_15\",\n \"iso8859_16\",\n \"iso8859_3\",\n \"iso8859_9\",\n \"latin_1\",\n \"mbcs\"\n ],\n \"language\": \"French\",\n \"alphabets\": [\n \"Basic Latin\",\n \"Latin-1 Supplement\"\n ],\n \"has_sig_or_bom\": false,\n \"chaos\": 0.149,\n \"coherence\": 97.152,\n \"unicode_path\": null,\n \"is_preferred\": true\n}\n```\n\n### Python\n*Just print out normalized text*\n```python\nfrom charset_normalizer import from_path\n\nresults = from_path('./my_subtitle.srt')\n\nprint(str(results.best()))\n```\n\n*Upgrade your code without effort*\n```python\nfrom charset_normalizer import detect\n```\n\nThe above code will behave the same as **chardet**. We ensure that we offer the best (reasonable) BC result possible.\n\nSee the docs for advanced usage : [readthedocs.io](https://charset-normalizer.readthedocs.io/en/latest/)\n\n## \ud83d\ude07 Why\n\nWhen I started using Chardet, I noticed that it was not suited to my expectations, and I wanted to propose a\nreliable alternative using a completely different method. Also! I never back down on a good challenge!\n\nI **don't care** about the **originating charset** encoding, because **two different tables** can\nproduce **two identical rendered string.**\nWhat I want is to get readable text, the best I can.\n\nIn a way, **I'm brute forcing text decoding.** How cool is that ? \ud83d\ude0e\n\nDon't confuse package **ftfy** with charset-normalizer or chardet. ftfy goal is to repair Unicode string whereas charset-normalizer to convert raw file in unknown encoding to unicode.\n\n## \ud83c\udf70 How\n\n - Discard all charset encoding table that could not fit the binary content.\n - Measure noise, or the mess once opened (by chunks) with a corresponding charset encoding.\n - Extract matches with the lowest mess detected.\n - Additionally, we measure coherence / probe for a language.\n\n**Wait a minute**, what is noise/mess and coherence according to **YOU ?**\n\n*Noise :* I opened hundred of text files, **written by humans**, with the wrong encoding table. **I observed**, then\n**I established** some ground rules about **what is obvious** when **it seems like** a mess (aka. defining noise in rendered text).\n I know that my interpretation of what is noise is probably incomplete, feel free to contribute in order to\n improve or rewrite it.\n\n*Coherence :* For each language there is on earth, we have computed ranked letter appearance occurrences (the best we can). So I thought\nthat intel is worth something here. So I use those records against decoded text to check if I can detect intelligent design.\n\n## \u26a1 Known limitations\n\n - Language detection is unreliable when text contains two or more languages sharing identical letters. (eg. HTML (english tags) + Turkish content (Sharing Latin characters))\n - Every charset detector heavily depends on sufficient content. In common cases, do not bother run detection on very tiny content.\n\n## \u26a0\ufe0f About Python EOLs\n\n**If you are running:**\n\n- Python >=2.7,<3.5: Unsupported\n- Python 3.5: charset-normalizer < 2.1\n- Python 3.6: charset-normalizer < 3.1\n- Python 3.7: charset-normalizer < 4.0\n\nUpgrade your Python interpreter as soon as possible.\n\n## \ud83d\udc64 Contributing\n\nContributions, issues and feature requests are very much welcome.
\nFeel free to check [issues page](https://github.com/ousret/charset_normalizer/issues) if you want to contribute.\n\n## \ud83d\udcdd License\n\nCopyright \u00a9 [Ahmed TAHRI @Ousret](https://github.com/Ousret).
\nThis project is [MIT](https://github.com/Ousret/charset_normalizer/blob/master/LICENSE) licensed.\n\nCharacters frequencies used in this project \u00a9 2012 [Denny Vrande\u010di\u0107](http://simia.net/letters/)\n\n## \ud83d\udcbc For Enterprise\n\nProfessional support for charset-normalizer is available as part of the [Tidelift\nSubscription][1]. Tidelift gives software development teams a single source for\npurchasing and maintaining their software, with professional grade assurances\nfrom the experts who know it best, while seamlessly integrating with existing\ntools.\n\n[1]: https://tidelift.com/subscription/pkg/pypi-charset-normalizer?utm_source=pypi-charset-normalizer&utm_medium=readme\n\n[![OpenSSF Best Practices](https://www.bestpractices.dev/projects/7297/badge)](https://www.bestpractices.dev/projects/7297)\n\n# Changelog\nAll notable changes to charset-normalizer will be documented in this file. This project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html).\nThe format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/).\n\n## [3.4.2](https://github.com/Ousret/charset_normalizer/compare/3.4.1...3.4.2) (2025-05-02)\n\n### Fixed\n- Addressed the DeprecationWarning in our CLI regarding `argparse.FileType` by backporting the target class into the package. (#591)\n- Improved the overall reliability of the detector with CJK Ideographs. (#605) (#587)\n\n### Changed\n- Optional mypyc compilation upgraded to version 1.15 for Python >= 3.8\n\n## [3.4.1](https://github.com/Ousret/charset_normalizer/compare/3.4.0...3.4.1) (2024-12-24)\n\n### Changed\n- Project metadata are now stored using `pyproject.toml` instead of `setup.cfg` using setuptools as the build backend.\n- Enforce annotation delayed loading for a simpler and consistent types in the project.\n- Optional mypyc compilation upgraded to version 1.14 for Python >= 3.8\n\n### Added\n- pre-commit configuration.\n- noxfile.\n\n### Removed\n- `build-requirements.txt` as per using `pyproject.toml` native build configuration.\n- `bin/integration.py` and `bin/serve.py` in favor of downstream integration test (see noxfile).\n- `setup.cfg` in favor of `pyproject.toml` metadata configuration.\n- Unused `utils.range_scan` function.\n\n### Fixed\n- Converting content to Unicode bytes may insert `utf_8` instead of preferred `utf-8`. (#572)\n- Deprecation warning \"'count' is passed as positional argument\" when converting to Unicode bytes on Python 3.13+\n\n## [3.4.0](https://github.com/Ousret/charset_normalizer/compare/3.3.2...3.4.0) (2024-10-08)\n\n### Added\n- Argument `--no-preemptive` in the CLI to prevent the detector to search for hints.\n- Support for Python 3.13 (#512)\n\n### Fixed\n- Relax the TypeError exception thrown when trying to compare a CharsetMatch with anything else than a CharsetMatch.\n- Improved the general reliability of the detector based on user feedbacks. (#520) (#509) (#498) (#407) (#537)\n- Declared charset in content (preemptive detection) not changed when converting to utf-8 bytes. (#381)\n\n## [3.3.2](https://github.com/Ousret/charset_normalizer/compare/3.3.1...3.3.2) (2023-10-31)\n\n### Fixed\n- Unintentional memory usage regression when using large payload that match several encoding (#376)\n- Regression on some detection case showcased in the documentation (#371)\n\n### Added\n- Noise (md) probe that identify malformed arabic representation due to the presence of letters in isolated form (credit to my wife)\n\n## [3.3.1](https://github.com/Ousret/charset_normalizer/compare/3.3.0...3.3.1) (2023-10-22)\n\n### Changed\n- Optional mypyc compilation upgraded to version 1.6.1 for Python >= 3.8\n- Improved the general detection reliability based on reports from the community\n\n## [3.3.0](https://github.com/Ousret/charset_normalizer/compare/3.2.0...3.3.0) (2023-09-30)\n\n### Added\n- Allow to execute the CLI (e.g. normalizer) through `python -m charset_normalizer.cli` or `python -m charset_normalizer`\n- Support for 9 forgotten encoding that are supported by Python but unlisted in `encoding.aliases` as they have no alias (#323)\n\n### Removed\n- (internal) Redundant utils.is_ascii function and unused function is_private_use_only\n- (internal) charset_normalizer.assets is moved inside charset_normalizer.constant\n\n### Changed\n- (internal) Unicode code blocks in constants are updated using the latest v15.0.0 definition to improve detection\n- Optional mypyc compilation upgraded to version 1.5.1 for Python >= 3.8\n\n### Fixed\n- Unable to properly sort CharsetMatch when both chaos/noise and coherence were close due to an unreachable condition in \\_\\_lt\\_\\_ (#350)\n\n## [3.2.0](https://github.com/Ousret/charset_normalizer/compare/3.1.0...3.2.0) (2023-06-07)\n\n### Changed\n- Typehint for function `from_path` no longer enforce `PathLike` as its first argument\n- Minor improvement over the global detection reliability\n\n### Added\n- Introduce function `is_binary` that relies on main capabilities, and optimized to detect binaries\n- Propagate `enable_fallback` argument throughout `from_bytes`, `from_path`, and `from_fp` that allow a deeper control over the detection (default True)\n- Explicit support for Python 3.12\n\n### Fixed\n- Edge case detection failure where a file would contain 'very-long' camel cased word (Issue #289)\n\n## [3.1.0](https://github.com/Ousret/charset_normalizer/compare/3.0.1...3.1.0) (2023-03-06)\n\n### Added\n- Argument `should_rename_legacy` for legacy function `detect` and disregard any new arguments without errors (PR #262)\n\n### Removed\n- Support for Python 3.6 (PR #260)\n\n### Changed\n- Optional speedup provided by mypy/c 1.0.1\n\n## [3.0.1](https://github.com/Ousret/charset_normalizer/compare/3.0.0...3.0.1) (2022-11-18)\n\n### Fixed\n- Multi-bytes cutter/chunk generator did not always cut correctly (PR #233)\n\n### Changed\n- Speedup provided by mypy/c 0.990 on Python >= 3.7\n\n## [3.0.0](https://github.com/Ousret/charset_normalizer/compare/2.1.1...3.0.0) (2022-10-20)\n\n### Added\n- Extend the capability of explain=True when cp_isolation contains at most two entries (min one), will log in details of the Mess-detector results\n- Support for alternative language frequency set in charset_normalizer.assets.FREQUENCIES\n- Add parameter `language_threshold` in `from_bytes`, `from_path` and `from_fp` to adjust the minimum expected coherence ratio\n- `normalizer --version` now specify if current version provide extra speedup (meaning mypyc compilation whl)\n\n### Changed\n- Build with static metadata using 'build' frontend\n- Make the language detection stricter\n- Optional: Module `md.py` can be compiled using Mypyc to provide an extra speedup up to 4x faster than v2.1\n\n### Fixed\n- CLI with opt --normalize fail when using full path for files\n- TooManyAccentuatedPlugin induce false positive on the mess detection when too few alpha character have been fed to it\n- Sphinx warnings when generating the documentation\n\n### Removed\n- Coherence detector no longer return 'Simple English' instead return 'English'\n- Coherence detector no longer return 'Classical Chinese' instead return 'Chinese'\n- Breaking: Method `first()` and `best()` from CharsetMatch\n- UTF-7 will no longer appear as \"detected\" without a recognized SIG/mark (is unreliable/conflict with ASCII)\n- Breaking: Class aliases CharsetDetector, CharsetDoctor, CharsetNormalizerMatch and CharsetNormalizerMatches\n- Breaking: Top-level function `normalize`\n- Breaking: Properties `chaos_secondary_pass`, `coherence_non_latin` and `w_counter` from CharsetMatch\n- Support for the backport `unicodedata2`\n\n## [3.0.0rc1](https://github.com/Ousret/charset_normalizer/compare/3.0.0b2...3.0.0rc1) (2022-10-18)\n\n### Added\n- Extend the capability of explain=True when cp_isolation contains at most two entries (min one), will log in details of the Mess-detector results\n- Support for alternative language frequency set in charset_normalizer.assets.FREQUENCIES\n- Add parameter `language_threshold` in `from_bytes`, `from_path` and `from_fp` to adjust the minimum expected coherence ratio\n\n### Changed\n- Build with static metadata using 'build' frontend\n- Make the language detection stricter\n\n### Fixed\n- CLI with opt --normalize fail when using full path for files\n- TooManyAccentuatedPlugin induce false positive on the mess detection when too few alpha character have been fed to it\n\n### Removed\n- Coherence detector no longer return 'Simple English' instead return 'English'\n- Coherence detector no longer return 'Classical Chinese' instead return 'Chinese'\n\n## [3.0.0b2](https://github.com/Ousret/charset_normalizer/compare/3.0.0b1...3.0.0b2) (2022-08-21)\n\n### Added\n- `normalizer --version` now specify if current version provide extra speedup (meaning mypyc compilation whl)\n\n### Removed\n- Breaking: Method `first()` and `best()` from CharsetMatch\n- UTF-7 will no longer appear as \"detected\" without a recognized SIG/mark (is unreliable/conflict with ASCII)\n\n### Fixed\n- Sphinx warnings when generating the documentation\n\n## [3.0.0b1](https://github.com/Ousret/charset_normalizer/compare/2.1.0...3.0.0b1) (2022-08-15)\n\n### Changed\n- Optional: Module `md.py` can be compiled using Mypyc to provide an extra speedup up to 4x faster than v2.1\n\n### Removed\n- Breaking: Class aliases CharsetDetector, CharsetDoctor, CharsetNormalizerMatch and CharsetNormalizerMatches\n- Breaking: Top-level function `normalize`\n- Breaking: Properties `chaos_secondary_pass`, `coherence_non_latin` and `w_counter` from CharsetMatch\n- Support for the backport `unicodedata2`\n\n## [2.1.1](https://github.com/Ousret/charset_normalizer/compare/2.1.0...2.1.1) (2022-08-19)\n\n### Deprecated\n- Function `normalize` scheduled for removal in 3.0\n\n### Changed\n- Removed useless call to decode in fn is_unprintable (#206)\n\n### Fixed\n- Third-party library (i18n xgettext) crashing not recognizing utf_8 (PEP 263) with underscore from [@aleksandernovikov](https://github.com/aleksandernovikov) (#204)\n\n## [2.1.0](https://github.com/Ousret/charset_normalizer/compare/2.0.12...2.1.0) (2022-06-19)\n\n### Added\n- Output the Unicode table version when running the CLI with `--version` (PR #194)\n\n### Changed\n- Re-use decoded buffer for single byte character sets from [@nijel](https://github.com/nijel) (PR #175)\n- Fixing some performance bottlenecks from [@deedy5](https://github.com/deedy5) (PR #183)\n\n### Fixed\n- Workaround potential bug in cpython with Zero Width No-Break Space located in Arabic Presentation Forms-B, Unicode 1.1 not acknowledged as space (PR #175)\n- CLI default threshold aligned with the API threshold from [@oleksandr-kuzmenko](https://github.com/oleksandr-kuzmenko) (PR #181)\n\n### Removed\n- Support for Python 3.5 (PR #192)\n\n### Deprecated\n- Use of backport unicodedata from `unicodedata2` as Python is quickly catching up, scheduled for removal in 3.0 (PR #194)\n\n## [2.0.12](https://github.com/Ousret/charset_normalizer/compare/2.0.11...2.0.12) (2022-02-12)\n\n### Fixed\n- ASCII miss-detection on rare cases (PR #170)\n\n## [2.0.11](https://github.com/Ousret/charset_normalizer/compare/2.0.10...2.0.11) (2022-01-30)\n\n### Added\n- Explicit support for Python 3.11 (PR #164)\n\n### Changed\n- The logging behavior have been completely reviewed, now using only TRACE and DEBUG levels (PR #163 #165)\n\n## [2.0.10](https://github.com/Ousret/charset_normalizer/compare/2.0.9...2.0.10) (2022-01-04)\n\n### Fixed\n- Fallback match entries might lead to UnicodeDecodeError for large bytes sequence (PR #154)\n\n### Changed\n- Skipping the language-detection (CD) on ASCII (PR #155)\n\n## [2.0.9](https://github.com/Ousret/charset_normalizer/compare/2.0.8...2.0.9) (2021-12-03)\n\n### Changed\n- Moderating the logging impact (since 2.0.8) for specific environments (PR #147)\n\n### Fixed\n- Wrong logging level applied when setting kwarg `explain` to True (PR #146)\n\n## [2.0.8](https://github.com/Ousret/charset_normalizer/compare/2.0.7...2.0.8) (2021-11-24)\n### Changed\n- Improvement over Vietnamese detection (PR #126)\n- MD improvement on trailing data and long foreign (non-pure latin) data (PR #124)\n- Efficiency improvements in cd/alphabet_languages from [@adbar](https://github.com/adbar) (PR #122)\n- call sum() without an intermediary list following PEP 289 recommendations from [@adbar](https://github.com/adbar) (PR #129)\n- Code style as refactored by Sourcery-AI (PR #131)\n- Minor adjustment on the MD around european words (PR #133)\n- Remove and replace SRTs from assets / tests (PR #139)\n- Initialize the library logger with a `NullHandler` by default from [@nmaynes](https://github.com/nmaynes) (PR #135)\n- Setting kwarg `explain` to True will add provisionally (bounded to function lifespan) a specific stream handler (PR #135)\n\n### Fixed\n- Fix large (misleading) sequence giving UnicodeDecodeError (PR #137)\n- Avoid using too insignificant chunk (PR #137)\n\n### Added\n- Add and expose function `set_logging_handler` to configure a specific StreamHandler from [@nmaynes](https://github.com/nmaynes) (PR #135)\n- Add `CHANGELOG.md` entries, format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/) (PR #141)\n\n## [2.0.7](https://github.com/Ousret/charset_normalizer/compare/2.0.6...2.0.7) (2021-10-11)\n### Added\n- Add support for Kazakh (Cyrillic) language detection (PR #109)\n\n### Changed\n- Further, improve inferring the language from a given single-byte code page (PR #112)\n- Vainly trying to leverage PEP263 when PEP3120 is not supported (PR #116)\n- Refactoring for potential performance improvements in loops from [@adbar](https://github.com/adbar) (PR #113)\n- Various detection improvement (MD+CD) (PR #117)\n\n### Removed\n- Remove redundant logging entry about detected language(s) (PR #115)\n\n### Fixed\n- Fix a minor inconsistency between Python 3.5 and other versions regarding language detection (PR #117 #102)\n\n## [2.0.6](https://github.com/Ousret/charset_normalizer/compare/2.0.5...2.0.6) (2021-09-18)\n### Fixed\n- Unforeseen regression with the loss of the backward-compatibility with some older minor of Python 3.5.x (PR #100)\n- Fix CLI crash when using --minimal output in certain cases (PR #103)\n\n### Changed\n- Minor improvement to the detection efficiency (less than 1%) (PR #106 #101)\n\n## [2.0.5](https://github.com/Ousret/charset_normalizer/compare/2.0.4...2.0.5) (2021-09-14)\n### Changed\n- The project now comply with: flake8, mypy, isort and black to ensure a better overall quality (PR #81)\n- The BC-support with v1.x was improved, the old staticmethods are restored (PR #82)\n- The Unicode detection is slightly improved (PR #93)\n- Add syntax sugar \\_\\_bool\\_\\_ for results CharsetMatches list-container (PR #91)\n\n### Removed\n- The project no longer raise warning on tiny content given for detection, will be simply logged as warning instead (PR #92)\n\n### Fixed\n- In some rare case, the chunks extractor could cut in the middle of a multi-byte character and could mislead the mess detection (PR #95)\n- Some rare 'space' characters could trip up the UnprintablePlugin/Mess detection (PR #96)\n- The MANIFEST.in was not exhaustive (PR #78)\n\n## [2.0.4](https://github.com/Ousret/charset_normalizer/compare/2.0.3...2.0.4) (2021-07-30)\n### Fixed\n- The CLI no longer raise an unexpected exception when no encoding has been found (PR #70)\n- Fix accessing the 'alphabets' property when the payload contains surrogate characters (PR #68)\n- The logger could mislead (explain=True) on detected languages and the impact of one MBCS match (PR #72)\n- Submatch factoring could be wrong in rare edge cases (PR #72)\n- Multiple files given to the CLI were ignored when publishing results to STDOUT. (After the first path) (PR #72)\n- Fix line endings from CRLF to LF for certain project files (PR #67)\n\n### Changed\n- Adjust the MD to lower the sensitivity, thus improving the global detection reliability (PR #69 #76)\n- Allow fallback on specified encoding if any (PR #71)\n\n## [2.0.3](https://github.com/Ousret/charset_normalizer/compare/2.0.2...2.0.3) (2021-07-16)\n### Changed\n- Part of the detection mechanism has been improved to be less sensitive, resulting in more accurate detection results. Especially ASCII. (PR #63)\n- According to the community wishes, the detection will fall back on ASCII or UTF-8 in a last-resort case. (PR #64)\n\n## [2.0.2](https://github.com/Ousret/charset_normalizer/compare/2.0.1...2.0.2) (2021-07-15)\n### Fixed\n- Empty/Too small JSON payload miss-detection fixed. Report from [@tseaver](https://github.com/tseaver) (PR #59)\n\n### Changed\n- Don't inject unicodedata2 into sys.modules from [@akx](https://github.com/akx) (PR #57)\n\n## [2.0.1](https://github.com/Ousret/charset_normalizer/compare/2.0.0...2.0.1) (2021-07-13)\n### Fixed\n- Make it work where there isn't a filesystem available, dropping assets frequencies.json. Report from [@sethmlarson](https://github.com/sethmlarson). (PR #55)\n- Using explain=False permanently disable the verbose output in the current runtime (PR #47)\n- One log entry (language target preemptive) was not show in logs when using explain=True (PR #47)\n- Fix undesired exception (ValueError) on getitem of instance CharsetMatches (PR #52)\n\n### Changed\n- Public function normalize default args values were not aligned with from_bytes (PR #53)\n\n### Added\n- You may now use charset aliases in cp_isolation and cp_exclusion arguments (PR #47)\n\n## [2.0.0](https://github.com/Ousret/charset_normalizer/compare/1.4.1...2.0.0) (2021-07-02)\n### Changed\n- 4x to 5 times faster than the previous 1.4.0 release. At least 2x faster than Chardet.\n- Accent has been made on UTF-8 detection, should perform rather instantaneous.\n- The backward compatibility with Chardet has been greatly improved. The legacy detect function returns an identical charset name whenever possible.\n- The detection mechanism has been slightly improved, now Turkish content is detected correctly (most of the time)\n- The program has been rewritten to ease the readability and maintainability. (+Using static typing)+\n- utf_7 detection has been reinstated.\n\n### Removed\n- This package no longer require anything when used with Python 3.5 (Dropped cached_property)\n- Removed support for these languages: Catalan, Esperanto, Kazakh, Baque, Volap\u00fck, Azeri, Galician, Nynorsk, Macedonian, and Serbocroatian.\n- The exception hook on UnicodeDecodeError has been removed.\n\n### Deprecated\n- Methods coherence_non_latin, w_counter, chaos_secondary_pass of the class CharsetMatch are now deprecated and scheduled for removal in v3.0\n\n### Fixed\n- The CLI output used the relative path of the file(s). Should be absolute.\n\n## [1.4.1](https://github.com/Ousret/charset_normalizer/compare/1.4.0...1.4.1) (2021-05-28)\n### Fixed\n- Logger configuration/usage no longer conflict with others (PR #44)\n\n## [1.4.0](https://github.com/Ousret/charset_normalizer/compare/1.3.9...1.4.0) (2021-05-21)\n### Removed\n- Using standard logging instead of using the package loguru.\n- Dropping nose test framework in favor of the maintained pytest.\n- Choose to not use dragonmapper package to help with gibberish Chinese/CJK text.\n- Require cached_property only for Python 3.5 due to constraint. Dropping for every other interpreter version.\n- Stop support for UTF-7 that does not contain a SIG.\n- Dropping PrettyTable, replaced with pure JSON output in CLI.\n\n### Fixed\n- BOM marker in a CharsetNormalizerMatch instance could be False in rare cases even if obviously present. Due to the sub-match factoring process.\n- Not searching properly for the BOM when trying utf32/16 parent codec.\n\n### Changed\n- Improving the package final size by compressing frequencies.json.\n- Huge improvement over the larges payload.\n\n### Added\n- CLI now produces JSON consumable output.\n- Return ASCII if given sequences fit. Given reasonable confidence.\n\n## [1.3.9](https://github.com/Ousret/charset_normalizer/compare/1.3.8...1.3.9) (2021-05-13)\n\n### Fixed\n- In some very rare cases, you may end up getting encode/decode errors due to a bad bytes payload (PR #40)\n\n## [1.3.8](https://github.com/Ousret/charset_normalizer/compare/1.3.7...1.3.8) (2021-05-12)\n\n### Fixed\n- Empty given payload for detection may cause an exception if trying to access the `alphabets` property. (PR #39)\n\n## [1.3.7](https://github.com/Ousret/charset_normalizer/compare/1.3.6...1.3.7) (2021-05-12)\n\n### Fixed\n- The legacy detect function should return UTF-8-SIG if sig is present in the payload. (PR #38)\n\n## [1.3.6](https://github.com/Ousret/charset_normalizer/compare/1.3.5...1.3.6) (2021-02-09)\n\n### Changed\n- Amend the previous release to allow prettytable 2.0 (PR #35)\n\n## [1.3.5](https://github.com/Ousret/charset_normalizer/compare/1.3.4...1.3.5) (2021-02-08)\n\n### Fixed\n- Fix error while using the package with a python pre-release interpreter (PR #33)\n\n### Changed\n- Dependencies refactoring, constraints revised.\n\n### Added\n- Add python 3.9 and 3.10 to the supported interpreters\n\nMIT License\n\nCopyright (c) 2025 TAHRI Ahmed R.\n\nPermission is hereby granted, free of charge, to any person obtaining a copy\nof this software and associated documentation files (the \"Software\"), to deal\nin the Software without restriction, including without limitation the rights\nto use, copy, modify, merge, publish, distribute, sublicense, and/or sell\ncopies of the Software, and to permit persons to whom the Software is\nfurnished to do so, subject to the following conditions:\n\nThe above copyright notice and this permission notice shall be included in all\ncopies or substantial portions of the Software.\n\nTHE SOFTWARE IS PROVIDED \"AS IS\", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR\nIMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,\nFITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE\nAUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER\nLIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,\nOUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE\nSOFTWARE.", + "release_date": "2025-05-02T08:32:40", "parties": [ { "type": "person", @@ -492,11 +491,11 @@ "Typing :: Typed" ], "homepage_url": null, - "download_url": "https://files.pythonhosted.org/packages/3e/a2/513f6cbe752421f16d969e32f3583762bfd583848b763913ddab8d9bfd4f/charset_normalizer-3.4.1-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", - "size": 145268, + "download_url": "https://files.pythonhosted.org/packages/8c/73/6ede2ec59bce19b3edf4209d70004253ec5f4e319f9a2e3f2f15601ed5f7/charset_normalizer-3.4.2-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", + "size": 148567, "sha1": null, - "md5": "1edb315f82fa657b8ee5d564117e057c", - "sha256": "bc2722592d8998c870fa4e290c2eec2c1569b87fe58618e67d38b4665dfa680d", + "md5": "fec698adfc210708df05f94ca9e09cc3", + "sha256": "4e594135de17ab3866138f496755f302b72157d115086d100c3f19370839dd3a", "sha512": null, "bug_tracking_url": null, "code_view_url": "https://github.com/jawah/charset_normalizer", @@ -516,9 +515,9 @@ "dependencies": [], "repository_homepage_url": null, "repository_download_url": null, - "api_data_url": "https://pypi.org/pypi/charset-normalizer/3.4.1/json", + "api_data_url": "https://pypi.org/pypi/charset-normalizer/3.4.2/json", "datasource_id": null, - "purl": "pkg:pypi/charset-normalizer@3.4.1" + "purl": "pkg:pypi/charset-normalizer@3.4.2" }, { "type": "pypi", @@ -1318,7 +1317,7 @@ ], "resolved_dependencies_graph": [ { - "package": "pkg:pypi/azure-core@1.33.0", + "package": "pkg:pypi/azure-core@1.34.0", "dependencies": [ "pkg:pypi/requests@2.32.3", "pkg:pypi/six@1.17.0", @@ -1334,7 +1333,7 @@ { "package": "pkg:pypi/azure-storage-blob@12.25.1", "dependencies": [ - "pkg:pypi/azure-core@1.33.0", + "pkg:pypi/azure-core@1.34.0", "pkg:pypi/cryptography@44.0.2", "pkg:pypi/isodate@0.7.2", "pkg:pypi/typing-extensions@4.13.2" @@ -1351,7 +1350,7 @@ ] }, { - "package": "pkg:pypi/charset-normalizer@3.4.1", + "package": "pkg:pypi/charset-normalizer@3.4.2", "dependencies": [] }, { @@ -1375,7 +1374,7 @@ { "package": "pkg:pypi/msrest@0.7.1", "dependencies": [ - "pkg:pypi/azure-core@1.33.0", + "pkg:pypi/azure-core@1.34.0", "pkg:pypi/certifi@2025.4.26", "pkg:pypi/isodate@0.7.2", "pkg:pypi/requests-oauthlib@2.0.0", @@ -1401,7 +1400,7 @@ "package": "pkg:pypi/requests@2.32.3", "dependencies": [ "pkg:pypi/certifi@2025.4.26", - "pkg:pypi/charset-normalizer@3.4.1", + "pkg:pypi/charset-normalizer@3.4.2", "pkg:pypi/idna@3.10", "pkg:pypi/urllib3@2.4.0" ] diff --git a/tests/data/azure-devops.req-313-expected.json b/tests/data/azure-devops.req-313-expected.json index 3c7e9a8..a03a1e0 100644 --- a/tests/data/azure-devops.req-313-expected.json +++ b/tests/data/azure-devops.req-313-expected.json @@ -126,12 +126,12 @@ "type": "pypi", "namespace": null, "name": "azure-core", - "version": "1.33.0", + "version": "1.34.0", "qualifiers": {}, "subpath": null, "primary_language": "Python", - "description": "Microsoft Azure Core Library for Python\n# Azure Core shared client library for Python\n\nAzure core provides shared exceptions and modules for Python SDK client libraries.\nThese libraries follow the [Azure SDK Design Guidelines for Python](https://azure.github.io/azure-sdk/python/guidelines/index.html) .\n\nIf you are a client library developer, please reference [client library developer reference](https://github.com/Azure/azure-sdk-for-python/blob/main/sdk/core/azure-core/CLIENT_LIBRARY_DEVELOPER.md) for more information.\n\n[Source code](https://github.com/Azure/azure-sdk-for-python/blob/main/sdk/core/azure-core/)\n| [Package (Pypi)][package]\n| [Package (Conda)](https://anaconda.org/microsoft/azure-core/)\n| [API reference documentation](https://learn.microsoft.com/python/api/overview/azure/core-readme)\n\n## Getting started\n\nTypically, you will not need to install azure core;\nit will be installed when you install one of the client libraries using it.\nIn case you want to install it explicitly (to implement your own client library, for example),\nyou can find it [here](https://pypi.org/project/azure-core/).\n\n## Key concepts\n\n### Azure Core Library Exceptions\n\n#### AzureError\n\nAzureError is the base exception for all errors.\n\n```python\nclass AzureError(Exception):\n def __init__(self, message, *args, **kwargs):\n self.inner_exception = kwargs.get(\"error\")\n self.exc_type, self.exc_value, self.exc_traceback = sys.exc_info()\n self.exc_type = self.exc_type.__name__ if self.exc_type else type(self.inner_exception)\n self.exc_msg = \"{}, {}: {}\".format(message, self.exc_type, self.exc_value) # type: ignore\n self.message = str(message)\n self.continuation_token = kwargs.get(\"continuation_token\")\n super(AzureError, self).__init__(self.message, *args)\n```\n\n*message* is any message (str) to be associated with the exception.\n\n*args* are any additional args to be included with exception.\n\n*kwargs* are keyword arguments to include with the exception. Use the keyword *error* to pass in an internal exception and *continuation_token* for a token reference to continue an incomplete operation.\n\n**The following exceptions inherit from AzureError:**\n\n#### ServiceRequestError\n\nAn error occurred while attempt to make a request to the service. No request was sent.\n\n#### ServiceResponseError\n\nThe request was sent, but the client failed to understand the response.\nThe connection may have timed out. These errors can be retried for idempotent or safe operations.\n\n#### HttpResponseError\n\nA request was made, and a non-success status code was received from the service.\n\n```python\nclass HttpResponseError(AzureError):\n def __init__(self, message=None, response=None, **kwargs):\n self.reason = None\n self.response = response\n if response:\n self.reason = response.reason\n self.status_code = response.status_code\n self.error = self._parse_odata_body(ODataV4Format, response) # type: Optional[ODataV4Format]\n if self.error:\n message = str(self.error)\n else:\n message = message or \"Operation returned an invalid status '{}'\".format(\n self.reason\n )\n\n super(HttpResponseError, self).__init__(message=message, **kwargs)\n```\n\n*message* is the HTTP response error message (optional)\n\n*response* is the HTTP response (optional).\n\n*kwargs* are keyword arguments to include with the exception.\n\n**The following exceptions inherit from HttpResponseError:**\n\n#### DecodeError\n\nAn error raised during response de-serialization.\n\n#### IncompleteReadError\n\nAn error raised if peer closes the connection before we have received the complete message body.\n\n#### ResourceExistsError\n\nAn error response with status code 4xx. This will not be raised directly by the Azure core pipeline.\n\n#### ResourceNotFoundError\n\nAn error response, typically triggered by a 412 response (for update) or 404 (for get/post).\n\n#### ResourceModifiedError\n\nAn error response with status code 4xx, typically 412 Conflict. This will not be raised directly by the Azure core pipeline.\n\n#### ResourceNotModifiedError\n\nAn error response with status code 304. This will not be raised directly by the Azure core pipeline.\n\n#### ClientAuthenticationError\n\nAn error response with status code 4xx. This will not be raised directly by the Azure core pipeline.\n\n#### TooManyRedirectsError\n\nAn error raised when the maximum number of redirect attempts is reached. The maximum amount of redirects can be configured in the RedirectPolicy.\n\n```python\nclass TooManyRedirectsError(HttpResponseError):\n def __init__(self, history, *args, **kwargs):\n self.history = history\n message = \"Reached maximum redirect attempts.\"\n super(TooManyRedirectsError, self).__init__(message, *args, **kwargs)\n```\n\n*history* is used to document the requests/responses that resulted in redirected requests.\n\n*args* are any additional args to be included with exception.\n\n*kwargs* are keyword arguments to include with the exception.\n\n#### StreamConsumedError\n\nAn error thrown if you try to access the stream of `azure.core.rest.HttpResponse` or `azure.core.rest.AsyncHttpResponse` once\nthe response stream has been consumed.\n\n#### StreamClosedError\n\nAn error thrown if you try to access the stream of the `azure.core.rest.HttpResponse` or `azure.core.rest.AsyncHttpResponse` once\nthe response stream has been closed.\n\n#### ResponseNotReadError\n\nAn error thrown if you try to access the `content` of `azure.core.rest.HttpResponse` or `azure.core.rest.AsyncHttpResponse` before\nreading in the response's bytes first.\n\n### Configurations\n\nWhen calling the methods, some properties can be configured by passing in as kwargs arguments.\n\n| Parameters | Description |\n| --- | --- |\n| headers | The HTTP Request headers. |\n| request_id | The request id to be added into header. |\n| user_agent | If specified, this will be added in front of the user agent string. |\n| logging_enable| Use to enable per operation. Defaults to `False`. |\n| logger | If specified, it will be used to log information. |\n| response_encoding | The encoding to use if known for this service (will disable auto-detection). |\n| raw_request_hook | Callback function. Will be invoked on request. |\n| raw_response_hook | Callback function. Will be invoked on response. |\n| network_span_namer | A callable to customize the span name. |\n| tracing_attributes | Attributes to set on all created spans. |\n| permit_redirects | Whether the client allows redirects. Defaults to `True`. |\n| redirect_max | The maximum allowed redirects. Defaults to `30`. |\n| retry_total | Total number of retries to allow. Takes precedence over other counts. Default value is `10`. |\n| retry_connect | How many connection-related errors to retry on. These are errors raised before the request is sent to the remote server, which we assume has not triggered the server to process the request. Default value is `3`. |\n| retry_read | How many times to retry on read errors. These errors are raised after the request was sent to the server, so the request may have side-effects. Default value is `3`. |\n| retry_status | How many times to retry on bad status codes. Default value is `3`. |\n| retry_backoff_factor | A backoff factor to apply between attempts after the second try (most errors are resolved immediately by a second try without a delay). Retry policy will sleep for: `{backoff factor} * (2 ** ({number of total retries} - 1))` seconds. If the backoff_factor is 0.1, then the retry will sleep for [0.0s, 0.2s, 0.4s, ...] between retries. The default value is `0.8`. |\n| retry_backoff_max | The maximum back off time. Default value is `120` seconds (2 minutes). |\n| retry_mode | Fixed or exponential delay between attempts, default is `Exponential`. |\n| timeout | Timeout setting for the operation in seconds, default is `604800`s (7 days). |\n| connection_timeout | A single float in seconds for the connection timeout. Defaults to `300` seconds. |\n| read_timeout | A single float in seconds for the read timeout. Defaults to `300` seconds. |\n| connection_verify | SSL certificate verification. Enabled by default. Set to False to disable, alternatively can be set to the path to a CA_BUNDLE file or directory with certificates of trusted CAs. |\n| connection_cert | Client-side certificates. You can specify a local cert to use as client side certificate, as a single file (containing the private key and the certificate) or as a tuple of both files' paths. |\n| proxies | Dictionary mapping protocol or protocol and hostname to the URL of the proxy. |\n| cookies | Dict or CookieJar object to send with the `Request`. |\n| connection_data_block_size | The block size of data sent over the connection. Defaults to `4096` bytes. |\n\n### Async transport\n\nThe async transport is designed to be opt-in. [AioHttp](https://pypi.org/project/aiohttp/) is one of the supported implementations of async transport. It is not installed by default. You need to install it separately.\n\n### Shared modules\n\n#### MatchConditions\n\nMatchConditions is an enum to describe match conditions.\n\n```python\nclass MatchConditions(Enum):\n Unconditionally = 1 # Matches any condition\n IfNotModified = 2 # If the target object is not modified. Usually it maps to etag=\n IfModified = 3 # Only if the target object is modified. Usually it maps to etag!=\n IfPresent = 4 # If the target object exists. Usually it maps to etag='*'\n IfMissing = 5 # If the target object does not exist. Usually it maps to etag!='*'\n```\n\n#### CaseInsensitiveEnumMeta\n\nA metaclass to support case-insensitive enums.\n\n```python\nfrom enum import Enum\n\nfrom azure.core import CaseInsensitiveEnumMeta\n\nclass MyCustomEnum(str, Enum, metaclass=CaseInsensitiveEnumMeta):\n FOO = 'foo'\n BAR = 'bar'\n```\n\n#### Null Sentinel Value\n\nA falsy sentinel object which is supposed to be used to specify attributes\nwith no data. This gets serialized to `null` on the wire.\n\n```python\nfrom azure.core.serialization import NULL\n\nassert bool(NULL) is False\n\nfoo = Foo(\n attr=NULL\n)\n```\n\n## Contributing\n\nThis project welcomes contributions and suggestions. Most contributions require\nyou to agree to a Contributor License Agreement (CLA) declaring that you have\nthe right to, and actually do, grant us the rights to use your contribution.\nFor details, visit [https://cla.microsoft.com](https://cla.microsoft.com).\n\nWhen you submit a pull request, a CLA-bot will automatically determine whether\nyou need to provide a CLA and decorate the PR appropriately (e.g., label,\ncomment). Simply follow the instructions provided by the bot. You will only\nneed to do this once across all repos using our CLA.\n\nThis project has adopted the\n[Microsoft Open Source Code of Conduct](https://opensource.microsoft.com/codeofconduct/).\nFor more information, see the\n[Code of Conduct FAQ](https://opensource.microsoft.com/codeofconduct/faq/)\nor contact [opencode@microsoft.com](mailto:opencode@microsoft.com) with any\nadditional questions or comments.\n\n\n[package]: https://pypi.org/project/azure-core/\n\n\n# Release History\n\n## 1.33.0 (2025-04-03)\n\n### Features Added\n\n- Added native OpenTelemetry tracing to Azure Core which enables users to use OpenTelemetry to trace Azure SDK operations without needing to install a plugin. #39563\n - To enable native OpenTelemetry tracing, users need to:\n 1. Have `opentelemetry-api` installed.\n 2. Ensure that `settings.tracing_implementation` is not set.\n 3. Ensure that `settings.tracing_enabled` is set to `True`.\n - If `setting.tracing_implementation` is set, the tracing plugin will be used instead of the native tracing.\n - If `settings.tracing_enabled` is set to `False`, tracing will be disabled.\n - The `OpenTelemetryTracer` class was added to the `azure.core.tracing.opentelemetry` module. This is a wrapper around the OpenTelemetry tracer that is used to create spans for Azure SDK operations.\n - Added a `get_tracer` method to the new `azure.core.instrumentation` module. This method returns an instance of the `OpenTelemetryTracer` class if OpenTelemetry is available.\n - A `TracingOptions` TypedDict class was added to define the options that SDK users can use to configure tracing per-operation. These options include the ability to enable or disable tracing and set additional attributes on spans.\n - Example usage: `client.method(tracing_options={\"enabled\": True, \"attributes\": {\"foo\": \"bar\"}})`\n - The `DistributedTracingPolicy` and `distributed_trace`/`distributed_trace_async` decorators now uses the OpenTelemetry tracer if it is available and native tracing is enabled.\n - SDK clients can define an `_instrumentation_config` class variable to configure the OpenTelemetry tracer used in method span creation. Possible configuration options are `library_name`, `library_version`, `schema_url`, and `attributes`.\n - `DistributedTracingPolicy` now accepts a `instrumentation_config` keyword argument to configure the OpenTelemetry tracer used in HTTP span creation.\n\n### Breaking Changes\n\n- Removed automatic tracing enablement for the OpenTelemetry plugin if `opentelemetry` was imported. To enable tracing with the plugin, please import `azure.core.settings.settings` and set `settings.tracing_implementation` to `\"opentelemetry\"`. #39563\n- In `DistributedTracingPolicy`, the default span name is now just the HTTP method (e.g., \"GET\", \"POST\") and no longer includes the URL path. This change was made to converge with the OpenTelemetry HTTP semantic conventions. The full URL is still included in the span attributes.\n- Renamed span attributes in `DistributedTracingPolicy`:\n - \"x-ms-client-request-id\" is now \"az.client_request_id\"\n - \"x-ms-request-id\" is now \"az.service_request_id\"\n\n### Bugs Fixed\n\n- Fixed an issue where the `traceparent` header was not being set correctly in the `DistributedTracingPolicy`. The `traceparent` header will now set based on the context of the HTTP client span. #40074\n\n### Other Changes\n\n- Added `opentelemetry-api` as an optional dependency for tracing. This can be installed with `pip install azure-core[tracing]`. #39563\n\n## 1.32.0 (2024-10-31)\n\n### Features Added\n\n- Added a default implementation to handle token challenges in `BearerTokenCredentialPolicy` and `AsyncBearerTokenCredentialPolicy`.\n\n### Bugs Fixed\n\n- Fixed an issue where the `tracing_attributes` keyword argument wasn't being handled at the request/method level. #38164\n\n### Other Changes\n\n- Log \"x-vss-e2eid\" and \"x-msedge-ref\" headers in `HttpLoggingPolicy`.\n\n## 1.31.0 (2024-09-12)\n\n### Features Added\n\n- Added azure.core.AzureClouds enum to represent the different Azure clouds.\n- Added two new credential protocol classes, `SupportsTokenInfo` and `AsyncSupportsTokenInfo`, to offer more extensibility in supporting various token acquisition scenarios. #36565\n - Each new protocol class defines a `get_token_info` method that returns an `AccessTokenInfo` object.\n- Added a new `TokenRequestOptions` class, which is a `TypedDict` with optional parameters, that can be used to define options for token requests through the `get_token_info` method. #36565\n- Added a new `AccessTokenInfo` class, which is returned by `get_token_info` implementations. This class contains the token, its expiration time, and optional additional information like when a token should be refreshed. #36565\n- `BearerTokenCredentialPolicy` and `AsyncBearerTokenCredentialPolicy` now first check if a credential has the `get_token_info` method defined. If so, the `get_token_info` method is used to acquire a token. Otherwise, the `get_token` method is used. #36565\n - These policies now also check the `refresh_on` attribute when determining if a new token request should be made.\n\n### Other Changes\n\n- The Azure Core OpenTelemetry tracing plugin will now be the preferred tracing plugin over the OpenCensus plugin. If both plugins are installed and `opentelemetry` is imported, then OpenTelemetry will be used to trace Azure SDK operations. #35050\n\n## 1.30.2 (2024-06-06)\n\n### Features Added\n\n- Tracing: `DistributedTracingPolicy` will now set an attribute, `http.request.resend_count`, on HTTP spans for resent requests to indicate the resend attempt number. #35069\n\n### Bugs Fixed\n\n- Raise correct exception if transport is used while already closed #35559\n\n### Other Changes\n\n- HTTP tracing spans will now include an `error.type` attribute if an error status code is returned. #34619\n- Minimum required Python version is now 3.8\n\n## 1.30.1 (2024-02-29)\n\n### Other Changes\n\n- Accept float for `retry_after` header. #34203\n\n## 1.30.0 (2024-02-01)\n\n### Features Added\n\n- Support tuple input for file values to `azure.core.rest.HttpRequest` #33948\n- Support tuple input to `files` with duplicate field names `azure.core.rest.HttpRequest` #34021\n\n## 1.29.7 (2024-01-18)\n\n### Other Changes\n\n- Removed dependency on `anyio`. #33282\n\n## 1.29.6 (2023-12-14)\n\n### Bugs Fixed\n\n- Adjusted `AsyncBearerTokenCredentialPolicy` to work properly with `trio` concurrency mechanisms. ([#33307](https://github.com/Azure/azure-sdk-for-python/pull/33307))\n\n### Other Changes\n\n- Added dependency on `anyio` >=3.0,<5.0\n- Bumped minimum dependency on `requests` to 2.21.0.\n\n## 1.29.5 (2023-10-19)\n\n### Bugs Fixed\n\n- Fixed an issue with `multipart/form-data` in the async transport where `data` was not getting encoded into the request body. #32473\n\n### Other Changes\n\n- Use ssl context from aiohttp by default.\n\n## 1.29.4 (2023-09-07)\n\n### Bugs Fixed\n\n- Fixed the issue that some urls trigger an infinite loop. #31346\n- Fixed issue where IndexError was raised if multipart responses did not match the number of requests. #31471\n- Fixed issue unbound variable exception if dict is invalid in CloudEvent.from_dict. #31835\n- Fixed issue asyncBearerTokenCredentialPolicy is not backward compatible with SansIOHTTPPolicy. #31836\n- Fixed issue mypy complains with new version of azure-core. #31564\n\n## 1.29.3 (2023-08-22)\n\n### Bugs Fixed\n\n- Typing fix: `message` cannot be `None` in `AzureError`. #31564\n\n## 1.29.2 (2023-08-14)\n\n### Bugs Fixed\n\n- Added a default implementation for `AsyncTokenCredential.__aexit__()` #31573\n\n### Other Changes\n\n- Bumped `typing-extensions` version to 4.6.0.\n\n## 1.29.1 (2023-08-09)\n\n### Bugs Fixed\n\n- Not pass `enabled_cae` unless it is explicitly enabled.\n\n## 1.29.0 (2023-08-03)\n\n### Features Added\n\n- A keyword argument `enable_cae` was added to the `get_token` method of the `TokenCredential` protocol. #31012\n- `BearerTokenCredentialPolicy` and `AsyncBearerTokenCredentialPolicy` now accept `enable_cae` keyword arguments in their constructors. This is used in determining if [Continuous Access Evaluation (CAE)](https://learn.microsoft.com/azure/active-directory/conditional-access/concept-continuous-access-evaluation) should be enabled for each `get_token` request. #31012\n\n## 1.28.0 (2023-07-06)\n\n### Features Added\n\n- Added header name parameter to `RequestIdPolicy`. #30772\n- Added `SensitiveHeaderCleanupPolicy` that cleans up sensitive headers if a redirect happens and the new destination is in another domain. #28349\n\n### Other Changes\n\n- Catch aiohttp errors and translate them into azure-core errors.\n\n## 1.27.1 (2023-06-13)\n\n### Bugs Fixed\n\n- Fix url building for some complex query parameters scenarios #30707\n\n## 1.27.0 (2023-06-01)\n\n### Features Added\n\n- Added support to use sync credentials in `AsyncBearerTokenCredentialPolicy`. #30381\n- Added \"prefix\" parameter to AzureKeyCredentialPolicy #29901\n\n### Bugs Fixed\n\n- Improve error message when providing the wrong credential type for AzureKeyCredential #30380\n\n## 1.26.4 (2023-04-06)\n\n### Features Added\n\n- Updated settings to include OpenTelemetry as a tracer provider. #29095\n\n### Other Changes\n\n- Improved typing\n\n## 1.26.3 (2023-02-02)\n\n### Bugs Fixed\n\n- Fixed deflate decompression for aiohttp #28483\n\n## 1.26.2 (2023-01-05)\n\n### Bugs Fixed\n\n- Fix 'ClientSession' object has no attribute 'auto_decompress' (thanks to @mghextreme for the contribution)\n\n### Other Changes\n\n- Add \"x-ms-error-code\" as secure header to log\n- Rename \"DEFAULT_HEADERS_WHITELIST\" to \"DEFAULT_HEADERS_ALLOWLIST\". Added a backward compatible alias.\n\n## 1.26.1 (2022-11-03)\n\n### Other Changes\n\n- Added example of RequestsTransport with custom session. (thanks to @inirudebwoy for the contribution) #26768\n- Added Python 3.11 support.\n\n## 1.26.0 (2022-10-06)\n\n### Other Changes\n\n- LRO polling will not wait anymore before doing the first status check #26376\n- Added extra dependency for [aio]. pip install azure-core[aio] installs aiohttp too.\n\n## 1.25.1 (2022-09-01)\n\n### Bugs Fixed\n\n- Added @runtime_checkable to `TokenCredential` protocol definitions #25187\n\n## 1.25.0 (2022-08-04)\n\nAzure-core is supported on Python 3.7 or later. For more details, please read our page on [Azure SDK for Python version support policy](https://github.com/Azure/azure-sdk-for-python/wiki/Azure-SDKs-Python-version-support-policy).\n\n### Features Added\n\n- Added `CaseInsensitiveDict` implementation in `azure.core.utils` removing dependency on `requests` and `aiohttp`\n\n## 1.24.2 (2022-06-30)\n\n### Bugs Fixed\n\n- Fixed the bug that azure-core could not be imported under Python 3.11.0b3 #24928\n- `ContentDecodePolicy` can now correctly deserialize more JSON bodies with different mime types #22410\n\n## 1.24.1 (2022-06-01)\n\n### Bugs Fixed\n\n- Declare method level span as INTERNAL by default #24492\n- Fixed type hints for `azure.core.paging.ItemPaged` #24548\n\n## 1.24.0 (2022-05-06)\n\n### Features Added\n\n- Add `SerializationError` and `DeserializationError` in `azure.core.exceptions` for errors raised during serialization / deserialization #24312\n\n## 1.23.1 (2022-03-31)\n\n### Bugs Fixed\n\n- Allow stream inputs to the `content` kwarg of `azure.core.rest.HttpRequest` from objects with a `read` method #23578\n\n## 1.23.0 (2022-03-03)\n\n### Features Added\n\n- Improve intellisense type hinting for service client methods. #22891\n\n- Add a case insensitive dict `case_insensitive_dict` in `azure.core.utils`. #23206\n\n### Bugs Fixed\n\n- Use \"\\n\" rather than \"/n\" for new line in log. #23261\n\n### Other Changes\n\n- Log \"WWW-Authenticate\" header in `HttpLoggingPolicy` #22990\n- Added dependency on `typing-extensions` >= 4.0.1\n\n## 1.22.1 (2022-02-09)\n\n### Bugs Fixed\n\n- Limiting `final-state-via` scope to POST until consuming SDKs has been fixed to use this option properly on PUT. #22989\n\n## 1.22.0 (2022-02-03)\n_[**This version is deprecated.**]_\n\n### Features Added\n\n- Add support for `final-state-via` LRO option in core. #22713\n\n### Bugs Fixed\n\n- Add response body to string representation of `HttpResponseError` if we're not able to parse out information #22302\n- Raise `AttributeError` when calling azure.core.pipeline.transport.\\_\\_bases__ #22469\n\n### Other Changes\n\n- Python 2.7 is no longer supported. Please use Python version 3.6 or later.\n\n## 1.21.1 (2021-12-06)\n\n### Other Changes\n\n- Revert change in str method #22023\n\n## 1.21.0 (2021-12-02)\n\n### Breaking Changes\n\n- Sync stream downloading now raises `azure.core.exceptions.DecodeError` rather than `requests.exceptions.ContentDecodingError`\n\n### Bugs Fixed\n\n- Add response body to string representation of `HttpResponseError` if we're not able to parse out information #21800\n\n## 1.20.1 (2021-11-08)\n\n### Bugs Fixed\n\n- Correctly set response's content to decompressed body when users are using aiohttp transport with decompression headers #21620\n\n## 1.20.0 (2021-11-04)\n\n### Features Added\n\n- GA `send_request` onto the `azure.core.PipelineClient` and `azure.core.AsyncPipelineClient`. This method takes in\nrequests and sends them through our pipelines.\n- GA `azure.core.rest`. `azure.core.rest` is our new public simple HTTP library in `azure.core` that users will use to create requests, and consume responses.\n- GA errors `StreamConsumedError`, `StreamClosedError`, and `ResponseNotReadError` to `azure.core.exceptions`. These errors\nare thrown if you mishandle streamed responses from the `azure.core.rest` module\n- add kwargs to the methods for `iter_raw` and `iter_bytes` #21529\n- no longer raise JSON errors if users pass in file descriptors of JSON to the `json` kwarg in `HttpRequest` #21504\n- Added new error type `IncompleteReadError` which is raised if peer closes the connection before we have received the complete message body.\n\n### Breaking Changes\n\n- SansIOHTTPPolicy.on_exception returns None instead of bool.\n\n### Bugs Fixed\n\n- The `Content-Length` header in a http response is strictly checked against the actual number of bytes in the body,\n rather than silently truncating data in case the underlying tcp connection is closed prematurely.\n (thanks to @jochen-ott-by for the contribution) #20412\n- UnboundLocalError when SansIOHTTPPolicy handles an exception #15222\n- Add default content type header of `text/plain` and content length header for users who pass unicode strings to the `content` kwarg of `HttpRequest` in 2.7 #21550\n\n## 1.19.1 (2021-11-01)\n\n### Bugs Fixed\n\n- respect text encoding specified in argument (thanks to @ryohji for the contribution) #20796\n- Fix \"coroutine x.read() was never awaited\" warning from `ContentDecodePolicy` #21318\n- fix type check for `data` input to `azure.core.rest` for python 2.7 users #21341\n- use `charset_normalizer` if `chardet` is not installed to migrate aiohttp 3.8.0 changes.\n\n### Other Changes\n\n- Refactor AzureJSONEncoder (thanks to @Codejune for the contribution) #21028\n\n## 1.19.0 (2021-09-30)\n\n### Breaking Changes in the Provisional `azure.core.rest` package\n\n- `azure.core.rest.HttpResponse` and `azure.core.rest.AsyncHttpResponse` are now abstract base classes. They should not be initialized directly, instead\nyour transport responses should inherit from them and implement them.\n- The properties of the `azure.core.rest` responses are now all read-only\n\n- HttpLoggingPolicy integrates logs into one record #19925\n\n## 1.18.0 (2021-09-02)\n\n### Features Added\n\n- `azure.core.serialization.AzureJSONEncoder` (introduced in 1.17.0) serializes `datetime.datetime` objects in ISO 8601 format, conforming to RFC 3339's specification. #20190\n- We now use `azure.core.serialization.AzureJSONEncoder` to serialize `json` input to `azure.core.rest.HttpRequest`.\n\n### Breaking Changes in the Provisional `azure.core.rest` package\n\n- The `text` property on `azure.core.rest.HttpResponse` and `azure.core.rest.AsyncHttpResponse` has changed to a method, which also takes\nan `encoding` parameter.\n- Removed `iter_text` and `iter_lines` from `azure.core.rest.HttpResponse` and `azure.core.rest.AsyncHttpResponse`\n\n### Bugs Fixed\n\n- The behaviour of the headers returned in `azure.core.rest` responses now aligns across sync and async. Items can now be checked case-insensitively and without raising an error for format.\n\n## 1.17.0 (2021-08-05)\n\n### Features Added\n\n- Cut hard dependency on requests library\n- Added a `from_json` method which now accepts storage QueueMessage, eventhub's EventData or ServiceBusMessage or simply json bytes to return a `CloudEvent`\n\n### Fixed\n\n- Not override \"x-ms-client-request-id\" if it already exists in the header. #17757\n\n### Breaking Changes in the Provisional `azure.core.rest` package\n\n- `azure.core.rest` will not try to guess the `charset` anymore if it was impossible to extract it from `HttpResponse` analysis. This removes our dependency on `charset`.\n\n## 1.16.0 (2021-07-01)\n\n### Features Added\n\n- Add new ***provisional*** methods `send_request` onto the `azure.core.PipelineClient` and `azure.core.AsyncPipelineClient`. This method takes in\nrequests and sends them through our pipelines.\n- Add new ***provisional*** module `azure.core.rest`. `azure.core.rest` is our new public simple HTTP library in `azure.core` that users will use to create requests, and consume responses.\n- Add new ***provisional*** errors `StreamConsumedError`, `StreamClosedError`, and `ResponseNotReadError` to `azure.core.exceptions`. These errors\nare thrown if you mishandle streamed responses from the provisional `azure.core.rest` module\n\n### Fixed\n\n- Improved error message in the `from_dict` method of `CloudEvent` when a wrong schema is sent.\n\n## 1.15.0 (2021-06-04)\n\n### New Features\n\n- Added `BearerTokenCredentialPolicy.on_challenge` and `.authorize_request` to allow subclasses to optionally handle authentication challenges\n\n### Bug Fixes\n\n- Retry policies don't sleep after operations time out\n- The `from_dict` methhod in the `CloudEvent` can now convert a datetime string to datetime object when microsecond exceeds the python limitation\n\n## 1.14.0 (2021-05-13)\n\n### New Features\n\n- Added `azure.core.credentials.AzureNamedKeyCredential` credential #17548.\n- Added `decompress` parameter for `stream_download` method. If it is set to `False`, will not do decompression upon the stream. #17920\n\n## 1.13.0 (2021-04-02)\n\nAzure core requires Python 2.7 or Python 3.6+ since this release.\n\n### New Features\n\n- Added `azure.core.utils.parse_connection_string` function to parse connection strings across SDKs, with common validation and support for case insensitive keys.\n- Supported adding custom policies #16519\n- Added `~azure.core.tracing.Link` that should be used while passing `Links` to `AbstractSpan`.\n- `AbstractSpan` constructor can now take in additional keyword only args.\n\n### Bug fixes\n\n- Make NetworkTraceLoggingPolicy show the auth token in plain text. #14191\n- Fixed RetryPolicy overriding default connection timeout with an extreme value #17481\n\n## 1.12.0 (2021-03-08)\n\nThis version will be the last version to officially support Python 3.5, future versions will require Python 2.7 or Python 3.6+.\n\n### Features\n\n- Added `azure.core.messaging.CloudEvent` model that follows the cloud event spec.\n- Added `azure.core.serialization.NULL` sentinel value\n- Improve `repr`s for `HttpRequest` and `HttpResponse`s #16972\n\n### Bug Fixes\n\n- Disable retry in stream downloading. (thanks to @jochen-ott-by @hoffmann for the contribution) #16723\n\n## 1.11.0 (2021-02-08)\n\n### Features\n\n- Added `CaseInsensitiveEnumMeta` class for case-insensitive enums. #16316\n- Add `raise_for_status` method onto `HttpResponse`. Calling `response.raise_for_status()` on a response with an error code\nwill raise an `HttpResponseError`. Calling it on a good response will do nothing #16399\n\n### Bug Fixes\n\n- Update conn.conn_kw rather than overriding it when setting block size. (thanks for @jiasli for the contribution) #16587\n\n## 1.10.0 (2021-01-11)\n\n### Features\n\n- Added `AzureSasCredential` and its respective policy. #15946\n\n## 1.9.0 (2020-11-09)\n\n### Features\n\n- Add a `continuation_token` attribute to the base `AzureError` exception, and set this value for errors raised\n during paged or long-running operations.\n\n### Bug Fixes\n\n- Set retry_interval to 1 second instead of 1000 seconds (thanks **vbarbaresi** for contributing) #14357\n\n\n## 1.8.2 (2020-10-05)\n\n### Bug Fixes\n\n- Fixed bug to allow polling in the case of parameterized endpoints with relative polling urls #14097\n\n\n## 1.8.1 (2020-09-08)\n\n### Bug fixes\n\n- SAS credential replicated \"/\" fix #13159\n\n## 1.8.0 (2020-08-10)\n\n### Features\n\n- Support params as list for exploding parameters #12410\n\n\n## 1.7.0 (2020-07-06)\n\n### Bug fixes\n\n- `AzureKeyCredentialPolicy` will now accept (and ignore) passed in kwargs #11963\n- Better error messages if passed endpoint is incorrect #12106\n- Do not JSON encore a string if content type is \"text\" #12137\n\n### Features\n\n- Added `http_logging_policy` property on the `Configuration` object, allowing users to individually\nset the http logging policy of the config #12218\n\n## 1.6.0 (2020-06-03)\n\n### Bug fixes\n\n- Fixed deadlocks in AsyncBearerTokenCredentialPolicy #11543\n- Fix AttributeException in StreamDownloadGenerator #11462\n\n### Features\n\n- Added support for changesets as part of multipart message support #10485\n- Add AsyncLROPoller in azure.core.polling #10801\n- Add get_continuation_token/from_continuation_token/polling_method methods in pollers (sync and async) #10801\n- HttpResponse and PipelineContext objects are now pickable #10801\n\n## 1.5.0 (2020-05-04)\n\n### Features\n\n- Support \"x-ms-retry-after-ms\" in response header #10743\n- `link` and `link_from_headers` now accepts attributes #10765\n\n### Bug fixes\n\n- Not retry if the status code is less than 400 #10778\n- \"x-ms-request-id\" is not considered safe header for logging #10967\n\n## 1.4.0 (2020-04-06)\n\n### Features\n\n- Support a default error type in map_error #9773\n- Added `AzureKeyCredential` and its respective policy. #10509\n- Added `azure.core.polling.base_polling` module with a \"Microsoft One API\" polling implementation #10090\n Also contains the async version in `azure.core.polling.async_base_polling`\n- Support kwarg `enforce_https` to disable HTTPS check on authentication #9821\n- Support additional kwargs in `HttpRequest.set_multipart_mixed` that will be passed into pipeline context.\n\n## 1.3.0 (2020-03-09)\n\n### Bug fixes\n\n- Appended RequestIdPolicy to the default pipeline #9841\n- Rewind the body position in async_retry #10117\n\n### Features\n\n- Add raw_request_hook support in custom_hook_policy #9958\n- Add timeout support in retry_policy #10011\n- Add OdataV4 error format auto-parsing in all exceptions ('error' attribute) #9738\n\n## 1.2.2 (2020-02-10)\n\n### Bug fixes\n\n- Fixed a bug that sends None as request_id #9545\n- Enable mypy for customers #9572\n- Handle TypeError in deep copy #9620\n- Fix text/plain content-type in decoder #9589\n\n## 1.2.1 (2020-01-14)\n\n### Bug fixes\n\n- Fixed a regression in 1.2.0 that was incompatible with azure-keyvault-* 4.0.0\n[#9462](https://github.com/Azure/azure-sdk-for-python/issues/9462)\n\n\n## 1.2.0 (2020-01-14)\n\n### Features\n\n- Add user_agent & sdk_moniker kwargs in UserAgentPolicy init #9355\n- Support OPTIONS HTTP verb #9322\n- Add tracing_attributes to tracing decorator #9297\n- Support auto_request_id in RequestIdPolicy #9163\n- Support fixed retry #6419\n- Support \"retry-after-ms\" in response header #9240\n\n### Bug fixes\n\n- Removed `__enter__` and `__exit__` from async context managers #9313\n\n## 1.1.1 (2019-12-03)\n\n### Bug fixes\n\n- Bearer token authorization requires HTTPS\n- Rewind the body position in retry #8307\n\n## 1.1.0 (2019-11-25)\n\n### Features\n\n- New RequestIdPolicy #8437\n- Enable logging policy in default pipeline #8053\n- Normalize transport timeout. #8000\n Now we have:\n * 'connection_timeout' - a single float in seconds for the connection timeout. Default 5min\n * 'read_timeout' - a single float in seconds for the read timeout. Default 5min\n\n### Bug fixes\n\n- RequestHistory: deepcopy fails if request contains a stream #7732\n- Retry: retry raises error if response does not have http_response #8629\n- Client kwargs are now passed to DistributedTracingPolicy correctly #8051\n- NetworkLoggingPolicy now logs correctly all requests in case of retry #8262\n\n## 1.0.0 (2019-10-29)\n\n### Features\n\n- Tracing: DistributedTracingPolicy now accepts kwargs network_span_namer to change network span name #7773\n- Tracing: Implementation of AbstractSpan can now use the mixin HttpSpanMixin to get HTTP span update automatically #7773\n- Tracing: AbstractSpan contract \"change_context\" introduced #7773\n- Introduce new policy HttpLoggingPolicy #7988\n\n### Bug fixes\n\n- Fix AsyncioRequestsTransport if input stream is an async generator #7743\n- Fix form-data with aiohttp transport #7749\n\n### Breaking changes\n\n- Tracing: AbstractSpan.set_current_span is longer supported. Use change_context instead. #7773\n- azure.core.pipeline.policies.ContentDecodePolicy.deserialize_from_text changed\n\n## 1.0.0b4 (2019-10-07)\n\n### Features\n\n- Tracing: network span context is available with the TRACING_CONTEXT in pipeline response #7252\n- Tracing: Span contract now has `kind`, `traceparent` and is a context manager #7252\n- SansIOHTTPPolicy methods can now be coroutines #7497\n- Add multipart/mixed support #7083:\n\n - HttpRequest now has a \"set_multipart_mixed\" method to set the parts of this request\n - HttpRequest now has a \"prepare_multipart_body\" method to build final body.\n - HttpResponse now has a \"parts\" method to return an iterator of parts\n - AsyncHttpResponse now has a \"parts\" methods to return an async iterator of parts\n - Note that multipart/mixed is a Python 3.x only feature\n\n### Bug fixes\n\n- Tracing: policy cannot fail the pipeline, even in the worst condition #7252\n- Tracing: policy pass correctly status message if exception #7252\n- Tracing: incorrect span if exception raised from decorated function #7133\n- Fixed urllib3 ConnectTimeoutError being raised by Requests during a socket timeout. Now this exception is caught and wrapped as a `ServiceRequestError` #7542\n\n### Breaking changes\n\n- Tracing: `azure.core.tracing.context` removed\n- Tracing: `azure.core.tracing.context.tracing_context.with_current_context` renamed to `azure.core.tracing.common.with_current_context` #7252\n- Tracing: `link` renamed `link_from_headers` and `link` takes now a string\n- Tracing: opencensus implementation has been moved to the package `azure-core-tracing-opencensus`\n- Some modules and classes that were importables from several different places have been removed:\n\n - `azure.core.HttpResponseError` is now only `azure.core.exceptions.HttpResponseError`\n - `azure.core.Configuration` is now only `azure.core.configuration.Configuration`\n - `azure.core.HttpRequest` is now only `azure.core.pipeline.transport.HttpRequest`\n - `azure.core.version` module has been removed. Use `azure.core.__version__` to get version number.\n - `azure.core.pipeline_client` has been removed. Import from `azure.core` instead.\n - `azure.core.pipeline_client_async` has been removed. Import from `azure.core` instead.\n - `azure.core.pipeline.base` has been removed. Import from `azure.core.pipeline` instead.\n - `azure.core.pipeline.base_async` has been removed. Import from `azure.core.pipeline` instead.\n - `azure.core.pipeline.policies.base` has been removed. Import from `azure.core.pipeline.policies` instead.\n - `azure.core.pipeline.policies.base_async` has been removed. Import from `azure.core.pipeline.policies` instead.\n - `azure.core.pipeline.policies.authentication` has been removed. Import from `azure.core.pipeline.policies` instead.\n - `azure.core.pipeline.policies.authentication_async` has been removed. Import from `azure.core.pipeline.policies` instead.\n - `azure.core.pipeline.policies.custom_hook` has been removed. Import from `azure.core.pipeline.policies` instead.\n - `azure.core.pipeline.policies.redirect` has been removed. Import from `azure.core.pipeline.policies` instead.\n - `azure.core.pipeline.policies.redirect_async` has been removed. Import from `azure.core.pipeline.policies` instead.\n - `azure.core.pipeline.policies.retry` has been removed. Import from `azure.core.pipeline.policies` instead.\n - `azure.core.pipeline.policies.retry_async` has been removed. Import from `azure.core.pipeline.policies` instead.\n - `azure.core.pipeline.policies.distributed_tracing` has been removed. Import from `azure.core.pipeline.policies` instead.\n - `azure.core.pipeline.policies.universal` has been removed. Import from `azure.core.pipeline.policies` instead.\n - `azure.core.tracing.abstract_span` has been removed. Import from `azure.core.tracing` instead.\n - `azure.core.pipeline.transport.base` has been removed. Import from `azure.core.pipeline.transport` instead.\n - `azure.core.pipeline.transport.base_async` has been removed. Import from `azure.core.pipeline.transport` instead.\n - `azure.core.pipeline.transport.requests_basic` has been removed. Import from `azure.core.pipeline.transport` instead.\n - `azure.core.pipeline.transport.requests_asyncio` has been removed. Import from `azure.core.pipeline.transport` instead.\n - `azure.core.pipeline.transport.requests_trio` has been removed. Import from `azure.core.pipeline.transport` instead.\n - `azure.core.pipeline.transport.aiohttp` has been removed. Import from `azure.core.pipeline.transport` instead.\n - `azure.core.polling.poller` has been removed. Import from `azure.core.polling` instead.\n - `azure.core.polling.async_poller` has been removed. Import from `azure.core.polling` instead.\n\n## 1.0.0b3 (2019-09-09)\n\n### Bug fixes\n\n- Fix aiohttp auto-headers #6992\n- Add tracing to policies module init #6951\n\n## 1.0.0b2 (2019-08-05)\n\n### Breaking changes\n\n- Transport classes don't take `config` parameter anymore (use kwargs instead) #6372\n- `azure.core.paging` has been completely refactored #6420\n- HttpResponse.content_type attribute is now a string (was a list) #6490\n- For `StreamDownloadGenerator` subclasses, `response` is now an `HttpResponse`, and not a transport response like `aiohttp.ClientResponse` or `requests.Response`. The transport response is available in `internal_response` attribute #6490\n\n### Bug fixes\n\n- aiohttp is not required to import async pipelines classes #6496\n- `AsyncioRequestsTransport.sleep` is now a coroutine as expected #6490\n- `RequestsTransport` is not tight to `ProxyPolicy` implementation details anymore #6372\n- `AiohttpTransport` does not raise on unexpected kwargs #6355\n\n### Features\n\n- New paging base classes that support `continuation_token` and `by_page()` #6420\n- Proxy support for `AiohttpTransport` #6372\n\n## 1.0.0b1 (2019-06-26)\n\n- Preview 1 release", - "release_date": "2025-04-03T23:51:03", + "description": "Microsoft Azure Core Library for Python\n# Azure Core shared client library for Python\n\nAzure core provides shared exceptions and modules for Python SDK client libraries.\nThese libraries follow the [Azure SDK Design Guidelines for Python](https://azure.github.io/azure-sdk/python/guidelines/index.html) .\n\nIf you are a client library developer, please reference [client library developer reference](https://github.com/Azure/azure-sdk-for-python/blob/main/sdk/core/azure-core/CLIENT_LIBRARY_DEVELOPER.md) for more information.\n\n[Source code](https://github.com/Azure/azure-sdk-for-python/blob/main/sdk/core/azure-core/)\n| [Package (Pypi)][package]\n| [Package (Conda)](https://anaconda.org/microsoft/azure-core/)\n| [API reference documentation](https://learn.microsoft.com/python/api/overview/azure/core-readme)\n\n## Getting started\n\nTypically, you will not need to install azure core;\nit will be installed when you install one of the client libraries using it.\nIn case you want to install it explicitly (to implement your own client library, for example),\nyou can find it [here](https://pypi.org/project/azure-core/).\n\n## Key concepts\n\n### Azure Core Library Exceptions\n\n#### AzureError\n\nAzureError is the base exception for all errors.\n\n```python\nclass AzureError(Exception):\n def __init__(self, message, *args, **kwargs):\n self.inner_exception = kwargs.get(\"error\")\n self.exc_type, self.exc_value, self.exc_traceback = sys.exc_info()\n self.exc_type = self.exc_type.__name__ if self.exc_type else type(self.inner_exception)\n self.exc_msg = \"{}, {}: {}\".format(message, self.exc_type, self.exc_value) # type: ignore\n self.message = str(message)\n self.continuation_token = kwargs.get(\"continuation_token\")\n super(AzureError, self).__init__(self.message, *args)\n```\n\n*message* is any message (str) to be associated with the exception.\n\n*args* are any additional args to be included with exception.\n\n*kwargs* are keyword arguments to include with the exception. Use the keyword *error* to pass in an internal exception and *continuation_token* for a token reference to continue an incomplete operation.\n\n**The following exceptions inherit from AzureError:**\n\n#### ServiceRequestError\n\nAn error occurred while attempt to make a request to the service. No request was sent.\n\n#### ServiceResponseError\n\nThe request was sent, but the client failed to understand the response.\nThe connection may have timed out. These errors can be retried for idempotent or safe operations.\n\n#### HttpResponseError\n\nA request was made, and a non-success status code was received from the service.\n\n```python\nclass HttpResponseError(AzureError):\n def __init__(self, message=None, response=None, **kwargs):\n self.reason = None\n self.response = response\n if response:\n self.reason = response.reason\n self.status_code = response.status_code\n self.error = self._parse_odata_body(ODataV4Format, response) # type: Optional[ODataV4Format]\n if self.error:\n message = str(self.error)\n else:\n message = message or \"Operation returned an invalid status '{}'\".format(\n self.reason\n )\n\n super(HttpResponseError, self).__init__(message=message, **kwargs)\n```\n\n*message* is the HTTP response error message (optional)\n\n*response* is the HTTP response (optional).\n\n*kwargs* are keyword arguments to include with the exception.\n\n**The following exceptions inherit from HttpResponseError:**\n\n#### DecodeError\n\nAn error raised during response de-serialization.\n\n#### IncompleteReadError\n\nAn error raised if peer closes the connection before we have received the complete message body.\n\n#### ResourceExistsError\n\nAn error response with status code 4xx. This will not be raised directly by the Azure core pipeline.\n\n#### ResourceNotFoundError\n\nAn error response, typically triggered by a 412 response (for update) or 404 (for get/post).\n\n#### ResourceModifiedError\n\nAn error response with status code 4xx, typically 412 Conflict. This will not be raised directly by the Azure core pipeline.\n\n#### ResourceNotModifiedError\n\nAn error response with status code 304. This will not be raised directly by the Azure core pipeline.\n\n#### ClientAuthenticationError\n\nAn error response with status code 4xx. This will not be raised directly by the Azure core pipeline.\n\n#### TooManyRedirectsError\n\nAn error raised when the maximum number of redirect attempts is reached. The maximum amount of redirects can be configured in the RedirectPolicy.\n\n```python\nclass TooManyRedirectsError(HttpResponseError):\n def __init__(self, history, *args, **kwargs):\n self.history = history\n message = \"Reached maximum redirect attempts.\"\n super(TooManyRedirectsError, self).__init__(message, *args, **kwargs)\n```\n\n*history* is used to document the requests/responses that resulted in redirected requests.\n\n*args* are any additional args to be included with exception.\n\n*kwargs* are keyword arguments to include with the exception.\n\n#### StreamConsumedError\n\nAn error thrown if you try to access the stream of `azure.core.rest.HttpResponse` or `azure.core.rest.AsyncHttpResponse` once\nthe response stream has been consumed.\n\n#### StreamClosedError\n\nAn error thrown if you try to access the stream of the `azure.core.rest.HttpResponse` or `azure.core.rest.AsyncHttpResponse` once\nthe response stream has been closed.\n\n#### ResponseNotReadError\n\nAn error thrown if you try to access the `content` of `azure.core.rest.HttpResponse` or `azure.core.rest.AsyncHttpResponse` before\nreading in the response's bytes first.\n\n### Configurations\n\nWhen calling the methods, some properties can be configured by passing in as kwargs arguments.\n\n| Parameters | Description |\n| --- | --- |\n| headers | The HTTP Request headers. |\n| request_id | The request id to be added into header. |\n| user_agent | If specified, this will be added in front of the user agent string. |\n| logging_enable| Use to enable per operation. Defaults to `False`. |\n| logger | If specified, it will be used to log information. |\n| response_encoding | The encoding to use if known for this service (will disable auto-detection). |\n| raw_request_hook | Callback function. Will be invoked on request. |\n| raw_response_hook | Callback function. Will be invoked on response. |\n| network_span_namer | A callable to customize the span name. |\n| tracing_attributes | Attributes to set on all created spans. |\n| permit_redirects | Whether the client allows redirects. Defaults to `True`. |\n| redirect_max | The maximum allowed redirects. Defaults to `30`. |\n| retry_total | Total number of retries to allow. Takes precedence over other counts. Default value is `10`. |\n| retry_connect | How many connection-related errors to retry on. These are errors raised before the request is sent to the remote server, which we assume has not triggered the server to process the request. Default value is `3`. |\n| retry_read | How many times to retry on read errors. These errors are raised after the request was sent to the server, so the request may have side-effects. Default value is `3`. |\n| retry_status | How many times to retry on bad status codes. Default value is `3`. |\n| retry_backoff_factor | A backoff factor to apply between attempts after the second try (most errors are resolved immediately by a second try without a delay). Retry policy will sleep for: `{backoff factor} * (2 ** ({number of total retries} - 1))` seconds. If the backoff_factor is 0.1, then the retry will sleep for [0.0s, 0.2s, 0.4s, ...] between retries. The default value is `0.8`. |\n| retry_backoff_max | The maximum back off time. Default value is `120` seconds (2 minutes). |\n| retry_mode | Fixed or exponential delay between attempts, default is `Exponential`. |\n| timeout | Timeout setting for the operation in seconds, default is `604800`s (7 days). |\n| connection_timeout | A single float in seconds for the connection timeout. Defaults to `300` seconds. |\n| read_timeout | A single float in seconds for the read timeout. Defaults to `300` seconds. |\n| connection_verify | SSL certificate verification. Enabled by default. Set to False to disable, alternatively can be set to the path to a CA_BUNDLE file or directory with certificates of trusted CAs. |\n| connection_cert | Client-side certificates. You can specify a local cert to use as client side certificate, as a single file (containing the private key and the certificate) or as a tuple of both files' paths. |\n| proxies | Dictionary mapping protocol or protocol and hostname to the URL of the proxy. |\n| cookies | Dict or CookieJar object to send with the `Request`. |\n| connection_data_block_size | The block size of data sent over the connection. Defaults to `4096` bytes. |\n\n### Async transport\n\nThe async transport is designed to be opt-in. [AioHttp](https://pypi.org/project/aiohttp/) is one of the supported implementations of async transport. It is not installed by default. You need to install it separately.\n\n### Shared modules\n\n#### MatchConditions\n\nMatchConditions is an enum to describe match conditions.\n\n```python\nclass MatchConditions(Enum):\n Unconditionally = 1 # Matches any condition\n IfNotModified = 2 # If the target object is not modified. Usually it maps to etag=\n IfModified = 3 # Only if the target object is modified. Usually it maps to etag!=\n IfPresent = 4 # If the target object exists. Usually it maps to etag='*'\n IfMissing = 5 # If the target object does not exist. Usually it maps to etag!='*'\n```\n\n#### CaseInsensitiveEnumMeta\n\nA metaclass to support case-insensitive enums.\n\n```python\nfrom enum import Enum\n\nfrom azure.core import CaseInsensitiveEnumMeta\n\nclass MyCustomEnum(str, Enum, metaclass=CaseInsensitiveEnumMeta):\n FOO = 'foo'\n BAR = 'bar'\n```\n\n#### Null Sentinel Value\n\nA falsy sentinel object which is supposed to be used to specify attributes\nwith no data. This gets serialized to `null` on the wire.\n\n```python\nfrom azure.core.serialization import NULL\n\nassert bool(NULL) is False\n\nfoo = Foo(\n attr=NULL\n)\n```\n\n## Contributing\n\nThis project welcomes contributions and suggestions. Most contributions require\nyou to agree to a Contributor License Agreement (CLA) declaring that you have\nthe right to, and actually do, grant us the rights to use your contribution.\nFor details, visit [https://cla.microsoft.com](https://cla.microsoft.com).\n\nWhen you submit a pull request, a CLA-bot will automatically determine whether\nyou need to provide a CLA and decorate the PR appropriately (e.g., label,\ncomment). Simply follow the instructions provided by the bot. You will only\nneed to do this once across all repos using our CLA.\n\nThis project has adopted the\n[Microsoft Open Source Code of Conduct](https://opensource.microsoft.com/codeofconduct/).\nFor more information, see the\n[Code of Conduct FAQ](https://opensource.microsoft.com/codeofconduct/faq/)\nor contact [opencode@microsoft.com](mailto:opencode@microsoft.com) with any\nadditional questions or comments.\n\n\n[package]: https://pypi.org/project/azure-core/\n\n\n# Release History\n\n## 1.34.0 (2025-05-01)\n\n### Features Added\n\n- Added a `set_span_error_status` method to the `OpenTelemetryTracer` class. This method allows users to set the status of a span to `ERROR` after it has been created. #40703\n\n### Other Changes\n\n- Python 3.8 is no longer supported. Please use Python version 3.9 or later.\n\n## 1.33.0 (2025-04-03)\n\n### Features Added\n\n- Added native OpenTelemetry tracing to Azure Core which enables users to use OpenTelemetry to trace Azure SDK operations without needing to install a plugin. #39563\n - To enable native OpenTelemetry tracing, users need to:\n 1. Have `opentelemetry-api` installed.\n 2. Ensure that `settings.tracing_implementation` is not set.\n 3. Ensure that `settings.tracing_enabled` is set to `True`.\n - If `setting.tracing_implementation` is set, the tracing plugin will be used instead of the native tracing.\n - If `settings.tracing_enabled` is set to `False`, tracing will be disabled.\n - The `OpenTelemetryTracer` class was added to the `azure.core.tracing.opentelemetry` module. This is a wrapper around the OpenTelemetry tracer that is used to create spans for Azure SDK operations.\n - Added a `get_tracer` method to the new `azure.core.instrumentation` module. This method returns an instance of the `OpenTelemetryTracer` class if OpenTelemetry is available.\n - A `TracingOptions` TypedDict class was added to define the options that SDK users can use to configure tracing per-operation. These options include the ability to enable or disable tracing and set additional attributes on spans.\n - Example usage: `client.method(tracing_options={\"enabled\": True, \"attributes\": {\"foo\": \"bar\"}})`\n - The `DistributedTracingPolicy` and `distributed_trace`/`distributed_trace_async` decorators now uses the OpenTelemetry tracer if it is available and native tracing is enabled.\n - SDK clients can define an `_instrumentation_config` class variable to configure the OpenTelemetry tracer used in method span creation. Possible configuration options are `library_name`, `library_version`, `schema_url`, and `attributes`.\n - `DistributedTracingPolicy` now accepts a `instrumentation_config` keyword argument to configure the OpenTelemetry tracer used in HTTP span creation.\n\n### Breaking Changes\n\n- Removed automatic tracing enablement for the OpenTelemetry plugin if `opentelemetry` was imported. To enable tracing with the plugin, please import `azure.core.settings.settings` and set `settings.tracing_implementation` to `\"opentelemetry\"`. #39563\n- In `DistributedTracingPolicy`, the default span name is now just the HTTP method (e.g., \"GET\", \"POST\") and no longer includes the URL path. This change was made to converge with the OpenTelemetry HTTP semantic conventions. The full URL is still included in the span attributes.\n- Renamed span attributes in `DistributedTracingPolicy`:\n - \"x-ms-client-request-id\" is now \"az.client_request_id\"\n - \"x-ms-request-id\" is now \"az.service_request_id\"\n\n### Bugs Fixed\n\n- Fixed an issue where the `traceparent` header was not being set correctly in the `DistributedTracingPolicy`. The `traceparent` header will now set based on the context of the HTTP client span. #40074\n\n### Other Changes\n\n- Added `opentelemetry-api` as an optional dependency for tracing. This can be installed with `pip install azure-core[tracing]`. #39563\n\n## 1.32.0 (2024-10-31)\n\n### Features Added\n\n- Added a default implementation to handle token challenges in `BearerTokenCredentialPolicy` and `AsyncBearerTokenCredentialPolicy`.\n\n### Bugs Fixed\n\n- Fixed an issue where the `tracing_attributes` keyword argument wasn't being handled at the request/method level. #38164\n\n### Other Changes\n\n- Log \"x-vss-e2eid\" and \"x-msedge-ref\" headers in `HttpLoggingPolicy`.\n\n## 1.31.0 (2024-09-12)\n\n### Features Added\n\n- Added azure.core.AzureClouds enum to represent the different Azure clouds.\n- Added two new credential protocol classes, `SupportsTokenInfo` and `AsyncSupportsTokenInfo`, to offer more extensibility in supporting various token acquisition scenarios. #36565\n - Each new protocol class defines a `get_token_info` method that returns an `AccessTokenInfo` object.\n- Added a new `TokenRequestOptions` class, which is a `TypedDict` with optional parameters, that can be used to define options for token requests through the `get_token_info` method. #36565\n- Added a new `AccessTokenInfo` class, which is returned by `get_token_info` implementations. This class contains the token, its expiration time, and optional additional information like when a token should be refreshed. #36565\n- `BearerTokenCredentialPolicy` and `AsyncBearerTokenCredentialPolicy` now first check if a credential has the `get_token_info` method defined. If so, the `get_token_info` method is used to acquire a token. Otherwise, the `get_token` method is used. #36565\n - These policies now also check the `refresh_on` attribute when determining if a new token request should be made.\n\n### Other Changes\n\n- The Azure Core OpenTelemetry tracing plugin will now be the preferred tracing plugin over the OpenCensus plugin. If both plugins are installed and `opentelemetry` is imported, then OpenTelemetry will be used to trace Azure SDK operations. #35050\n\n## 1.30.2 (2024-06-06)\n\n### Features Added\n\n- Tracing: `DistributedTracingPolicy` will now set an attribute, `http.request.resend_count`, on HTTP spans for resent requests to indicate the resend attempt number. #35069\n\n### Bugs Fixed\n\n- Raise correct exception if transport is used while already closed #35559\n\n### Other Changes\n\n- HTTP tracing spans will now include an `error.type` attribute if an error status code is returned. #34619\n- Minimum required Python version is now 3.8\n\n## 1.30.1 (2024-02-29)\n\n### Other Changes\n\n- Accept float for `retry_after` header. #34203\n\n## 1.30.0 (2024-02-01)\n\n### Features Added\n\n- Support tuple input for file values to `azure.core.rest.HttpRequest` #33948\n- Support tuple input to `files` with duplicate field names `azure.core.rest.HttpRequest` #34021\n\n## 1.29.7 (2024-01-18)\n\n### Other Changes\n\n- Removed dependency on `anyio`. #33282\n\n## 1.29.6 (2023-12-14)\n\n### Bugs Fixed\n\n- Adjusted `AsyncBearerTokenCredentialPolicy` to work properly with `trio` concurrency mechanisms. ([#33307](https://github.com/Azure/azure-sdk-for-python/pull/33307))\n\n### Other Changes\n\n- Added dependency on `anyio` >=3.0,<5.0\n- Bumped minimum dependency on `requests` to 2.21.0.\n\n## 1.29.5 (2023-10-19)\n\n### Bugs Fixed\n\n- Fixed an issue with `multipart/form-data` in the async transport where `data` was not getting encoded into the request body. #32473\n\n### Other Changes\n\n- Use ssl context from aiohttp by default.\n\n## 1.29.4 (2023-09-07)\n\n### Bugs Fixed\n\n- Fixed the issue that some urls trigger an infinite loop. #31346\n- Fixed issue where IndexError was raised if multipart responses did not match the number of requests. #31471\n- Fixed issue unbound variable exception if dict is invalid in CloudEvent.from_dict. #31835\n- Fixed issue asyncBearerTokenCredentialPolicy is not backward compatible with SansIOHTTPPolicy. #31836\n- Fixed issue mypy complains with new version of azure-core. #31564\n\n## 1.29.3 (2023-08-22)\n\n### Bugs Fixed\n\n- Typing fix: `message` cannot be `None` in `AzureError`. #31564\n\n## 1.29.2 (2023-08-14)\n\n### Bugs Fixed\n\n- Added a default implementation for `AsyncTokenCredential.__aexit__()` #31573\n\n### Other Changes\n\n- Bumped `typing-extensions` version to 4.6.0.\n\n## 1.29.1 (2023-08-09)\n\n### Bugs Fixed\n\n- Not pass `enabled_cae` unless it is explicitly enabled.\n\n## 1.29.0 (2023-08-03)\n\n### Features Added\n\n- A keyword argument `enable_cae` was added to the `get_token` method of the `TokenCredential` protocol. #31012\n- `BearerTokenCredentialPolicy` and `AsyncBearerTokenCredentialPolicy` now accept `enable_cae` keyword arguments in their constructors. This is used in determining if [Continuous Access Evaluation (CAE)](https://learn.microsoft.com/azure/active-directory/conditional-access/concept-continuous-access-evaluation) should be enabled for each `get_token` request. #31012\n\n## 1.28.0 (2023-07-06)\n\n### Features Added\n\n- Added header name parameter to `RequestIdPolicy`. #30772\n- Added `SensitiveHeaderCleanupPolicy` that cleans up sensitive headers if a redirect happens and the new destination is in another domain. #28349\n\n### Other Changes\n\n- Catch aiohttp errors and translate them into azure-core errors.\n\n## 1.27.1 (2023-06-13)\n\n### Bugs Fixed\n\n- Fix url building for some complex query parameters scenarios #30707\n\n## 1.27.0 (2023-06-01)\n\n### Features Added\n\n- Added support to use sync credentials in `AsyncBearerTokenCredentialPolicy`. #30381\n- Added \"prefix\" parameter to AzureKeyCredentialPolicy #29901\n\n### Bugs Fixed\n\n- Improve error message when providing the wrong credential type for AzureKeyCredential #30380\n\n## 1.26.4 (2023-04-06)\n\n### Features Added\n\n- Updated settings to include OpenTelemetry as a tracer provider. #29095\n\n### Other Changes\n\n- Improved typing\n\n## 1.26.3 (2023-02-02)\n\n### Bugs Fixed\n\n- Fixed deflate decompression for aiohttp #28483\n\n## 1.26.2 (2023-01-05)\n\n### Bugs Fixed\n\n- Fix 'ClientSession' object has no attribute 'auto_decompress' (thanks to @mghextreme for the contribution)\n\n### Other Changes\n\n- Add \"x-ms-error-code\" as secure header to log\n- Rename \"DEFAULT_HEADERS_WHITELIST\" to \"DEFAULT_HEADERS_ALLOWLIST\". Added a backward compatible alias.\n\n## 1.26.1 (2022-11-03)\n\n### Other Changes\n\n- Added example of RequestsTransport with custom session. (thanks to @inirudebwoy for the contribution) #26768\n- Added Python 3.11 support.\n\n## 1.26.0 (2022-10-06)\n\n### Other Changes\n\n- LRO polling will not wait anymore before doing the first status check #26376\n- Added extra dependency for [aio]. pip install azure-core[aio] installs aiohttp too.\n\n## 1.25.1 (2022-09-01)\n\n### Bugs Fixed\n\n- Added @runtime_checkable to `TokenCredential` protocol definitions #25187\n\n## 1.25.0 (2022-08-04)\n\nAzure-core is supported on Python 3.7 or later. For more details, please read our page on [Azure SDK for Python version support policy](https://github.com/Azure/azure-sdk-for-python/wiki/Azure-SDKs-Python-version-support-policy).\n\n### Features Added\n\n- Added `CaseInsensitiveDict` implementation in `azure.core.utils` removing dependency on `requests` and `aiohttp`\n\n## 1.24.2 (2022-06-30)\n\n### Bugs Fixed\n\n- Fixed the bug that azure-core could not be imported under Python 3.11.0b3 #24928\n- `ContentDecodePolicy` can now correctly deserialize more JSON bodies with different mime types #22410\n\n## 1.24.1 (2022-06-01)\n\n### Bugs Fixed\n\n- Declare method level span as INTERNAL by default #24492\n- Fixed type hints for `azure.core.paging.ItemPaged` #24548\n\n## 1.24.0 (2022-05-06)\n\n### Features Added\n\n- Add `SerializationError` and `DeserializationError` in `azure.core.exceptions` for errors raised during serialization / deserialization #24312\n\n## 1.23.1 (2022-03-31)\n\n### Bugs Fixed\n\n- Allow stream inputs to the `content` kwarg of `azure.core.rest.HttpRequest` from objects with a `read` method #23578\n\n## 1.23.0 (2022-03-03)\n\n### Features Added\n\n- Improve intellisense type hinting for service client methods. #22891\n\n- Add a case insensitive dict `case_insensitive_dict` in `azure.core.utils`. #23206\n\n### Bugs Fixed\n\n- Use \"\\n\" rather than \"/n\" for new line in log. #23261\n\n### Other Changes\n\n- Log \"WWW-Authenticate\" header in `HttpLoggingPolicy` #22990\n- Added dependency on `typing-extensions` >= 4.0.1\n\n## 1.22.1 (2022-02-09)\n\n### Bugs Fixed\n\n- Limiting `final-state-via` scope to POST until consuming SDKs has been fixed to use this option properly on PUT. #22989\n\n## 1.22.0 (2022-02-03)\n_[**This version is deprecated.**]_\n\n### Features Added\n\n- Add support for `final-state-via` LRO option in core. #22713\n\n### Bugs Fixed\n\n- Add response body to string representation of `HttpResponseError` if we're not able to parse out information #22302\n- Raise `AttributeError` when calling azure.core.pipeline.transport.\\_\\_bases__ #22469\n\n### Other Changes\n\n- Python 2.7 is no longer supported. Please use Python version 3.6 or later.\n\n## 1.21.1 (2021-12-06)\n\n### Other Changes\n\n- Revert change in str method #22023\n\n## 1.21.0 (2021-12-02)\n\n### Breaking Changes\n\n- Sync stream downloading now raises `azure.core.exceptions.DecodeError` rather than `requests.exceptions.ContentDecodingError`\n\n### Bugs Fixed\n\n- Add response body to string representation of `HttpResponseError` if we're not able to parse out information #21800\n\n## 1.20.1 (2021-11-08)\n\n### Bugs Fixed\n\n- Correctly set response's content to decompressed body when users are using aiohttp transport with decompression headers #21620\n\n## 1.20.0 (2021-11-04)\n\n### Features Added\n\n- GA `send_request` onto the `azure.core.PipelineClient` and `azure.core.AsyncPipelineClient`. This method takes in\nrequests and sends them through our pipelines.\n- GA `azure.core.rest`. `azure.core.rest` is our new public simple HTTP library in `azure.core` that users will use to create requests, and consume responses.\n- GA errors `StreamConsumedError`, `StreamClosedError`, and `ResponseNotReadError` to `azure.core.exceptions`. These errors\nare thrown if you mishandle streamed responses from the `azure.core.rest` module\n- add kwargs to the methods for `iter_raw` and `iter_bytes` #21529\n- no longer raise JSON errors if users pass in file descriptors of JSON to the `json` kwarg in `HttpRequest` #21504\n- Added new error type `IncompleteReadError` which is raised if peer closes the connection before we have received the complete message body.\n\n### Breaking Changes\n\n- SansIOHTTPPolicy.on_exception returns None instead of bool.\n\n### Bugs Fixed\n\n- The `Content-Length` header in a http response is strictly checked against the actual number of bytes in the body,\n rather than silently truncating data in case the underlying tcp connection is closed prematurely.\n (thanks to @jochen-ott-by for the contribution) #20412\n- UnboundLocalError when SansIOHTTPPolicy handles an exception #15222\n- Add default content type header of `text/plain` and content length header for users who pass unicode strings to the `content` kwarg of `HttpRequest` in 2.7 #21550\n\n## 1.19.1 (2021-11-01)\n\n### Bugs Fixed\n\n- respect text encoding specified in argument (thanks to @ryohji for the contribution) #20796\n- Fix \"coroutine x.read() was never awaited\" warning from `ContentDecodePolicy` #21318\n- fix type check for `data` input to `azure.core.rest` for python 2.7 users #21341\n- use `charset_normalizer` if `chardet` is not installed to migrate aiohttp 3.8.0 changes.\n\n### Other Changes\n\n- Refactor AzureJSONEncoder (thanks to @Codejune for the contribution) #21028\n\n## 1.19.0 (2021-09-30)\n\n### Breaking Changes in the Provisional `azure.core.rest` package\n\n- `azure.core.rest.HttpResponse` and `azure.core.rest.AsyncHttpResponse` are now abstract base classes. They should not be initialized directly, instead\nyour transport responses should inherit from them and implement them.\n- The properties of the `azure.core.rest` responses are now all read-only\n\n- HttpLoggingPolicy integrates logs into one record #19925\n\n## 1.18.0 (2021-09-02)\n\n### Features Added\n\n- `azure.core.serialization.AzureJSONEncoder` (introduced in 1.17.0) serializes `datetime.datetime` objects in ISO 8601 format, conforming to RFC 3339's specification. #20190\n- We now use `azure.core.serialization.AzureJSONEncoder` to serialize `json` input to `azure.core.rest.HttpRequest`.\n\n### Breaking Changes in the Provisional `azure.core.rest` package\n\n- The `text` property on `azure.core.rest.HttpResponse` and `azure.core.rest.AsyncHttpResponse` has changed to a method, which also takes\nan `encoding` parameter.\n- Removed `iter_text` and `iter_lines` from `azure.core.rest.HttpResponse` and `azure.core.rest.AsyncHttpResponse`\n\n### Bugs Fixed\n\n- The behaviour of the headers returned in `azure.core.rest` responses now aligns across sync and async. Items can now be checked case-insensitively and without raising an error for format.\n\n## 1.17.0 (2021-08-05)\n\n### Features Added\n\n- Cut hard dependency on requests library\n- Added a `from_json` method which now accepts storage QueueMessage, eventhub's EventData or ServiceBusMessage or simply json bytes to return a `CloudEvent`\n\n### Fixed\n\n- Not override \"x-ms-client-request-id\" if it already exists in the header. #17757\n\n### Breaking Changes in the Provisional `azure.core.rest` package\n\n- `azure.core.rest` will not try to guess the `charset` anymore if it was impossible to extract it from `HttpResponse` analysis. This removes our dependency on `charset`.\n\n## 1.16.0 (2021-07-01)\n\n### Features Added\n\n- Add new ***provisional*** methods `send_request` onto the `azure.core.PipelineClient` and `azure.core.AsyncPipelineClient`. This method takes in\nrequests and sends them through our pipelines.\n- Add new ***provisional*** module `azure.core.rest`. `azure.core.rest` is our new public simple HTTP library in `azure.core` that users will use to create requests, and consume responses.\n- Add new ***provisional*** errors `StreamConsumedError`, `StreamClosedError`, and `ResponseNotReadError` to `azure.core.exceptions`. These errors\nare thrown if you mishandle streamed responses from the provisional `azure.core.rest` module\n\n### Fixed\n\n- Improved error message in the `from_dict` method of `CloudEvent` when a wrong schema is sent.\n\n## 1.15.0 (2021-06-04)\n\n### New Features\n\n- Added `BearerTokenCredentialPolicy.on_challenge` and `.authorize_request` to allow subclasses to optionally handle authentication challenges\n\n### Bug Fixes\n\n- Retry policies don't sleep after operations time out\n- The `from_dict` methhod in the `CloudEvent` can now convert a datetime string to datetime object when microsecond exceeds the python limitation\n\n## 1.14.0 (2021-05-13)\n\n### New Features\n\n- Added `azure.core.credentials.AzureNamedKeyCredential` credential #17548.\n- Added `decompress` parameter for `stream_download` method. If it is set to `False`, will not do decompression upon the stream. #17920\n\n## 1.13.0 (2021-04-02)\n\nAzure core requires Python 2.7 or Python 3.6+ since this release.\n\n### New Features\n\n- Added `azure.core.utils.parse_connection_string` function to parse connection strings across SDKs, with common validation and support for case insensitive keys.\n- Supported adding custom policies #16519\n- Added `~azure.core.tracing.Link` that should be used while passing `Links` to `AbstractSpan`.\n- `AbstractSpan` constructor can now take in additional keyword only args.\n\n### Bug fixes\n\n- Make NetworkTraceLoggingPolicy show the auth token in plain text. #14191\n- Fixed RetryPolicy overriding default connection timeout with an extreme value #17481\n\n## 1.12.0 (2021-03-08)\n\nThis version will be the last version to officially support Python 3.5, future versions will require Python 2.7 or Python 3.6+.\n\n### Features\n\n- Added `azure.core.messaging.CloudEvent` model that follows the cloud event spec.\n- Added `azure.core.serialization.NULL` sentinel value\n- Improve `repr`s for `HttpRequest` and `HttpResponse`s #16972\n\n### Bug Fixes\n\n- Disable retry in stream downloading. (thanks to @jochen-ott-by @hoffmann for the contribution) #16723\n\n## 1.11.0 (2021-02-08)\n\n### Features\n\n- Added `CaseInsensitiveEnumMeta` class for case-insensitive enums. #16316\n- Add `raise_for_status` method onto `HttpResponse`. Calling `response.raise_for_status()` on a response with an error code\nwill raise an `HttpResponseError`. Calling it on a good response will do nothing #16399\n\n### Bug Fixes\n\n- Update conn.conn_kw rather than overriding it when setting block size. (thanks for @jiasli for the contribution) #16587\n\n## 1.10.0 (2021-01-11)\n\n### Features\n\n- Added `AzureSasCredential` and its respective policy. #15946\n\n## 1.9.0 (2020-11-09)\n\n### Features\n\n- Add a `continuation_token` attribute to the base `AzureError` exception, and set this value for errors raised\n during paged or long-running operations.\n\n### Bug Fixes\n\n- Set retry_interval to 1 second instead of 1000 seconds (thanks **vbarbaresi** for contributing) #14357\n\n\n## 1.8.2 (2020-10-05)\n\n### Bug Fixes\n\n- Fixed bug to allow polling in the case of parameterized endpoints with relative polling urls #14097\n\n\n## 1.8.1 (2020-09-08)\n\n### Bug fixes\n\n- SAS credential replicated \"/\" fix #13159\n\n## 1.8.0 (2020-08-10)\n\n### Features\n\n- Support params as list for exploding parameters #12410\n\n\n## 1.7.0 (2020-07-06)\n\n### Bug fixes\n\n- `AzureKeyCredentialPolicy` will now accept (and ignore) passed in kwargs #11963\n- Better error messages if passed endpoint is incorrect #12106\n- Do not JSON encore a string if content type is \"text\" #12137\n\n### Features\n\n- Added `http_logging_policy` property on the `Configuration` object, allowing users to individually\nset the http logging policy of the config #12218\n\n## 1.6.0 (2020-06-03)\n\n### Bug fixes\n\n- Fixed deadlocks in AsyncBearerTokenCredentialPolicy #11543\n- Fix AttributeException in StreamDownloadGenerator #11462\n\n### Features\n\n- Added support for changesets as part of multipart message support #10485\n- Add AsyncLROPoller in azure.core.polling #10801\n- Add get_continuation_token/from_continuation_token/polling_method methods in pollers (sync and async) #10801\n- HttpResponse and PipelineContext objects are now pickable #10801\n\n## 1.5.0 (2020-05-04)\n\n### Features\n\n- Support \"x-ms-retry-after-ms\" in response header #10743\n- `link` and `link_from_headers` now accepts attributes #10765\n\n### Bug fixes\n\n- Not retry if the status code is less than 400 #10778\n- \"x-ms-request-id\" is not considered safe header for logging #10967\n\n## 1.4.0 (2020-04-06)\n\n### Features\n\n- Support a default error type in map_error #9773\n- Added `AzureKeyCredential` and its respective policy. #10509\n- Added `azure.core.polling.base_polling` module with a \"Microsoft One API\" polling implementation #10090\n Also contains the async version in `azure.core.polling.async_base_polling`\n- Support kwarg `enforce_https` to disable HTTPS check on authentication #9821\n- Support additional kwargs in `HttpRequest.set_multipart_mixed` that will be passed into pipeline context.\n\n## 1.3.0 (2020-03-09)\n\n### Bug fixes\n\n- Appended RequestIdPolicy to the default pipeline #9841\n- Rewind the body position in async_retry #10117\n\n### Features\n\n- Add raw_request_hook support in custom_hook_policy #9958\n- Add timeout support in retry_policy #10011\n- Add OdataV4 error format auto-parsing in all exceptions ('error' attribute) #9738\n\n## 1.2.2 (2020-02-10)\n\n### Bug fixes\n\n- Fixed a bug that sends None as request_id #9545\n- Enable mypy for customers #9572\n- Handle TypeError in deep copy #9620\n- Fix text/plain content-type in decoder #9589\n\n## 1.2.1 (2020-01-14)\n\n### Bug fixes\n\n- Fixed a regression in 1.2.0 that was incompatible with azure-keyvault-* 4.0.0\n[#9462](https://github.com/Azure/azure-sdk-for-python/issues/9462)\n\n\n## 1.2.0 (2020-01-14)\n\n### Features\n\n- Add user_agent & sdk_moniker kwargs in UserAgentPolicy init #9355\n- Support OPTIONS HTTP verb #9322\n- Add tracing_attributes to tracing decorator #9297\n- Support auto_request_id in RequestIdPolicy #9163\n- Support fixed retry #6419\n- Support \"retry-after-ms\" in response header #9240\n\n### Bug fixes\n\n- Removed `__enter__` and `__exit__` from async context managers #9313\n\n## 1.1.1 (2019-12-03)\n\n### Bug fixes\n\n- Bearer token authorization requires HTTPS\n- Rewind the body position in retry #8307\n\n## 1.1.0 (2019-11-25)\n\n### Features\n\n- New RequestIdPolicy #8437\n- Enable logging policy in default pipeline #8053\n- Normalize transport timeout. #8000\n Now we have:\n * 'connection_timeout' - a single float in seconds for the connection timeout. Default 5min\n * 'read_timeout' - a single float in seconds for the read timeout. Default 5min\n\n### Bug fixes\n\n- RequestHistory: deepcopy fails if request contains a stream #7732\n- Retry: retry raises error if response does not have http_response #8629\n- Client kwargs are now passed to DistributedTracingPolicy correctly #8051\n- NetworkLoggingPolicy now logs correctly all requests in case of retry #8262\n\n## 1.0.0 (2019-10-29)\n\n### Features\n\n- Tracing: DistributedTracingPolicy now accepts kwargs network_span_namer to change network span name #7773\n- Tracing: Implementation of AbstractSpan can now use the mixin HttpSpanMixin to get HTTP span update automatically #7773\n- Tracing: AbstractSpan contract \"change_context\" introduced #7773\n- Introduce new policy HttpLoggingPolicy #7988\n\n### Bug fixes\n\n- Fix AsyncioRequestsTransport if input stream is an async generator #7743\n- Fix form-data with aiohttp transport #7749\n\n### Breaking changes\n\n- Tracing: AbstractSpan.set_current_span is longer supported. Use change_context instead. #7773\n- azure.core.pipeline.policies.ContentDecodePolicy.deserialize_from_text changed\n\n## 1.0.0b4 (2019-10-07)\n\n### Features\n\n- Tracing: network span context is available with the TRACING_CONTEXT in pipeline response #7252\n- Tracing: Span contract now has `kind`, `traceparent` and is a context manager #7252\n- SansIOHTTPPolicy methods can now be coroutines #7497\n- Add multipart/mixed support #7083:\n\n - HttpRequest now has a \"set_multipart_mixed\" method to set the parts of this request\n - HttpRequest now has a \"prepare_multipart_body\" method to build final body.\n - HttpResponse now has a \"parts\" method to return an iterator of parts\n - AsyncHttpResponse now has a \"parts\" methods to return an async iterator of parts\n - Note that multipart/mixed is a Python 3.x only feature\n\n### Bug fixes\n\n- Tracing: policy cannot fail the pipeline, even in the worst condition #7252\n- Tracing: policy pass correctly status message if exception #7252\n- Tracing: incorrect span if exception raised from decorated function #7133\n- Fixed urllib3 ConnectTimeoutError being raised by Requests during a socket timeout. Now this exception is caught and wrapped as a `ServiceRequestError` #7542\n\n### Breaking changes\n\n- Tracing: `azure.core.tracing.context` removed\n- Tracing: `azure.core.tracing.context.tracing_context.with_current_context` renamed to `azure.core.tracing.common.with_current_context` #7252\n- Tracing: `link` renamed `link_from_headers` and `link` takes now a string\n- Tracing: opencensus implementation has been moved to the package `azure-core-tracing-opencensus`\n- Some modules and classes that were importables from several different places have been removed:\n\n - `azure.core.HttpResponseError` is now only `azure.core.exceptions.HttpResponseError`\n - `azure.core.Configuration` is now only `azure.core.configuration.Configuration`\n - `azure.core.HttpRequest` is now only `azure.core.pipeline.transport.HttpRequest`\n - `azure.core.version` module has been removed. Use `azure.core.__version__` to get version number.\n - `azure.core.pipeline_client` has been removed. Import from `azure.core` instead.\n - `azure.core.pipeline_client_async` has been removed. Import from `azure.core` instead.\n - `azure.core.pipeline.base` has been removed. Import from `azure.core.pipeline` instead.\n - `azure.core.pipeline.base_async` has been removed. Import from `azure.core.pipeline` instead.\n - `azure.core.pipeline.policies.base` has been removed. Import from `azure.core.pipeline.policies` instead.\n - `azure.core.pipeline.policies.base_async` has been removed. Import from `azure.core.pipeline.policies` instead.\n - `azure.core.pipeline.policies.authentication` has been removed. Import from `azure.core.pipeline.policies` instead.\n - `azure.core.pipeline.policies.authentication_async` has been removed. Import from `azure.core.pipeline.policies` instead.\n - `azure.core.pipeline.policies.custom_hook` has been removed. Import from `azure.core.pipeline.policies` instead.\n - `azure.core.pipeline.policies.redirect` has been removed. Import from `azure.core.pipeline.policies` instead.\n - `azure.core.pipeline.policies.redirect_async` has been removed. Import from `azure.core.pipeline.policies` instead.\n - `azure.core.pipeline.policies.retry` has been removed. Import from `azure.core.pipeline.policies` instead.\n - `azure.core.pipeline.policies.retry_async` has been removed. Import from `azure.core.pipeline.policies` instead.\n - `azure.core.pipeline.policies.distributed_tracing` has been removed. Import from `azure.core.pipeline.policies` instead.\n - `azure.core.pipeline.policies.universal` has been removed. Import from `azure.core.pipeline.policies` instead.\n - `azure.core.tracing.abstract_span` has been removed. Import from `azure.core.tracing` instead.\n - `azure.core.pipeline.transport.base` has been removed. Import from `azure.core.pipeline.transport` instead.\n - `azure.core.pipeline.transport.base_async` has been removed. Import from `azure.core.pipeline.transport` instead.\n - `azure.core.pipeline.transport.requests_basic` has been removed. Import from `azure.core.pipeline.transport` instead.\n - `azure.core.pipeline.transport.requests_asyncio` has been removed. Import from `azure.core.pipeline.transport` instead.\n - `azure.core.pipeline.transport.requests_trio` has been removed. Import from `azure.core.pipeline.transport` instead.\n - `azure.core.pipeline.transport.aiohttp` has been removed. Import from `azure.core.pipeline.transport` instead.\n - `azure.core.polling.poller` has been removed. Import from `azure.core.polling` instead.\n - `azure.core.polling.async_poller` has been removed. Import from `azure.core.polling` instead.\n\n## 1.0.0b3 (2019-09-09)\n\n### Bug fixes\n\n- Fix aiohttp auto-headers #6992\n- Add tracing to policies module init #6951\n\n## 1.0.0b2 (2019-08-05)\n\n### Breaking changes\n\n- Transport classes don't take `config` parameter anymore (use kwargs instead) #6372\n- `azure.core.paging` has been completely refactored #6420\n- HttpResponse.content_type attribute is now a string (was a list) #6490\n- For `StreamDownloadGenerator` subclasses, `response` is now an `HttpResponse`, and not a transport response like `aiohttp.ClientResponse` or `requests.Response`. The transport response is available in `internal_response` attribute #6490\n\n### Bug fixes\n\n- aiohttp is not required to import async pipelines classes #6496\n- `AsyncioRequestsTransport.sleep` is now a coroutine as expected #6490\n- `RequestsTransport` is not tight to `ProxyPolicy` implementation details anymore #6372\n- `AiohttpTransport` does not raise on unexpected kwargs #6355\n\n### Features\n\n- New paging base classes that support `continuation_token` and `by_page()` #6420\n- Proxy support for `AiohttpTransport` #6372\n\n## 1.0.0b1 (2019-06-26)\n\n- Preview 1 release", + "release_date": "2025-05-01T23:17:29", "parties": [ { "type": "person", @@ -152,15 +152,14 @@ "Programming Language :: Python :: 3.11", "Programming Language :: Python :: 3.12", "Programming Language :: Python :: 3.13", - "Programming Language :: Python :: 3.8", "Programming Language :: Python :: 3.9" ], "homepage_url": "https://github.com/Azure/azure-sdk-for-python/tree/main/sdk/core/azure-core", - "download_url": "https://files.pythonhosted.org/packages/07/b7/76b7e144aa53bd206bf1ce34fa75350472c3f69bf30e5c8c18bc9881035d/azure_core-1.33.0-py3-none-any.whl", - "size": 207071, + "download_url": "https://files.pythonhosted.org/packages/84/9e/5c87b49f65bb16571599bc789857d0ded2f53014d3392bc88a5d1f3ad779/azure_core-1.34.0-py3-none-any.whl", + "size": 207409, "sha1": null, - "md5": "ddd17415a5573ca4c435d7db0ff5631f", - "sha256": "9b5b6d0223a1d38c37500e6971118c1e0f13f54951e6893968b38910bc9cda8f", + "md5": "b03f983d92f026edffb1a5c50ee5a64b", + "sha256": "0615d3b756beccdb6624d1c0ae97284f38b78fb59a2a9839bf927c66fbbdddd6", "sha512": null, "bug_tracking_url": null, "code_view_url": null, @@ -180,9 +179,9 @@ "dependencies": [], "repository_homepage_url": null, "repository_download_url": null, - "api_data_url": "https://pypi.org/pypi/azure-core/1.33.0/json", + "api_data_url": "https://pypi.org/pypi/azure-core/1.34.0/json", "datasource_id": null, - "purl": "pkg:pypi/azure-core@1.33.0" + "purl": "pkg:pypi/azure-core@1.34.0" }, { "type": "pypi", @@ -441,12 +440,12 @@ "type": "pypi", "namespace": null, "name": "charset-normalizer", - "version": "3.4.1", + "version": "3.4.2", "qualifiers": {}, "subpath": null, "primary_language": "Python", - "description": "The Real First Universal Charset Detector. Open, modern and actively maintained alternative to Chardet.\n

Charset Detection, for Everyone \ud83d\udc4b

\n\n

\n The Real First Universal Charset Detector
\n \n \n \n \n \"Download\n \n \n \n \n

\n

\n Featured Packages
\n \n \"Static\n \n \n \"Static\n \n

\n

\n In other language (unofficial port - by the community)
\n \n \"Static\n \n

\n\n> A library that helps you read text from an unknown charset encoding.
Motivated by `chardet`,\n> I'm trying to resolve the issue by taking a new approach.\n> All IANA character set names for which the Python core library provides codecs are supported.\n\n

\n >>>>> \ud83d\udc49 Try Me Online Now, Then Adopt Me \ud83d\udc48 <<<<<\n

\n\nThis project offers you an alternative to **Universal Charset Encoding Detector**, also known as **Chardet**.\n\n| Feature | [Chardet](https://github.com/chardet/chardet) | Charset Normalizer | [cChardet](https://github.com/PyYoshi/cChardet) |\n|--------------------------------------------------|:---------------------------------------------:|:--------------------------------------------------------------------------------------------------:|:-----------------------------------------------:|\n| `Fast` | \u274c | \u2705 | \u2705 |\n| `Universal**` | \u274c | \u2705 | \u274c |\n| `Reliable` **without** distinguishable standards | \u274c | \u2705 | \u2705 |\n| `Reliable` **with** distinguishable standards | \u2705 | \u2705 | \u2705 |\n| `License` | LGPL-2.1
_restrictive_ | MIT | MPL-1.1
_restrictive_ |\n| `Native Python` | \u2705 | \u2705 | \u274c |\n| `Detect spoken language` | \u274c | \u2705 | N/A |\n| `UnicodeDecodeError Safety` | \u274c | \u2705 | \u274c |\n| `Whl Size (min)` | 193.6 kB | 42 kB | ~200 kB |\n| `Supported Encoding` | 33 | \ud83c\udf89 [99](https://charset-normalizer.readthedocs.io/en/latest/user/support.html#supported-encodings) | 40 |\n\n

\n\"Reading\"Cat\n

\n\n*\\*\\* : They are clearly using specific code for a specific encoding even if covering most of used one*
\n\n## \u26a1 Performance\n\nThis package offer better performance than its counterpart Chardet. Here are some numbers.\n\n| Package | Accuracy | Mean per file (ms) | File per sec (est) |\n|-----------------------------------------------|:--------:|:------------------:|:------------------:|\n| [chardet](https://github.com/chardet/chardet) | 86 % | 63 ms | 16 file/sec |\n| charset-normalizer | **98 %** | **10 ms** | 100 file/sec |\n\n| Package | 99th percentile | 95th percentile | 50th percentile |\n|-----------------------------------------------|:---------------:|:---------------:|:---------------:|\n| [chardet](https://github.com/chardet/chardet) | 265 ms | 71 ms | 7 ms |\n| charset-normalizer | 100 ms | 50 ms | 5 ms |\n\n_updated as of december 2024 using CPython 3.12_\n\nChardet's performance on larger file (1MB+) are very poor. Expect huge difference on large payload.\n\n> Stats are generated using 400+ files using default parameters. More details on used files, see GHA workflows.\n> And yes, these results might change at any time. The dataset can be updated to include more files.\n> The actual delays heavily depends on your CPU capabilities. The factors should remain the same.\n> Keep in mind that the stats are generous and that Chardet accuracy vs our is measured using Chardet initial capability\n> (e.g. Supported Encoding) Challenge-them if you want.\n\n## \u2728 Installation\n\nUsing pip:\n\n```sh\npip install charset-normalizer -U\n```\n\n## \ud83d\ude80 Basic Usage\n\n### CLI\nThis package comes with a CLI.\n\n```\nusage: normalizer [-h] [-v] [-a] [-n] [-m] [-r] [-f] [-t THRESHOLD]\n file [file ...]\n\nThe Real First Universal Charset Detector. Discover originating encoding used\non text file. Normalize text to unicode.\n\npositional arguments:\n files File(s) to be analysed\n\noptional arguments:\n -h, --help show this help message and exit\n -v, --verbose Display complementary information about file if any.\n Stdout will contain logs about the detection process.\n -a, --with-alternative\n Output complementary possibilities if any. Top-level\n JSON WILL be a list.\n -n, --normalize Permit to normalize input file. If not set, program\n does not write anything.\n -m, --minimal Only output the charset detected to STDOUT. Disabling\n JSON output.\n -r, --replace Replace file when trying to normalize it instead of\n creating a new one.\n -f, --force Replace file without asking if you are sure, use this\n flag with caution.\n -t THRESHOLD, --threshold THRESHOLD\n Define a custom maximum amount of chaos allowed in\n decoded content. 0. <= chaos <= 1.\n --version Show version information and exit.\n```\n\n```bash\nnormalizer ./data/sample.1.fr.srt\n```\n\nor\n\n```bash\npython -m charset_normalizer ./data/sample.1.fr.srt\n```\n\n\ud83c\udf89 Since version 1.4.0 the CLI produce easily usable stdout result in JSON format.\n\n```json\n{\n \"path\": \"/home/default/projects/charset_normalizer/data/sample.1.fr.srt\",\n \"encoding\": \"cp1252\",\n \"encoding_aliases\": [\n \"1252\",\n \"windows_1252\"\n ],\n \"alternative_encodings\": [\n \"cp1254\",\n \"cp1256\",\n \"cp1258\",\n \"iso8859_14\",\n \"iso8859_15\",\n \"iso8859_16\",\n \"iso8859_3\",\n \"iso8859_9\",\n \"latin_1\",\n \"mbcs\"\n ],\n \"language\": \"French\",\n \"alphabets\": [\n \"Basic Latin\",\n \"Latin-1 Supplement\"\n ],\n \"has_sig_or_bom\": false,\n \"chaos\": 0.149,\n \"coherence\": 97.152,\n \"unicode_path\": null,\n \"is_preferred\": true\n}\n```\n\n### Python\n*Just print out normalized text*\n```python\nfrom charset_normalizer import from_path\n\nresults = from_path('./my_subtitle.srt')\n\nprint(str(results.best()))\n```\n\n*Upgrade your code without effort*\n```python\nfrom charset_normalizer import detect\n```\n\nThe above code will behave the same as **chardet**. We ensure that we offer the best (reasonable) BC result possible.\n\nSee the docs for advanced usage : [readthedocs.io](https://charset-normalizer.readthedocs.io/en/latest/)\n\n## \ud83d\ude07 Why\n\nWhen I started using Chardet, I noticed that it was not suited to my expectations, and I wanted to propose a\nreliable alternative using a completely different method. Also! I never back down on a good challenge!\n\nI **don't care** about the **originating charset** encoding, because **two different tables** can\nproduce **two identical rendered string.**\nWhat I want is to get readable text, the best I can.\n\nIn a way, **I'm brute forcing text decoding.** How cool is that ? \ud83d\ude0e\n\nDon't confuse package **ftfy** with charset-normalizer or chardet. ftfy goal is to repair Unicode string whereas charset-normalizer to convert raw file in unknown encoding to unicode.\n\n## \ud83c\udf70 How\n\n - Discard all charset encoding table that could not fit the binary content.\n - Measure noise, or the mess once opened (by chunks) with a corresponding charset encoding.\n - Extract matches with the lowest mess detected.\n - Additionally, we measure coherence / probe for a language.\n\n**Wait a minute**, what is noise/mess and coherence according to **YOU ?**\n\n*Noise :* I opened hundred of text files, **written by humans**, with the wrong encoding table. **I observed**, then\n**I established** some ground rules about **what is obvious** when **it seems like** a mess (aka. defining noise in rendered text).\n I know that my interpretation of what is noise is probably incomplete, feel free to contribute in order to\n improve or rewrite it.\n\n*Coherence :* For each language there is on earth, we have computed ranked letter appearance occurrences (the best we can). So I thought\nthat intel is worth something here. So I use those records against decoded text to check if I can detect intelligent design.\n\n## \u26a1 Known limitations\n\n - Language detection is unreliable when text contains two or more languages sharing identical letters. (eg. HTML (english tags) + Turkish content (Sharing Latin characters))\n - Every charset detector heavily depends on sufficient content. In common cases, do not bother run detection on very tiny content.\n\n## \u26a0\ufe0f About Python EOLs\n\n**If you are running:**\n\n- Python >=2.7,<3.5: Unsupported\n- Python 3.5: charset-normalizer < 2.1\n- Python 3.6: charset-normalizer < 3.1\n- Python 3.7: charset-normalizer < 4.0\n\nUpgrade your Python interpreter as soon as possible.\n\n## \ud83d\udc64 Contributing\n\nContributions, issues and feature requests are very much welcome.
\nFeel free to check [issues page](https://github.com/ousret/charset_normalizer/issues) if you want to contribute.\n\n## \ud83d\udcdd License\n\nCopyright \u00a9 [Ahmed TAHRI @Ousret](https://github.com/Ousret).
\nThis project is [MIT](https://github.com/Ousret/charset_normalizer/blob/master/LICENSE) licensed.\n\nCharacters frequencies used in this project \u00a9 2012 [Denny Vrande\u010di\u0107](http://simia.net/letters/)\n\n## \ud83d\udcbc For Enterprise\n\nProfessional support for charset-normalizer is available as part of the [Tidelift\nSubscription][1]. Tidelift gives software development teams a single source for\npurchasing and maintaining their software, with professional grade assurances\nfrom the experts who know it best, while seamlessly integrating with existing\ntools.\n\n[1]: https://tidelift.com/subscription/pkg/pypi-charset-normalizer?utm_source=pypi-charset-normalizer&utm_medium=readme\n\n[![OpenSSF Best Practices](https://www.bestpractices.dev/projects/7297/badge)](https://www.bestpractices.dev/projects/7297)\n\n# Changelog\nAll notable changes to charset-normalizer will be documented in this file. This project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html).\nThe format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/).\n\n## [3.4.1](https://github.com/Ousret/charset_normalizer/compare/3.4.0...3.4.1) (2024-12-24)\n\n### Changed\n- Project metadata are now stored using `pyproject.toml` instead of `setup.cfg` using setuptools as the build backend.\n- Enforce annotation delayed loading for a simpler and consistent types in the project.\n- Optional mypyc compilation upgraded to version 1.14 for Python >= 3.8\n\n### Added\n- pre-commit configuration.\n- noxfile.\n\n### Removed\n- `build-requirements.txt` as per using `pyproject.toml` native build configuration.\n- `bin/integration.py` and `bin/serve.py` in favor of downstream integration test (see noxfile).\n- `setup.cfg` in favor of `pyproject.toml` metadata configuration.\n- Unused `utils.range_scan` function.\n\n### Fixed\n- Converting content to Unicode bytes may insert `utf_8` instead of preferred `utf-8`. (#572)\n- Deprecation warning \"'count' is passed as positional argument\" when converting to Unicode bytes on Python 3.13+\n\n## [3.4.0](https://github.com/Ousret/charset_normalizer/compare/3.3.2...3.4.0) (2024-10-08)\n\n### Added\n- Argument `--no-preemptive` in the CLI to prevent the detector to search for hints.\n- Support for Python 3.13 (#512)\n\n### Fixed\n- Relax the TypeError exception thrown when trying to compare a CharsetMatch with anything else than a CharsetMatch.\n- Improved the general reliability of the detector based on user feedbacks. (#520) (#509) (#498) (#407) (#537)\n- Declared charset in content (preemptive detection) not changed when converting to utf-8 bytes. (#381)\n\n## [3.3.2](https://github.com/Ousret/charset_normalizer/compare/3.3.1...3.3.2) (2023-10-31)\n\n### Fixed\n- Unintentional memory usage regression when using large payload that match several encoding (#376)\n- Regression on some detection case showcased in the documentation (#371)\n\n### Added\n- Noise (md) probe that identify malformed arabic representation due to the presence of letters in isolated form (credit to my wife)\n\n## [3.3.1](https://github.com/Ousret/charset_normalizer/compare/3.3.0...3.3.1) (2023-10-22)\n\n### Changed\n- Optional mypyc compilation upgraded to version 1.6.1 for Python >= 3.8\n- Improved the general detection reliability based on reports from the community\n\n## [3.3.0](https://github.com/Ousret/charset_normalizer/compare/3.2.0...3.3.0) (2023-09-30)\n\n### Added\n- Allow to execute the CLI (e.g. normalizer) through `python -m charset_normalizer.cli` or `python -m charset_normalizer`\n- Support for 9 forgotten encoding that are supported by Python but unlisted in `encoding.aliases` as they have no alias (#323)\n\n### Removed\n- (internal) Redundant utils.is_ascii function and unused function is_private_use_only\n- (internal) charset_normalizer.assets is moved inside charset_normalizer.constant\n\n### Changed\n- (internal) Unicode code blocks in constants are updated using the latest v15.0.0 definition to improve detection\n- Optional mypyc compilation upgraded to version 1.5.1 for Python >= 3.8\n\n### Fixed\n- Unable to properly sort CharsetMatch when both chaos/noise and coherence were close due to an unreachable condition in \\_\\_lt\\_\\_ (#350)\n\n## [3.2.0](https://github.com/Ousret/charset_normalizer/compare/3.1.0...3.2.0) (2023-06-07)\n\n### Changed\n- Typehint for function `from_path` no longer enforce `PathLike` as its first argument\n- Minor improvement over the global detection reliability\n\n### Added\n- Introduce function `is_binary` that relies on main capabilities, and optimized to detect binaries\n- Propagate `enable_fallback` argument throughout `from_bytes`, `from_path`, and `from_fp` that allow a deeper control over the detection (default True)\n- Explicit support for Python 3.12\n\n### Fixed\n- Edge case detection failure where a file would contain 'very-long' camel cased word (Issue #289)\n\n## [3.1.0](https://github.com/Ousret/charset_normalizer/compare/3.0.1...3.1.0) (2023-03-06)\n\n### Added\n- Argument `should_rename_legacy` for legacy function `detect` and disregard any new arguments without errors (PR #262)\n\n### Removed\n- Support for Python 3.6 (PR #260)\n\n### Changed\n- Optional speedup provided by mypy/c 1.0.1\n\n## [3.0.1](https://github.com/Ousret/charset_normalizer/compare/3.0.0...3.0.1) (2022-11-18)\n\n### Fixed\n- Multi-bytes cutter/chunk generator did not always cut correctly (PR #233)\n\n### Changed\n- Speedup provided by mypy/c 0.990 on Python >= 3.7\n\n## [3.0.0](https://github.com/Ousret/charset_normalizer/compare/2.1.1...3.0.0) (2022-10-20)\n\n### Added\n- Extend the capability of explain=True when cp_isolation contains at most two entries (min one), will log in details of the Mess-detector results\n- Support for alternative language frequency set in charset_normalizer.assets.FREQUENCIES\n- Add parameter `language_threshold` in `from_bytes`, `from_path` and `from_fp` to adjust the minimum expected coherence ratio\n- `normalizer --version` now specify if current version provide extra speedup (meaning mypyc compilation whl)\n\n### Changed\n- Build with static metadata using 'build' frontend\n- Make the language detection stricter\n- Optional: Module `md.py` can be compiled using Mypyc to provide an extra speedup up to 4x faster than v2.1\n\n### Fixed\n- CLI with opt --normalize fail when using full path for files\n- TooManyAccentuatedPlugin induce false positive on the mess detection when too few alpha character have been fed to it\n- Sphinx warnings when generating the documentation\n\n### Removed\n- Coherence detector no longer return 'Simple English' instead return 'English'\n- Coherence detector no longer return 'Classical Chinese' instead return 'Chinese'\n- Breaking: Method `first()` and `best()` from CharsetMatch\n- UTF-7 will no longer appear as \"detected\" without a recognized SIG/mark (is unreliable/conflict with ASCII)\n- Breaking: Class aliases CharsetDetector, CharsetDoctor, CharsetNormalizerMatch and CharsetNormalizerMatches\n- Breaking: Top-level function `normalize`\n- Breaking: Properties `chaos_secondary_pass`, `coherence_non_latin` and `w_counter` from CharsetMatch\n- Support for the backport `unicodedata2`\n\n## [3.0.0rc1](https://github.com/Ousret/charset_normalizer/compare/3.0.0b2...3.0.0rc1) (2022-10-18)\n\n### Added\n- Extend the capability of explain=True when cp_isolation contains at most two entries (min one), will log in details of the Mess-detector results\n- Support for alternative language frequency set in charset_normalizer.assets.FREQUENCIES\n- Add parameter `language_threshold` in `from_bytes`, `from_path` and `from_fp` to adjust the minimum expected coherence ratio\n\n### Changed\n- Build with static metadata using 'build' frontend\n- Make the language detection stricter\n\n### Fixed\n- CLI with opt --normalize fail when using full path for files\n- TooManyAccentuatedPlugin induce false positive on the mess detection when too few alpha character have been fed to it\n\n### Removed\n- Coherence detector no longer return 'Simple English' instead return 'English'\n- Coherence detector no longer return 'Classical Chinese' instead return 'Chinese'\n\n## [3.0.0b2](https://github.com/Ousret/charset_normalizer/compare/3.0.0b1...3.0.0b2) (2022-08-21)\n\n### Added\n- `normalizer --version` now specify if current version provide extra speedup (meaning mypyc compilation whl)\n\n### Removed\n- Breaking: Method `first()` and `best()` from CharsetMatch\n- UTF-7 will no longer appear as \"detected\" without a recognized SIG/mark (is unreliable/conflict with ASCII)\n\n### Fixed\n- Sphinx warnings when generating the documentation\n\n## [3.0.0b1](https://github.com/Ousret/charset_normalizer/compare/2.1.0...3.0.0b1) (2022-08-15)\n\n### Changed\n- Optional: Module `md.py` can be compiled using Mypyc to provide an extra speedup up to 4x faster than v2.1\n\n### Removed\n- Breaking: Class aliases CharsetDetector, CharsetDoctor, CharsetNormalizerMatch and CharsetNormalizerMatches\n- Breaking: Top-level function `normalize`\n- Breaking: Properties `chaos_secondary_pass`, `coherence_non_latin` and `w_counter` from CharsetMatch\n- Support for the backport `unicodedata2`\n\n## [2.1.1](https://github.com/Ousret/charset_normalizer/compare/2.1.0...2.1.1) (2022-08-19)\n\n### Deprecated\n- Function `normalize` scheduled for removal in 3.0\n\n### Changed\n- Removed useless call to decode in fn is_unprintable (#206)\n\n### Fixed\n- Third-party library (i18n xgettext) crashing not recognizing utf_8 (PEP 263) with underscore from [@aleksandernovikov](https://github.com/aleksandernovikov) (#204)\n\n## [2.1.0](https://github.com/Ousret/charset_normalizer/compare/2.0.12...2.1.0) (2022-06-19)\n\n### Added\n- Output the Unicode table version when running the CLI with `--version` (PR #194)\n\n### Changed\n- Re-use decoded buffer for single byte character sets from [@nijel](https://github.com/nijel) (PR #175)\n- Fixing some performance bottlenecks from [@deedy5](https://github.com/deedy5) (PR #183)\n\n### Fixed\n- Workaround potential bug in cpython with Zero Width No-Break Space located in Arabic Presentation Forms-B, Unicode 1.1 not acknowledged as space (PR #175)\n- CLI default threshold aligned with the API threshold from [@oleksandr-kuzmenko](https://github.com/oleksandr-kuzmenko) (PR #181)\n\n### Removed\n- Support for Python 3.5 (PR #192)\n\n### Deprecated\n- Use of backport unicodedata from `unicodedata2` as Python is quickly catching up, scheduled for removal in 3.0 (PR #194)\n\n## [2.0.12](https://github.com/Ousret/charset_normalizer/compare/2.0.11...2.0.12) (2022-02-12)\n\n### Fixed\n- ASCII miss-detection on rare cases (PR #170)\n\n## [2.0.11](https://github.com/Ousret/charset_normalizer/compare/2.0.10...2.0.11) (2022-01-30)\n\n### Added\n- Explicit support for Python 3.11 (PR #164)\n\n### Changed\n- The logging behavior have been completely reviewed, now using only TRACE and DEBUG levels (PR #163 #165)\n\n## [2.0.10](https://github.com/Ousret/charset_normalizer/compare/2.0.9...2.0.10) (2022-01-04)\n\n### Fixed\n- Fallback match entries might lead to UnicodeDecodeError for large bytes sequence (PR #154)\n\n### Changed\n- Skipping the language-detection (CD) on ASCII (PR #155)\n\n## [2.0.9](https://github.com/Ousret/charset_normalizer/compare/2.0.8...2.0.9) (2021-12-03)\n\n### Changed\n- Moderating the logging impact (since 2.0.8) for specific environments (PR #147)\n\n### Fixed\n- Wrong logging level applied when setting kwarg `explain` to True (PR #146)\n\n## [2.0.8](https://github.com/Ousret/charset_normalizer/compare/2.0.7...2.0.8) (2021-11-24)\n### Changed\n- Improvement over Vietnamese detection (PR #126)\n- MD improvement on trailing data and long foreign (non-pure latin) data (PR #124)\n- Efficiency improvements in cd/alphabet_languages from [@adbar](https://github.com/adbar) (PR #122)\n- call sum() without an intermediary list following PEP 289 recommendations from [@adbar](https://github.com/adbar) (PR #129)\n- Code style as refactored by Sourcery-AI (PR #131)\n- Minor adjustment on the MD around european words (PR #133)\n- Remove and replace SRTs from assets / tests (PR #139)\n- Initialize the library logger with a `NullHandler` by default from [@nmaynes](https://github.com/nmaynes) (PR #135)\n- Setting kwarg `explain` to True will add provisionally (bounded to function lifespan) a specific stream handler (PR #135)\n\n### Fixed\n- Fix large (misleading) sequence giving UnicodeDecodeError (PR #137)\n- Avoid using too insignificant chunk (PR #137)\n\n### Added\n- Add and expose function `set_logging_handler` to configure a specific StreamHandler from [@nmaynes](https://github.com/nmaynes) (PR #135)\n- Add `CHANGELOG.md` entries, format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/) (PR #141)\n\n## [2.0.7](https://github.com/Ousret/charset_normalizer/compare/2.0.6...2.0.7) (2021-10-11)\n### Added\n- Add support for Kazakh (Cyrillic) language detection (PR #109)\n\n### Changed\n- Further, improve inferring the language from a given single-byte code page (PR #112)\n- Vainly trying to leverage PEP263 when PEP3120 is not supported (PR #116)\n- Refactoring for potential performance improvements in loops from [@adbar](https://github.com/adbar) (PR #113)\n- Various detection improvement (MD+CD) (PR #117)\n\n### Removed\n- Remove redundant logging entry about detected language(s) (PR #115)\n\n### Fixed\n- Fix a minor inconsistency between Python 3.5 and other versions regarding language detection (PR #117 #102)\n\n## [2.0.6](https://github.com/Ousret/charset_normalizer/compare/2.0.5...2.0.6) (2021-09-18)\n### Fixed\n- Unforeseen regression with the loss of the backward-compatibility with some older minor of Python 3.5.x (PR #100)\n- Fix CLI crash when using --minimal output in certain cases (PR #103)\n\n### Changed\n- Minor improvement to the detection efficiency (less than 1%) (PR #106 #101)\n\n## [2.0.5](https://github.com/Ousret/charset_normalizer/compare/2.0.4...2.0.5) (2021-09-14)\n### Changed\n- The project now comply with: flake8, mypy, isort and black to ensure a better overall quality (PR #81)\n- The BC-support with v1.x was improved, the old staticmethods are restored (PR #82)\n- The Unicode detection is slightly improved (PR #93)\n- Add syntax sugar \\_\\_bool\\_\\_ for results CharsetMatches list-container (PR #91)\n\n### Removed\n- The project no longer raise warning on tiny content given for detection, will be simply logged as warning instead (PR #92)\n\n### Fixed\n- In some rare case, the chunks extractor could cut in the middle of a multi-byte character and could mislead the mess detection (PR #95)\n- Some rare 'space' characters could trip up the UnprintablePlugin/Mess detection (PR #96)\n- The MANIFEST.in was not exhaustive (PR #78)\n\n## [2.0.4](https://github.com/Ousret/charset_normalizer/compare/2.0.3...2.0.4) (2021-07-30)\n### Fixed\n- The CLI no longer raise an unexpected exception when no encoding has been found (PR #70)\n- Fix accessing the 'alphabets' property when the payload contains surrogate characters (PR #68)\n- The logger could mislead (explain=True) on detected languages and the impact of one MBCS match (PR #72)\n- Submatch factoring could be wrong in rare edge cases (PR #72)\n- Multiple files given to the CLI were ignored when publishing results to STDOUT. (After the first path) (PR #72)\n- Fix line endings from CRLF to LF for certain project files (PR #67)\n\n### Changed\n- Adjust the MD to lower the sensitivity, thus improving the global detection reliability (PR #69 #76)\n- Allow fallback on specified encoding if any (PR #71)\n\n## [2.0.3](https://github.com/Ousret/charset_normalizer/compare/2.0.2...2.0.3) (2021-07-16)\n### Changed\n- Part of the detection mechanism has been improved to be less sensitive, resulting in more accurate detection results. Especially ASCII. (PR #63)\n- According to the community wishes, the detection will fall back on ASCII or UTF-8 in a last-resort case. (PR #64)\n\n## [2.0.2](https://github.com/Ousret/charset_normalizer/compare/2.0.1...2.0.2) (2021-07-15)\n### Fixed\n- Empty/Too small JSON payload miss-detection fixed. Report from [@tseaver](https://github.com/tseaver) (PR #59)\n\n### Changed\n- Don't inject unicodedata2 into sys.modules from [@akx](https://github.com/akx) (PR #57)\n\n## [2.0.1](https://github.com/Ousret/charset_normalizer/compare/2.0.0...2.0.1) (2021-07-13)\n### Fixed\n- Make it work where there isn't a filesystem available, dropping assets frequencies.json. Report from [@sethmlarson](https://github.com/sethmlarson). (PR #55)\n- Using explain=False permanently disable the verbose output in the current runtime (PR #47)\n- One log entry (language target preemptive) was not show in logs when using explain=True (PR #47)\n- Fix undesired exception (ValueError) on getitem of instance CharsetMatches (PR #52)\n\n### Changed\n- Public function normalize default args values were not aligned with from_bytes (PR #53)\n\n### Added\n- You may now use charset aliases in cp_isolation and cp_exclusion arguments (PR #47)\n\n## [2.0.0](https://github.com/Ousret/charset_normalizer/compare/1.4.1...2.0.0) (2021-07-02)\n### Changed\n- 4x to 5 times faster than the previous 1.4.0 release. At least 2x faster than Chardet.\n- Accent has been made on UTF-8 detection, should perform rather instantaneous.\n- The backward compatibility with Chardet has been greatly improved. The legacy detect function returns an identical charset name whenever possible.\n- The detection mechanism has been slightly improved, now Turkish content is detected correctly (most of the time)\n- The program has been rewritten to ease the readability and maintainability. (+Using static typing)+\n- utf_7 detection has been reinstated.\n\n### Removed\n- This package no longer require anything when used with Python 3.5 (Dropped cached_property)\n- Removed support for these languages: Catalan, Esperanto, Kazakh, Baque, Volap\u00fck, Azeri, Galician, Nynorsk, Macedonian, and Serbocroatian.\n- The exception hook on UnicodeDecodeError has been removed.\n\n### Deprecated\n- Methods coherence_non_latin, w_counter, chaos_secondary_pass of the class CharsetMatch are now deprecated and scheduled for removal in v3.0\n\n### Fixed\n- The CLI output used the relative path of the file(s). Should be absolute.\n\n## [1.4.1](https://github.com/Ousret/charset_normalizer/compare/1.4.0...1.4.1) (2021-05-28)\n### Fixed\n- Logger configuration/usage no longer conflict with others (PR #44)\n\n## [1.4.0](https://github.com/Ousret/charset_normalizer/compare/1.3.9...1.4.0) (2021-05-21)\n### Removed\n- Using standard logging instead of using the package loguru.\n- Dropping nose test framework in favor of the maintained pytest.\n- Choose to not use dragonmapper package to help with gibberish Chinese/CJK text.\n- Require cached_property only for Python 3.5 due to constraint. Dropping for every other interpreter version.\n- Stop support for UTF-7 that does not contain a SIG.\n- Dropping PrettyTable, replaced with pure JSON output in CLI.\n\n### Fixed\n- BOM marker in a CharsetNormalizerMatch instance could be False in rare cases even if obviously present. Due to the sub-match factoring process.\n- Not searching properly for the BOM when trying utf32/16 parent codec.\n\n### Changed\n- Improving the package final size by compressing frequencies.json.\n- Huge improvement over the larges payload.\n\n### Added\n- CLI now produces JSON consumable output.\n- Return ASCII if given sequences fit. Given reasonable confidence.\n\n## [1.3.9](https://github.com/Ousret/charset_normalizer/compare/1.3.8...1.3.9) (2021-05-13)\n\n### Fixed\n- In some very rare cases, you may end up getting encode/decode errors due to a bad bytes payload (PR #40)\n\n## [1.3.8](https://github.com/Ousret/charset_normalizer/compare/1.3.7...1.3.8) (2021-05-12)\n\n### Fixed\n- Empty given payload for detection may cause an exception if trying to access the `alphabets` property. (PR #39)\n\n## [1.3.7](https://github.com/Ousret/charset_normalizer/compare/1.3.6...1.3.7) (2021-05-12)\n\n### Fixed\n- The legacy detect function should return UTF-8-SIG if sig is present in the payload. (PR #38)\n\n## [1.3.6](https://github.com/Ousret/charset_normalizer/compare/1.3.5...1.3.6) (2021-02-09)\n\n### Changed\n- Amend the previous release to allow prettytable 2.0 (PR #35)\n\n## [1.3.5](https://github.com/Ousret/charset_normalizer/compare/1.3.4...1.3.5) (2021-02-08)\n\n### Fixed\n- Fix error while using the package with a python pre-release interpreter (PR #33)\n\n### Changed\n- Dependencies refactoring, constraints revised.\n\n### Added\n- Add python 3.9 and 3.10 to the supported interpreters\n\nMIT License\n\nCopyright (c) 2025 TAHRI Ahmed R.\n\nPermission is hereby granted, free of charge, to any person obtaining a copy\nof this software and associated documentation files (the \"Software\"), to deal\nin the Software without restriction, including without limitation the rights\nto use, copy, modify, merge, publish, distribute, sublicense, and/or sell\ncopies of the Software, and to permit persons to whom the Software is\nfurnished to do so, subject to the following conditions:\n\nThe above copyright notice and this permission notice shall be included in all\ncopies or substantial portions of the Software.\n\nTHE SOFTWARE IS PROVIDED \"AS IS\", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR\nIMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,\nFITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE\nAUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER\nLIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,\nOUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE\nSOFTWARE.", - "release_date": "2024-12-24T18:11:22", + "description": "The Real First Universal Charset Detector. Open, modern and actively maintained alternative to Chardet.\n

Charset Detection, for Everyone \ud83d\udc4b

\n\n

\n The Real First Universal Charset Detector
\n \n \n \n \n \"Download\n \n \n \n \n

\n

\n Featured Packages
\n \n \"Static\n \n \n \"Static\n \n

\n

\n In other language (unofficial port - by the community)
\n \n \"Static\n \n

\n\n> A library that helps you read text from an unknown charset encoding.
Motivated by `chardet`,\n> I'm trying to resolve the issue by taking a new approach.\n> All IANA character set names for which the Python core library provides codecs are supported.\n\n

\n >>>>> \ud83d\udc49 Try Me Online Now, Then Adopt Me \ud83d\udc48 <<<<<\n

\n\nThis project offers you an alternative to **Universal Charset Encoding Detector**, also known as **Chardet**.\n\n| Feature | [Chardet](https://github.com/chardet/chardet) | Charset Normalizer | [cChardet](https://github.com/PyYoshi/cChardet) |\n|--------------------------------------------------|:---------------------------------------------:|:--------------------------------------------------------------------------------------------------:|:-----------------------------------------------:|\n| `Fast` | \u274c | \u2705 | \u2705 |\n| `Universal**` | \u274c | \u2705 | \u274c |\n| `Reliable` **without** distinguishable standards | \u274c | \u2705 | \u2705 |\n| `Reliable` **with** distinguishable standards | \u2705 | \u2705 | \u2705 |\n| `License` | LGPL-2.1
_restrictive_ | MIT | MPL-1.1
_restrictive_ |\n| `Native Python` | \u2705 | \u2705 | \u274c |\n| `Detect spoken language` | \u274c | \u2705 | N/A |\n| `UnicodeDecodeError Safety` | \u274c | \u2705 | \u274c |\n| `Whl Size (min)` | 193.6 kB | 42 kB | ~200 kB |\n| `Supported Encoding` | 33 | \ud83c\udf89 [99](https://charset-normalizer.readthedocs.io/en/latest/user/support.html#supported-encodings) | 40 |\n\n

\n\"Reading\"Cat\n

\n\n*\\*\\* : They are clearly using specific code for a specific encoding even if covering most of used one*
\n\n## \u26a1 Performance\n\nThis package offer better performance than its counterpart Chardet. Here are some numbers.\n\n| Package | Accuracy | Mean per file (ms) | File per sec (est) |\n|-----------------------------------------------|:--------:|:------------------:|:------------------:|\n| [chardet](https://github.com/chardet/chardet) | 86 % | 63 ms | 16 file/sec |\n| charset-normalizer | **98 %** | **10 ms** | 100 file/sec |\n\n| Package | 99th percentile | 95th percentile | 50th percentile |\n|-----------------------------------------------|:---------------:|:---------------:|:---------------:|\n| [chardet](https://github.com/chardet/chardet) | 265 ms | 71 ms | 7 ms |\n| charset-normalizer | 100 ms | 50 ms | 5 ms |\n\n_updated as of december 2024 using CPython 3.12_\n\nChardet's performance on larger file (1MB+) are very poor. Expect huge difference on large payload.\n\n> Stats are generated using 400+ files using default parameters. More details on used files, see GHA workflows.\n> And yes, these results might change at any time. The dataset can be updated to include more files.\n> The actual delays heavily depends on your CPU capabilities. The factors should remain the same.\n> Keep in mind that the stats are generous and that Chardet accuracy vs our is measured using Chardet initial capability\n> (e.g. Supported Encoding) Challenge-them if you want.\n\n## \u2728 Installation\n\nUsing pip:\n\n```sh\npip install charset-normalizer -U\n```\n\n## \ud83d\ude80 Basic Usage\n\n### CLI\nThis package comes with a CLI.\n\n```\nusage: normalizer [-h] [-v] [-a] [-n] [-m] [-r] [-f] [-t THRESHOLD]\n file [file ...]\n\nThe Real First Universal Charset Detector. Discover originating encoding used\non text file. Normalize text to unicode.\n\npositional arguments:\n files File(s) to be analysed\n\noptional arguments:\n -h, --help show this help message and exit\n -v, --verbose Display complementary information about file if any.\n Stdout will contain logs about the detection process.\n -a, --with-alternative\n Output complementary possibilities if any. Top-level\n JSON WILL be a list.\n -n, --normalize Permit to normalize input file. If not set, program\n does not write anything.\n -m, --minimal Only output the charset detected to STDOUT. Disabling\n JSON output.\n -r, --replace Replace file when trying to normalize it instead of\n creating a new one.\n -f, --force Replace file without asking if you are sure, use this\n flag with caution.\n -t THRESHOLD, --threshold THRESHOLD\n Define a custom maximum amount of chaos allowed in\n decoded content. 0. <= chaos <= 1.\n --version Show version information and exit.\n```\n\n```bash\nnormalizer ./data/sample.1.fr.srt\n```\n\nor\n\n```bash\npython -m charset_normalizer ./data/sample.1.fr.srt\n```\n\n\ud83c\udf89 Since version 1.4.0 the CLI produce easily usable stdout result in JSON format.\n\n```json\n{\n \"path\": \"/home/default/projects/charset_normalizer/data/sample.1.fr.srt\",\n \"encoding\": \"cp1252\",\n \"encoding_aliases\": [\n \"1252\",\n \"windows_1252\"\n ],\n \"alternative_encodings\": [\n \"cp1254\",\n \"cp1256\",\n \"cp1258\",\n \"iso8859_14\",\n \"iso8859_15\",\n \"iso8859_16\",\n \"iso8859_3\",\n \"iso8859_9\",\n \"latin_1\",\n \"mbcs\"\n ],\n \"language\": \"French\",\n \"alphabets\": [\n \"Basic Latin\",\n \"Latin-1 Supplement\"\n ],\n \"has_sig_or_bom\": false,\n \"chaos\": 0.149,\n \"coherence\": 97.152,\n \"unicode_path\": null,\n \"is_preferred\": true\n}\n```\n\n### Python\n*Just print out normalized text*\n```python\nfrom charset_normalizer import from_path\n\nresults = from_path('./my_subtitle.srt')\n\nprint(str(results.best()))\n```\n\n*Upgrade your code without effort*\n```python\nfrom charset_normalizer import detect\n```\n\nThe above code will behave the same as **chardet**. We ensure that we offer the best (reasonable) BC result possible.\n\nSee the docs for advanced usage : [readthedocs.io](https://charset-normalizer.readthedocs.io/en/latest/)\n\n## \ud83d\ude07 Why\n\nWhen I started using Chardet, I noticed that it was not suited to my expectations, and I wanted to propose a\nreliable alternative using a completely different method. Also! I never back down on a good challenge!\n\nI **don't care** about the **originating charset** encoding, because **two different tables** can\nproduce **two identical rendered string.**\nWhat I want is to get readable text, the best I can.\n\nIn a way, **I'm brute forcing text decoding.** How cool is that ? \ud83d\ude0e\n\nDon't confuse package **ftfy** with charset-normalizer or chardet. ftfy goal is to repair Unicode string whereas charset-normalizer to convert raw file in unknown encoding to unicode.\n\n## \ud83c\udf70 How\n\n - Discard all charset encoding table that could not fit the binary content.\n - Measure noise, or the mess once opened (by chunks) with a corresponding charset encoding.\n - Extract matches with the lowest mess detected.\n - Additionally, we measure coherence / probe for a language.\n\n**Wait a minute**, what is noise/mess and coherence according to **YOU ?**\n\n*Noise :* I opened hundred of text files, **written by humans**, with the wrong encoding table. **I observed**, then\n**I established** some ground rules about **what is obvious** when **it seems like** a mess (aka. defining noise in rendered text).\n I know that my interpretation of what is noise is probably incomplete, feel free to contribute in order to\n improve or rewrite it.\n\n*Coherence :* For each language there is on earth, we have computed ranked letter appearance occurrences (the best we can). So I thought\nthat intel is worth something here. So I use those records against decoded text to check if I can detect intelligent design.\n\n## \u26a1 Known limitations\n\n - Language detection is unreliable when text contains two or more languages sharing identical letters. (eg. HTML (english tags) + Turkish content (Sharing Latin characters))\n - Every charset detector heavily depends on sufficient content. In common cases, do not bother run detection on very tiny content.\n\n## \u26a0\ufe0f About Python EOLs\n\n**If you are running:**\n\n- Python >=2.7,<3.5: Unsupported\n- Python 3.5: charset-normalizer < 2.1\n- Python 3.6: charset-normalizer < 3.1\n- Python 3.7: charset-normalizer < 4.0\n\nUpgrade your Python interpreter as soon as possible.\n\n## \ud83d\udc64 Contributing\n\nContributions, issues and feature requests are very much welcome.
\nFeel free to check [issues page](https://github.com/ousret/charset_normalizer/issues) if you want to contribute.\n\n## \ud83d\udcdd License\n\nCopyright \u00a9 [Ahmed TAHRI @Ousret](https://github.com/Ousret).
\nThis project is [MIT](https://github.com/Ousret/charset_normalizer/blob/master/LICENSE) licensed.\n\nCharacters frequencies used in this project \u00a9 2012 [Denny Vrande\u010di\u0107](http://simia.net/letters/)\n\n## \ud83d\udcbc For Enterprise\n\nProfessional support for charset-normalizer is available as part of the [Tidelift\nSubscription][1]. Tidelift gives software development teams a single source for\npurchasing and maintaining their software, with professional grade assurances\nfrom the experts who know it best, while seamlessly integrating with existing\ntools.\n\n[1]: https://tidelift.com/subscription/pkg/pypi-charset-normalizer?utm_source=pypi-charset-normalizer&utm_medium=readme\n\n[![OpenSSF Best Practices](https://www.bestpractices.dev/projects/7297/badge)](https://www.bestpractices.dev/projects/7297)\n\n# Changelog\nAll notable changes to charset-normalizer will be documented in this file. This project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html).\nThe format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/).\n\n## [3.4.2](https://github.com/Ousret/charset_normalizer/compare/3.4.1...3.4.2) (2025-05-02)\n\n### Fixed\n- Addressed the DeprecationWarning in our CLI regarding `argparse.FileType` by backporting the target class into the package. (#591)\n- Improved the overall reliability of the detector with CJK Ideographs. (#605) (#587)\n\n### Changed\n- Optional mypyc compilation upgraded to version 1.15 for Python >= 3.8\n\n## [3.4.1](https://github.com/Ousret/charset_normalizer/compare/3.4.0...3.4.1) (2024-12-24)\n\n### Changed\n- Project metadata are now stored using `pyproject.toml` instead of `setup.cfg` using setuptools as the build backend.\n- Enforce annotation delayed loading for a simpler and consistent types in the project.\n- Optional mypyc compilation upgraded to version 1.14 for Python >= 3.8\n\n### Added\n- pre-commit configuration.\n- noxfile.\n\n### Removed\n- `build-requirements.txt` as per using `pyproject.toml` native build configuration.\n- `bin/integration.py` and `bin/serve.py` in favor of downstream integration test (see noxfile).\n- `setup.cfg` in favor of `pyproject.toml` metadata configuration.\n- Unused `utils.range_scan` function.\n\n### Fixed\n- Converting content to Unicode bytes may insert `utf_8` instead of preferred `utf-8`. (#572)\n- Deprecation warning \"'count' is passed as positional argument\" when converting to Unicode bytes on Python 3.13+\n\n## [3.4.0](https://github.com/Ousret/charset_normalizer/compare/3.3.2...3.4.0) (2024-10-08)\n\n### Added\n- Argument `--no-preemptive` in the CLI to prevent the detector to search for hints.\n- Support for Python 3.13 (#512)\n\n### Fixed\n- Relax the TypeError exception thrown when trying to compare a CharsetMatch with anything else than a CharsetMatch.\n- Improved the general reliability of the detector based on user feedbacks. (#520) (#509) (#498) (#407) (#537)\n- Declared charset in content (preemptive detection) not changed when converting to utf-8 bytes. (#381)\n\n## [3.3.2](https://github.com/Ousret/charset_normalizer/compare/3.3.1...3.3.2) (2023-10-31)\n\n### Fixed\n- Unintentional memory usage regression when using large payload that match several encoding (#376)\n- Regression on some detection case showcased in the documentation (#371)\n\n### Added\n- Noise (md) probe that identify malformed arabic representation due to the presence of letters in isolated form (credit to my wife)\n\n## [3.3.1](https://github.com/Ousret/charset_normalizer/compare/3.3.0...3.3.1) (2023-10-22)\n\n### Changed\n- Optional mypyc compilation upgraded to version 1.6.1 for Python >= 3.8\n- Improved the general detection reliability based on reports from the community\n\n## [3.3.0](https://github.com/Ousret/charset_normalizer/compare/3.2.0...3.3.0) (2023-09-30)\n\n### Added\n- Allow to execute the CLI (e.g. normalizer) through `python -m charset_normalizer.cli` or `python -m charset_normalizer`\n- Support for 9 forgotten encoding that are supported by Python but unlisted in `encoding.aliases` as they have no alias (#323)\n\n### Removed\n- (internal) Redundant utils.is_ascii function and unused function is_private_use_only\n- (internal) charset_normalizer.assets is moved inside charset_normalizer.constant\n\n### Changed\n- (internal) Unicode code blocks in constants are updated using the latest v15.0.0 definition to improve detection\n- Optional mypyc compilation upgraded to version 1.5.1 for Python >= 3.8\n\n### Fixed\n- Unable to properly sort CharsetMatch when both chaos/noise and coherence were close due to an unreachable condition in \\_\\_lt\\_\\_ (#350)\n\n## [3.2.0](https://github.com/Ousret/charset_normalizer/compare/3.1.0...3.2.0) (2023-06-07)\n\n### Changed\n- Typehint for function `from_path` no longer enforce `PathLike` as its first argument\n- Minor improvement over the global detection reliability\n\n### Added\n- Introduce function `is_binary` that relies on main capabilities, and optimized to detect binaries\n- Propagate `enable_fallback` argument throughout `from_bytes`, `from_path`, and `from_fp` that allow a deeper control over the detection (default True)\n- Explicit support for Python 3.12\n\n### Fixed\n- Edge case detection failure where a file would contain 'very-long' camel cased word (Issue #289)\n\n## [3.1.0](https://github.com/Ousret/charset_normalizer/compare/3.0.1...3.1.0) (2023-03-06)\n\n### Added\n- Argument `should_rename_legacy` for legacy function `detect` and disregard any new arguments without errors (PR #262)\n\n### Removed\n- Support for Python 3.6 (PR #260)\n\n### Changed\n- Optional speedup provided by mypy/c 1.0.1\n\n## [3.0.1](https://github.com/Ousret/charset_normalizer/compare/3.0.0...3.0.1) (2022-11-18)\n\n### Fixed\n- Multi-bytes cutter/chunk generator did not always cut correctly (PR #233)\n\n### Changed\n- Speedup provided by mypy/c 0.990 on Python >= 3.7\n\n## [3.0.0](https://github.com/Ousret/charset_normalizer/compare/2.1.1...3.0.0) (2022-10-20)\n\n### Added\n- Extend the capability of explain=True when cp_isolation contains at most two entries (min one), will log in details of the Mess-detector results\n- Support for alternative language frequency set in charset_normalizer.assets.FREQUENCIES\n- Add parameter `language_threshold` in `from_bytes`, `from_path` and `from_fp` to adjust the minimum expected coherence ratio\n- `normalizer --version` now specify if current version provide extra speedup (meaning mypyc compilation whl)\n\n### Changed\n- Build with static metadata using 'build' frontend\n- Make the language detection stricter\n- Optional: Module `md.py` can be compiled using Mypyc to provide an extra speedup up to 4x faster than v2.1\n\n### Fixed\n- CLI with opt --normalize fail when using full path for files\n- TooManyAccentuatedPlugin induce false positive on the mess detection when too few alpha character have been fed to it\n- Sphinx warnings when generating the documentation\n\n### Removed\n- Coherence detector no longer return 'Simple English' instead return 'English'\n- Coherence detector no longer return 'Classical Chinese' instead return 'Chinese'\n- Breaking: Method `first()` and `best()` from CharsetMatch\n- UTF-7 will no longer appear as \"detected\" without a recognized SIG/mark (is unreliable/conflict with ASCII)\n- Breaking: Class aliases CharsetDetector, CharsetDoctor, CharsetNormalizerMatch and CharsetNormalizerMatches\n- Breaking: Top-level function `normalize`\n- Breaking: Properties `chaos_secondary_pass`, `coherence_non_latin` and `w_counter` from CharsetMatch\n- Support for the backport `unicodedata2`\n\n## [3.0.0rc1](https://github.com/Ousret/charset_normalizer/compare/3.0.0b2...3.0.0rc1) (2022-10-18)\n\n### Added\n- Extend the capability of explain=True when cp_isolation contains at most two entries (min one), will log in details of the Mess-detector results\n- Support for alternative language frequency set in charset_normalizer.assets.FREQUENCIES\n- Add parameter `language_threshold` in `from_bytes`, `from_path` and `from_fp` to adjust the minimum expected coherence ratio\n\n### Changed\n- Build with static metadata using 'build' frontend\n- Make the language detection stricter\n\n### Fixed\n- CLI with opt --normalize fail when using full path for files\n- TooManyAccentuatedPlugin induce false positive on the mess detection when too few alpha character have been fed to it\n\n### Removed\n- Coherence detector no longer return 'Simple English' instead return 'English'\n- Coherence detector no longer return 'Classical Chinese' instead return 'Chinese'\n\n## [3.0.0b2](https://github.com/Ousret/charset_normalizer/compare/3.0.0b1...3.0.0b2) (2022-08-21)\n\n### Added\n- `normalizer --version` now specify if current version provide extra speedup (meaning mypyc compilation whl)\n\n### Removed\n- Breaking: Method `first()` and `best()` from CharsetMatch\n- UTF-7 will no longer appear as \"detected\" without a recognized SIG/mark (is unreliable/conflict with ASCII)\n\n### Fixed\n- Sphinx warnings when generating the documentation\n\n## [3.0.0b1](https://github.com/Ousret/charset_normalizer/compare/2.1.0...3.0.0b1) (2022-08-15)\n\n### Changed\n- Optional: Module `md.py` can be compiled using Mypyc to provide an extra speedup up to 4x faster than v2.1\n\n### Removed\n- Breaking: Class aliases CharsetDetector, CharsetDoctor, CharsetNormalizerMatch and CharsetNormalizerMatches\n- Breaking: Top-level function `normalize`\n- Breaking: Properties `chaos_secondary_pass`, `coherence_non_latin` and `w_counter` from CharsetMatch\n- Support for the backport `unicodedata2`\n\n## [2.1.1](https://github.com/Ousret/charset_normalizer/compare/2.1.0...2.1.1) (2022-08-19)\n\n### Deprecated\n- Function `normalize` scheduled for removal in 3.0\n\n### Changed\n- Removed useless call to decode in fn is_unprintable (#206)\n\n### Fixed\n- Third-party library (i18n xgettext) crashing not recognizing utf_8 (PEP 263) with underscore from [@aleksandernovikov](https://github.com/aleksandernovikov) (#204)\n\n## [2.1.0](https://github.com/Ousret/charset_normalizer/compare/2.0.12...2.1.0) (2022-06-19)\n\n### Added\n- Output the Unicode table version when running the CLI with `--version` (PR #194)\n\n### Changed\n- Re-use decoded buffer for single byte character sets from [@nijel](https://github.com/nijel) (PR #175)\n- Fixing some performance bottlenecks from [@deedy5](https://github.com/deedy5) (PR #183)\n\n### Fixed\n- Workaround potential bug in cpython with Zero Width No-Break Space located in Arabic Presentation Forms-B, Unicode 1.1 not acknowledged as space (PR #175)\n- CLI default threshold aligned with the API threshold from [@oleksandr-kuzmenko](https://github.com/oleksandr-kuzmenko) (PR #181)\n\n### Removed\n- Support for Python 3.5 (PR #192)\n\n### Deprecated\n- Use of backport unicodedata from `unicodedata2` as Python is quickly catching up, scheduled for removal in 3.0 (PR #194)\n\n## [2.0.12](https://github.com/Ousret/charset_normalizer/compare/2.0.11...2.0.12) (2022-02-12)\n\n### Fixed\n- ASCII miss-detection on rare cases (PR #170)\n\n## [2.0.11](https://github.com/Ousret/charset_normalizer/compare/2.0.10...2.0.11) (2022-01-30)\n\n### Added\n- Explicit support for Python 3.11 (PR #164)\n\n### Changed\n- The logging behavior have been completely reviewed, now using only TRACE and DEBUG levels (PR #163 #165)\n\n## [2.0.10](https://github.com/Ousret/charset_normalizer/compare/2.0.9...2.0.10) (2022-01-04)\n\n### Fixed\n- Fallback match entries might lead to UnicodeDecodeError for large bytes sequence (PR #154)\n\n### Changed\n- Skipping the language-detection (CD) on ASCII (PR #155)\n\n## [2.0.9](https://github.com/Ousret/charset_normalizer/compare/2.0.8...2.0.9) (2021-12-03)\n\n### Changed\n- Moderating the logging impact (since 2.0.8) for specific environments (PR #147)\n\n### Fixed\n- Wrong logging level applied when setting kwarg `explain` to True (PR #146)\n\n## [2.0.8](https://github.com/Ousret/charset_normalizer/compare/2.0.7...2.0.8) (2021-11-24)\n### Changed\n- Improvement over Vietnamese detection (PR #126)\n- MD improvement on trailing data and long foreign (non-pure latin) data (PR #124)\n- Efficiency improvements in cd/alphabet_languages from [@adbar](https://github.com/adbar) (PR #122)\n- call sum() without an intermediary list following PEP 289 recommendations from [@adbar](https://github.com/adbar) (PR #129)\n- Code style as refactored by Sourcery-AI (PR #131)\n- Minor adjustment on the MD around european words (PR #133)\n- Remove and replace SRTs from assets / tests (PR #139)\n- Initialize the library logger with a `NullHandler` by default from [@nmaynes](https://github.com/nmaynes) (PR #135)\n- Setting kwarg `explain` to True will add provisionally (bounded to function lifespan) a specific stream handler (PR #135)\n\n### Fixed\n- Fix large (misleading) sequence giving UnicodeDecodeError (PR #137)\n- Avoid using too insignificant chunk (PR #137)\n\n### Added\n- Add and expose function `set_logging_handler` to configure a specific StreamHandler from [@nmaynes](https://github.com/nmaynes) (PR #135)\n- Add `CHANGELOG.md` entries, format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/) (PR #141)\n\n## [2.0.7](https://github.com/Ousret/charset_normalizer/compare/2.0.6...2.0.7) (2021-10-11)\n### Added\n- Add support for Kazakh (Cyrillic) language detection (PR #109)\n\n### Changed\n- Further, improve inferring the language from a given single-byte code page (PR #112)\n- Vainly trying to leverage PEP263 when PEP3120 is not supported (PR #116)\n- Refactoring for potential performance improvements in loops from [@adbar](https://github.com/adbar) (PR #113)\n- Various detection improvement (MD+CD) (PR #117)\n\n### Removed\n- Remove redundant logging entry about detected language(s) (PR #115)\n\n### Fixed\n- Fix a minor inconsistency between Python 3.5 and other versions regarding language detection (PR #117 #102)\n\n## [2.0.6](https://github.com/Ousret/charset_normalizer/compare/2.0.5...2.0.6) (2021-09-18)\n### Fixed\n- Unforeseen regression with the loss of the backward-compatibility with some older minor of Python 3.5.x (PR #100)\n- Fix CLI crash when using --minimal output in certain cases (PR #103)\n\n### Changed\n- Minor improvement to the detection efficiency (less than 1%) (PR #106 #101)\n\n## [2.0.5](https://github.com/Ousret/charset_normalizer/compare/2.0.4...2.0.5) (2021-09-14)\n### Changed\n- The project now comply with: flake8, mypy, isort and black to ensure a better overall quality (PR #81)\n- The BC-support with v1.x was improved, the old staticmethods are restored (PR #82)\n- The Unicode detection is slightly improved (PR #93)\n- Add syntax sugar \\_\\_bool\\_\\_ for results CharsetMatches list-container (PR #91)\n\n### Removed\n- The project no longer raise warning on tiny content given for detection, will be simply logged as warning instead (PR #92)\n\n### Fixed\n- In some rare case, the chunks extractor could cut in the middle of a multi-byte character and could mislead the mess detection (PR #95)\n- Some rare 'space' characters could trip up the UnprintablePlugin/Mess detection (PR #96)\n- The MANIFEST.in was not exhaustive (PR #78)\n\n## [2.0.4](https://github.com/Ousret/charset_normalizer/compare/2.0.3...2.0.4) (2021-07-30)\n### Fixed\n- The CLI no longer raise an unexpected exception when no encoding has been found (PR #70)\n- Fix accessing the 'alphabets' property when the payload contains surrogate characters (PR #68)\n- The logger could mislead (explain=True) on detected languages and the impact of one MBCS match (PR #72)\n- Submatch factoring could be wrong in rare edge cases (PR #72)\n- Multiple files given to the CLI were ignored when publishing results to STDOUT. (After the first path) (PR #72)\n- Fix line endings from CRLF to LF for certain project files (PR #67)\n\n### Changed\n- Adjust the MD to lower the sensitivity, thus improving the global detection reliability (PR #69 #76)\n- Allow fallback on specified encoding if any (PR #71)\n\n## [2.0.3](https://github.com/Ousret/charset_normalizer/compare/2.0.2...2.0.3) (2021-07-16)\n### Changed\n- Part of the detection mechanism has been improved to be less sensitive, resulting in more accurate detection results. Especially ASCII. (PR #63)\n- According to the community wishes, the detection will fall back on ASCII or UTF-8 in a last-resort case. (PR #64)\n\n## [2.0.2](https://github.com/Ousret/charset_normalizer/compare/2.0.1...2.0.2) (2021-07-15)\n### Fixed\n- Empty/Too small JSON payload miss-detection fixed. Report from [@tseaver](https://github.com/tseaver) (PR #59)\n\n### Changed\n- Don't inject unicodedata2 into sys.modules from [@akx](https://github.com/akx) (PR #57)\n\n## [2.0.1](https://github.com/Ousret/charset_normalizer/compare/2.0.0...2.0.1) (2021-07-13)\n### Fixed\n- Make it work where there isn't a filesystem available, dropping assets frequencies.json. Report from [@sethmlarson](https://github.com/sethmlarson). (PR #55)\n- Using explain=False permanently disable the verbose output in the current runtime (PR #47)\n- One log entry (language target preemptive) was not show in logs when using explain=True (PR #47)\n- Fix undesired exception (ValueError) on getitem of instance CharsetMatches (PR #52)\n\n### Changed\n- Public function normalize default args values were not aligned with from_bytes (PR #53)\n\n### Added\n- You may now use charset aliases in cp_isolation and cp_exclusion arguments (PR #47)\n\n## [2.0.0](https://github.com/Ousret/charset_normalizer/compare/1.4.1...2.0.0) (2021-07-02)\n### Changed\n- 4x to 5 times faster than the previous 1.4.0 release. At least 2x faster than Chardet.\n- Accent has been made on UTF-8 detection, should perform rather instantaneous.\n- The backward compatibility with Chardet has been greatly improved. The legacy detect function returns an identical charset name whenever possible.\n- The detection mechanism has been slightly improved, now Turkish content is detected correctly (most of the time)\n- The program has been rewritten to ease the readability and maintainability. (+Using static typing)+\n- utf_7 detection has been reinstated.\n\n### Removed\n- This package no longer require anything when used with Python 3.5 (Dropped cached_property)\n- Removed support for these languages: Catalan, Esperanto, Kazakh, Baque, Volap\u00fck, Azeri, Galician, Nynorsk, Macedonian, and Serbocroatian.\n- The exception hook on UnicodeDecodeError has been removed.\n\n### Deprecated\n- Methods coherence_non_latin, w_counter, chaos_secondary_pass of the class CharsetMatch are now deprecated and scheduled for removal in v3.0\n\n### Fixed\n- The CLI output used the relative path of the file(s). Should be absolute.\n\n## [1.4.1](https://github.com/Ousret/charset_normalizer/compare/1.4.0...1.4.1) (2021-05-28)\n### Fixed\n- Logger configuration/usage no longer conflict with others (PR #44)\n\n## [1.4.0](https://github.com/Ousret/charset_normalizer/compare/1.3.9...1.4.0) (2021-05-21)\n### Removed\n- Using standard logging instead of using the package loguru.\n- Dropping nose test framework in favor of the maintained pytest.\n- Choose to not use dragonmapper package to help with gibberish Chinese/CJK text.\n- Require cached_property only for Python 3.5 due to constraint. Dropping for every other interpreter version.\n- Stop support for UTF-7 that does not contain a SIG.\n- Dropping PrettyTable, replaced with pure JSON output in CLI.\n\n### Fixed\n- BOM marker in a CharsetNormalizerMatch instance could be False in rare cases even if obviously present. Due to the sub-match factoring process.\n- Not searching properly for the BOM when trying utf32/16 parent codec.\n\n### Changed\n- Improving the package final size by compressing frequencies.json.\n- Huge improvement over the larges payload.\n\n### Added\n- CLI now produces JSON consumable output.\n- Return ASCII if given sequences fit. Given reasonable confidence.\n\n## [1.3.9](https://github.com/Ousret/charset_normalizer/compare/1.3.8...1.3.9) (2021-05-13)\n\n### Fixed\n- In some very rare cases, you may end up getting encode/decode errors due to a bad bytes payload (PR #40)\n\n## [1.3.8](https://github.com/Ousret/charset_normalizer/compare/1.3.7...1.3.8) (2021-05-12)\n\n### Fixed\n- Empty given payload for detection may cause an exception if trying to access the `alphabets` property. (PR #39)\n\n## [1.3.7](https://github.com/Ousret/charset_normalizer/compare/1.3.6...1.3.7) (2021-05-12)\n\n### Fixed\n- The legacy detect function should return UTF-8-SIG if sig is present in the payload. (PR #38)\n\n## [1.3.6](https://github.com/Ousret/charset_normalizer/compare/1.3.5...1.3.6) (2021-02-09)\n\n### Changed\n- Amend the previous release to allow prettytable 2.0 (PR #35)\n\n## [1.3.5](https://github.com/Ousret/charset_normalizer/compare/1.3.4...1.3.5) (2021-02-08)\n\n### Fixed\n- Fix error while using the package with a python pre-release interpreter (PR #33)\n\n### Changed\n- Dependencies refactoring, constraints revised.\n\n### Added\n- Add python 3.9 and 3.10 to the supported interpreters\n\nMIT License\n\nCopyright (c) 2025 TAHRI Ahmed R.\n\nPermission is hereby granted, free of charge, to any person obtaining a copy\nof this software and associated documentation files (the \"Software\"), to deal\nin the Software without restriction, including without limitation the rights\nto use, copy, modify, merge, publish, distribute, sublicense, and/or sell\ncopies of the Software, and to permit persons to whom the Software is\nfurnished to do so, subject to the following conditions:\n\nThe above copyright notice and this permission notice shall be included in all\ncopies or substantial portions of the Software.\n\nTHE SOFTWARE IS PROVIDED \"AS IS\", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR\nIMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,\nFITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE\nAUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER\nLIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,\nOUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE\nSOFTWARE.", + "release_date": "2025-05-02T08:33:04", "parties": [ { "type": "person", @@ -492,11 +491,11 @@ "Typing :: Typed" ], "homepage_url": null, - "download_url": "https://files.pythonhosted.org/packages/e4/93/946a86ce20790e11312c87c75ba68d5f6ad2208cfb52b2d6a2c32840d922/charset_normalizer-3.4.1-cp313-cp313-musllinux_1_2_x86_64.whl", - "size": 145732, + "download_url": "https://files.pythonhosted.org/packages/e2/28/ffc026b26f441fc67bd21ab7f03b313ab3fe46714a14b516f931abe1a2d8/charset_normalizer-3.4.2-cp313-cp313-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", + "size": 148243, "sha1": null, - "md5": "c4597b636aa48efa186df97d36de5b61", - "sha256": "fa88b843d6e211393a37219e6a1c1df99d35e8fd90446f1118f4216e307e48cd", + "md5": "33e977b54a1ee45dbead0da58594fa8f", + "sha256": "6c9379d65defcab82d07b2a9dfbfc2e95bc8fe0ebb1b176a3190230a3ef0e07c", "sha512": null, "bug_tracking_url": null, "code_view_url": "https://github.com/jawah/charset_normalizer", @@ -516,9 +515,9 @@ "dependencies": [], "repository_homepage_url": null, "repository_download_url": null, - "api_data_url": "https://pypi.org/pypi/charset-normalizer/3.4.1/json", + "api_data_url": "https://pypi.org/pypi/charset-normalizer/3.4.2/json", "datasource_id": null, - "purl": "pkg:pypi/charset-normalizer@3.4.1" + "purl": "pkg:pypi/charset-normalizer@3.4.2" }, { "type": "pypi", @@ -1318,7 +1317,7 @@ ], "resolved_dependencies_graph": [ { - "package": "pkg:pypi/azure-core@1.33.0", + "package": "pkg:pypi/azure-core@1.34.0", "dependencies": [ "pkg:pypi/requests@2.32.3", "pkg:pypi/six@1.17.0", @@ -1334,7 +1333,7 @@ { "package": "pkg:pypi/azure-storage-blob@12.25.1", "dependencies": [ - "pkg:pypi/azure-core@1.33.0", + "pkg:pypi/azure-core@1.34.0", "pkg:pypi/cryptography@44.0.2", "pkg:pypi/isodate@0.7.2", "pkg:pypi/typing-extensions@4.13.2" @@ -1351,7 +1350,7 @@ ] }, { - "package": "pkg:pypi/charset-normalizer@3.4.1", + "package": "pkg:pypi/charset-normalizer@3.4.2", "dependencies": [] }, { @@ -1375,7 +1374,7 @@ { "package": "pkg:pypi/msrest@0.7.1", "dependencies": [ - "pkg:pypi/azure-core@1.33.0", + "pkg:pypi/azure-core@1.34.0", "pkg:pypi/certifi@2025.4.26", "pkg:pypi/isodate@0.7.2", "pkg:pypi/requests-oauthlib@2.0.0", @@ -1401,7 +1400,7 @@ "package": "pkg:pypi/requests@2.32.3", "dependencies": [ "pkg:pypi/certifi@2025.4.26", - "pkg:pypi/charset-normalizer@3.4.1", + "pkg:pypi/charset-normalizer@3.4.2", "pkg:pypi/idna@3.10", "pkg:pypi/urllib3@2.4.0" ] diff --git a/tests/data/azure-devops.req-38-expected.json b/tests/data/azure-devops.req-38-expected.json index 965f693..30f38d7 100644 --- a/tests/data/azure-devops.req-38-expected.json +++ b/tests/data/azure-devops.req-38-expected.json @@ -441,12 +441,12 @@ "type": "pypi", "namespace": null, "name": "charset-normalizer", - "version": "3.4.1", + "version": "3.4.2", "qualifiers": {}, "subpath": null, "primary_language": "Python", - "description": "The Real First Universal Charset Detector. Open, modern and actively maintained alternative to Chardet.\n

Charset Detection, for Everyone \ud83d\udc4b

\n\n

\n The Real First Universal Charset Detector
\n \n \n \n \n \"Download\n \n \n \n \n

\n

\n Featured Packages
\n \n \"Static\n \n \n \"Static\n \n

\n

\n In other language (unofficial port - by the community)
\n \n \"Static\n \n

\n\n> A library that helps you read text from an unknown charset encoding.
Motivated by `chardet`,\n> I'm trying to resolve the issue by taking a new approach.\n> All IANA character set names for which the Python core library provides codecs are supported.\n\n

\n >>>>> \ud83d\udc49 Try Me Online Now, Then Adopt Me \ud83d\udc48 <<<<<\n

\n\nThis project offers you an alternative to **Universal Charset Encoding Detector**, also known as **Chardet**.\n\n| Feature | [Chardet](https://github.com/chardet/chardet) | Charset Normalizer | [cChardet](https://github.com/PyYoshi/cChardet) |\n|--------------------------------------------------|:---------------------------------------------:|:--------------------------------------------------------------------------------------------------:|:-----------------------------------------------:|\n| `Fast` | \u274c | \u2705 | \u2705 |\n| `Universal**` | \u274c | \u2705 | \u274c |\n| `Reliable` **without** distinguishable standards | \u274c | \u2705 | \u2705 |\n| `Reliable` **with** distinguishable standards | \u2705 | \u2705 | \u2705 |\n| `License` | LGPL-2.1
_restrictive_ | MIT | MPL-1.1
_restrictive_ |\n| `Native Python` | \u2705 | \u2705 | \u274c |\n| `Detect spoken language` | \u274c | \u2705 | N/A |\n| `UnicodeDecodeError Safety` | \u274c | \u2705 | \u274c |\n| `Whl Size (min)` | 193.6 kB | 42 kB | ~200 kB |\n| `Supported Encoding` | 33 | \ud83c\udf89 [99](https://charset-normalizer.readthedocs.io/en/latest/user/support.html#supported-encodings) | 40 |\n\n

\n\"Reading\"Cat\n

\n\n*\\*\\* : They are clearly using specific code for a specific encoding even if covering most of used one*
\n\n## \u26a1 Performance\n\nThis package offer better performance than its counterpart Chardet. Here are some numbers.\n\n| Package | Accuracy | Mean per file (ms) | File per sec (est) |\n|-----------------------------------------------|:--------:|:------------------:|:------------------:|\n| [chardet](https://github.com/chardet/chardet) | 86 % | 63 ms | 16 file/sec |\n| charset-normalizer | **98 %** | **10 ms** | 100 file/sec |\n\n| Package | 99th percentile | 95th percentile | 50th percentile |\n|-----------------------------------------------|:---------------:|:---------------:|:---------------:|\n| [chardet](https://github.com/chardet/chardet) | 265 ms | 71 ms | 7 ms |\n| charset-normalizer | 100 ms | 50 ms | 5 ms |\n\n_updated as of december 2024 using CPython 3.12_\n\nChardet's performance on larger file (1MB+) are very poor. Expect huge difference on large payload.\n\n> Stats are generated using 400+ files using default parameters. More details on used files, see GHA workflows.\n> And yes, these results might change at any time. The dataset can be updated to include more files.\n> The actual delays heavily depends on your CPU capabilities. The factors should remain the same.\n> Keep in mind that the stats are generous and that Chardet accuracy vs our is measured using Chardet initial capability\n> (e.g. Supported Encoding) Challenge-them if you want.\n\n## \u2728 Installation\n\nUsing pip:\n\n```sh\npip install charset-normalizer -U\n```\n\n## \ud83d\ude80 Basic Usage\n\n### CLI\nThis package comes with a CLI.\n\n```\nusage: normalizer [-h] [-v] [-a] [-n] [-m] [-r] [-f] [-t THRESHOLD]\n file [file ...]\n\nThe Real First Universal Charset Detector. Discover originating encoding used\non text file. Normalize text to unicode.\n\npositional arguments:\n files File(s) to be analysed\n\noptional arguments:\n -h, --help show this help message and exit\n -v, --verbose Display complementary information about file if any.\n Stdout will contain logs about the detection process.\n -a, --with-alternative\n Output complementary possibilities if any. Top-level\n JSON WILL be a list.\n -n, --normalize Permit to normalize input file. If not set, program\n does not write anything.\n -m, --minimal Only output the charset detected to STDOUT. Disabling\n JSON output.\n -r, --replace Replace file when trying to normalize it instead of\n creating a new one.\n -f, --force Replace file without asking if you are sure, use this\n flag with caution.\n -t THRESHOLD, --threshold THRESHOLD\n Define a custom maximum amount of chaos allowed in\n decoded content. 0. <= chaos <= 1.\n --version Show version information and exit.\n```\n\n```bash\nnormalizer ./data/sample.1.fr.srt\n```\n\nor\n\n```bash\npython -m charset_normalizer ./data/sample.1.fr.srt\n```\n\n\ud83c\udf89 Since version 1.4.0 the CLI produce easily usable stdout result in JSON format.\n\n```json\n{\n \"path\": \"/home/default/projects/charset_normalizer/data/sample.1.fr.srt\",\n \"encoding\": \"cp1252\",\n \"encoding_aliases\": [\n \"1252\",\n \"windows_1252\"\n ],\n \"alternative_encodings\": [\n \"cp1254\",\n \"cp1256\",\n \"cp1258\",\n \"iso8859_14\",\n \"iso8859_15\",\n \"iso8859_16\",\n \"iso8859_3\",\n \"iso8859_9\",\n \"latin_1\",\n \"mbcs\"\n ],\n \"language\": \"French\",\n \"alphabets\": [\n \"Basic Latin\",\n \"Latin-1 Supplement\"\n ],\n \"has_sig_or_bom\": false,\n \"chaos\": 0.149,\n \"coherence\": 97.152,\n \"unicode_path\": null,\n \"is_preferred\": true\n}\n```\n\n### Python\n*Just print out normalized text*\n```python\nfrom charset_normalizer import from_path\n\nresults = from_path('./my_subtitle.srt')\n\nprint(str(results.best()))\n```\n\n*Upgrade your code without effort*\n```python\nfrom charset_normalizer import detect\n```\n\nThe above code will behave the same as **chardet**. We ensure that we offer the best (reasonable) BC result possible.\n\nSee the docs for advanced usage : [readthedocs.io](https://charset-normalizer.readthedocs.io/en/latest/)\n\n## \ud83d\ude07 Why\n\nWhen I started using Chardet, I noticed that it was not suited to my expectations, and I wanted to propose a\nreliable alternative using a completely different method. Also! I never back down on a good challenge!\n\nI **don't care** about the **originating charset** encoding, because **two different tables** can\nproduce **two identical rendered string.**\nWhat I want is to get readable text, the best I can.\n\nIn a way, **I'm brute forcing text decoding.** How cool is that ? \ud83d\ude0e\n\nDon't confuse package **ftfy** with charset-normalizer or chardet. ftfy goal is to repair Unicode string whereas charset-normalizer to convert raw file in unknown encoding to unicode.\n\n## \ud83c\udf70 How\n\n - Discard all charset encoding table that could not fit the binary content.\n - Measure noise, or the mess once opened (by chunks) with a corresponding charset encoding.\n - Extract matches with the lowest mess detected.\n - Additionally, we measure coherence / probe for a language.\n\n**Wait a minute**, what is noise/mess and coherence according to **YOU ?**\n\n*Noise :* I opened hundred of text files, **written by humans**, with the wrong encoding table. **I observed**, then\n**I established** some ground rules about **what is obvious** when **it seems like** a mess (aka. defining noise in rendered text).\n I know that my interpretation of what is noise is probably incomplete, feel free to contribute in order to\n improve or rewrite it.\n\n*Coherence :* For each language there is on earth, we have computed ranked letter appearance occurrences (the best we can). So I thought\nthat intel is worth something here. So I use those records against decoded text to check if I can detect intelligent design.\n\n## \u26a1 Known limitations\n\n - Language detection is unreliable when text contains two or more languages sharing identical letters. (eg. HTML (english tags) + Turkish content (Sharing Latin characters))\n - Every charset detector heavily depends on sufficient content. In common cases, do not bother run detection on very tiny content.\n\n## \u26a0\ufe0f About Python EOLs\n\n**If you are running:**\n\n- Python >=2.7,<3.5: Unsupported\n- Python 3.5: charset-normalizer < 2.1\n- Python 3.6: charset-normalizer < 3.1\n- Python 3.7: charset-normalizer < 4.0\n\nUpgrade your Python interpreter as soon as possible.\n\n## \ud83d\udc64 Contributing\n\nContributions, issues and feature requests are very much welcome.
\nFeel free to check [issues page](https://github.com/ousret/charset_normalizer/issues) if you want to contribute.\n\n## \ud83d\udcdd License\n\nCopyright \u00a9 [Ahmed TAHRI @Ousret](https://github.com/Ousret).
\nThis project is [MIT](https://github.com/Ousret/charset_normalizer/blob/master/LICENSE) licensed.\n\nCharacters frequencies used in this project \u00a9 2012 [Denny Vrande\u010di\u0107](http://simia.net/letters/)\n\n## \ud83d\udcbc For Enterprise\n\nProfessional support for charset-normalizer is available as part of the [Tidelift\nSubscription][1]. Tidelift gives software development teams a single source for\npurchasing and maintaining their software, with professional grade assurances\nfrom the experts who know it best, while seamlessly integrating with existing\ntools.\n\n[1]: https://tidelift.com/subscription/pkg/pypi-charset-normalizer?utm_source=pypi-charset-normalizer&utm_medium=readme\n\n[![OpenSSF Best Practices](https://www.bestpractices.dev/projects/7297/badge)](https://www.bestpractices.dev/projects/7297)\n\n# Changelog\nAll notable changes to charset-normalizer will be documented in this file. This project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html).\nThe format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/).\n\n## [3.4.1](https://github.com/Ousret/charset_normalizer/compare/3.4.0...3.4.1) (2024-12-24)\n\n### Changed\n- Project metadata are now stored using `pyproject.toml` instead of `setup.cfg` using setuptools as the build backend.\n- Enforce annotation delayed loading for a simpler and consistent types in the project.\n- Optional mypyc compilation upgraded to version 1.14 for Python >= 3.8\n\n### Added\n- pre-commit configuration.\n- noxfile.\n\n### Removed\n- `build-requirements.txt` as per using `pyproject.toml` native build configuration.\n- `bin/integration.py` and `bin/serve.py` in favor of downstream integration test (see noxfile).\n- `setup.cfg` in favor of `pyproject.toml` metadata configuration.\n- Unused `utils.range_scan` function.\n\n### Fixed\n- Converting content to Unicode bytes may insert `utf_8` instead of preferred `utf-8`. (#572)\n- Deprecation warning \"'count' is passed as positional argument\" when converting to Unicode bytes on Python 3.13+\n\n## [3.4.0](https://github.com/Ousret/charset_normalizer/compare/3.3.2...3.4.0) (2024-10-08)\n\n### Added\n- Argument `--no-preemptive` in the CLI to prevent the detector to search for hints.\n- Support for Python 3.13 (#512)\n\n### Fixed\n- Relax the TypeError exception thrown when trying to compare a CharsetMatch with anything else than a CharsetMatch.\n- Improved the general reliability of the detector based on user feedbacks. (#520) (#509) (#498) (#407) (#537)\n- Declared charset in content (preemptive detection) not changed when converting to utf-8 bytes. (#381)\n\n## [3.3.2](https://github.com/Ousret/charset_normalizer/compare/3.3.1...3.3.2) (2023-10-31)\n\n### Fixed\n- Unintentional memory usage regression when using large payload that match several encoding (#376)\n- Regression on some detection case showcased in the documentation (#371)\n\n### Added\n- Noise (md) probe that identify malformed arabic representation due to the presence of letters in isolated form (credit to my wife)\n\n## [3.3.1](https://github.com/Ousret/charset_normalizer/compare/3.3.0...3.3.1) (2023-10-22)\n\n### Changed\n- Optional mypyc compilation upgraded to version 1.6.1 for Python >= 3.8\n- Improved the general detection reliability based on reports from the community\n\n## [3.3.0](https://github.com/Ousret/charset_normalizer/compare/3.2.0...3.3.0) (2023-09-30)\n\n### Added\n- Allow to execute the CLI (e.g. normalizer) through `python -m charset_normalizer.cli` or `python -m charset_normalizer`\n- Support for 9 forgotten encoding that are supported by Python but unlisted in `encoding.aliases` as they have no alias (#323)\n\n### Removed\n- (internal) Redundant utils.is_ascii function and unused function is_private_use_only\n- (internal) charset_normalizer.assets is moved inside charset_normalizer.constant\n\n### Changed\n- (internal) Unicode code blocks in constants are updated using the latest v15.0.0 definition to improve detection\n- Optional mypyc compilation upgraded to version 1.5.1 for Python >= 3.8\n\n### Fixed\n- Unable to properly sort CharsetMatch when both chaos/noise and coherence were close due to an unreachable condition in \\_\\_lt\\_\\_ (#350)\n\n## [3.2.0](https://github.com/Ousret/charset_normalizer/compare/3.1.0...3.2.0) (2023-06-07)\n\n### Changed\n- Typehint for function `from_path` no longer enforce `PathLike` as its first argument\n- Minor improvement over the global detection reliability\n\n### Added\n- Introduce function `is_binary` that relies on main capabilities, and optimized to detect binaries\n- Propagate `enable_fallback` argument throughout `from_bytes`, `from_path`, and `from_fp` that allow a deeper control over the detection (default True)\n- Explicit support for Python 3.12\n\n### Fixed\n- Edge case detection failure where a file would contain 'very-long' camel cased word (Issue #289)\n\n## [3.1.0](https://github.com/Ousret/charset_normalizer/compare/3.0.1...3.1.0) (2023-03-06)\n\n### Added\n- Argument `should_rename_legacy` for legacy function `detect` and disregard any new arguments without errors (PR #262)\n\n### Removed\n- Support for Python 3.6 (PR #260)\n\n### Changed\n- Optional speedup provided by mypy/c 1.0.1\n\n## [3.0.1](https://github.com/Ousret/charset_normalizer/compare/3.0.0...3.0.1) (2022-11-18)\n\n### Fixed\n- Multi-bytes cutter/chunk generator did not always cut correctly (PR #233)\n\n### Changed\n- Speedup provided by mypy/c 0.990 on Python >= 3.7\n\n## [3.0.0](https://github.com/Ousret/charset_normalizer/compare/2.1.1...3.0.0) (2022-10-20)\n\n### Added\n- Extend the capability of explain=True when cp_isolation contains at most two entries (min one), will log in details of the Mess-detector results\n- Support for alternative language frequency set in charset_normalizer.assets.FREQUENCIES\n- Add parameter `language_threshold` in `from_bytes`, `from_path` and `from_fp` to adjust the minimum expected coherence ratio\n- `normalizer --version` now specify if current version provide extra speedup (meaning mypyc compilation whl)\n\n### Changed\n- Build with static metadata using 'build' frontend\n- Make the language detection stricter\n- Optional: Module `md.py` can be compiled using Mypyc to provide an extra speedup up to 4x faster than v2.1\n\n### Fixed\n- CLI with opt --normalize fail when using full path for files\n- TooManyAccentuatedPlugin induce false positive on the mess detection when too few alpha character have been fed to it\n- Sphinx warnings when generating the documentation\n\n### Removed\n- Coherence detector no longer return 'Simple English' instead return 'English'\n- Coherence detector no longer return 'Classical Chinese' instead return 'Chinese'\n- Breaking: Method `first()` and `best()` from CharsetMatch\n- UTF-7 will no longer appear as \"detected\" without a recognized SIG/mark (is unreliable/conflict with ASCII)\n- Breaking: Class aliases CharsetDetector, CharsetDoctor, CharsetNormalizerMatch and CharsetNormalizerMatches\n- Breaking: Top-level function `normalize`\n- Breaking: Properties `chaos_secondary_pass`, `coherence_non_latin` and `w_counter` from CharsetMatch\n- Support for the backport `unicodedata2`\n\n## [3.0.0rc1](https://github.com/Ousret/charset_normalizer/compare/3.0.0b2...3.0.0rc1) (2022-10-18)\n\n### Added\n- Extend the capability of explain=True when cp_isolation contains at most two entries (min one), will log in details of the Mess-detector results\n- Support for alternative language frequency set in charset_normalizer.assets.FREQUENCIES\n- Add parameter `language_threshold` in `from_bytes`, `from_path` and `from_fp` to adjust the minimum expected coherence ratio\n\n### Changed\n- Build with static metadata using 'build' frontend\n- Make the language detection stricter\n\n### Fixed\n- CLI with opt --normalize fail when using full path for files\n- TooManyAccentuatedPlugin induce false positive on the mess detection when too few alpha character have been fed to it\n\n### Removed\n- Coherence detector no longer return 'Simple English' instead return 'English'\n- Coherence detector no longer return 'Classical Chinese' instead return 'Chinese'\n\n## [3.0.0b2](https://github.com/Ousret/charset_normalizer/compare/3.0.0b1...3.0.0b2) (2022-08-21)\n\n### Added\n- `normalizer --version` now specify if current version provide extra speedup (meaning mypyc compilation whl)\n\n### Removed\n- Breaking: Method `first()` and `best()` from CharsetMatch\n- UTF-7 will no longer appear as \"detected\" without a recognized SIG/mark (is unreliable/conflict with ASCII)\n\n### Fixed\n- Sphinx warnings when generating the documentation\n\n## [3.0.0b1](https://github.com/Ousret/charset_normalizer/compare/2.1.0...3.0.0b1) (2022-08-15)\n\n### Changed\n- Optional: Module `md.py` can be compiled using Mypyc to provide an extra speedup up to 4x faster than v2.1\n\n### Removed\n- Breaking: Class aliases CharsetDetector, CharsetDoctor, CharsetNormalizerMatch and CharsetNormalizerMatches\n- Breaking: Top-level function `normalize`\n- Breaking: Properties `chaos_secondary_pass`, `coherence_non_latin` and `w_counter` from CharsetMatch\n- Support for the backport `unicodedata2`\n\n## [2.1.1](https://github.com/Ousret/charset_normalizer/compare/2.1.0...2.1.1) (2022-08-19)\n\n### Deprecated\n- Function `normalize` scheduled for removal in 3.0\n\n### Changed\n- Removed useless call to decode in fn is_unprintable (#206)\n\n### Fixed\n- Third-party library (i18n xgettext) crashing not recognizing utf_8 (PEP 263) with underscore from [@aleksandernovikov](https://github.com/aleksandernovikov) (#204)\n\n## [2.1.0](https://github.com/Ousret/charset_normalizer/compare/2.0.12...2.1.0) (2022-06-19)\n\n### Added\n- Output the Unicode table version when running the CLI with `--version` (PR #194)\n\n### Changed\n- Re-use decoded buffer for single byte character sets from [@nijel](https://github.com/nijel) (PR #175)\n- Fixing some performance bottlenecks from [@deedy5](https://github.com/deedy5) (PR #183)\n\n### Fixed\n- Workaround potential bug in cpython with Zero Width No-Break Space located in Arabic Presentation Forms-B, Unicode 1.1 not acknowledged as space (PR #175)\n- CLI default threshold aligned with the API threshold from [@oleksandr-kuzmenko](https://github.com/oleksandr-kuzmenko) (PR #181)\n\n### Removed\n- Support for Python 3.5 (PR #192)\n\n### Deprecated\n- Use of backport unicodedata from `unicodedata2` as Python is quickly catching up, scheduled for removal in 3.0 (PR #194)\n\n## [2.0.12](https://github.com/Ousret/charset_normalizer/compare/2.0.11...2.0.12) (2022-02-12)\n\n### Fixed\n- ASCII miss-detection on rare cases (PR #170)\n\n## [2.0.11](https://github.com/Ousret/charset_normalizer/compare/2.0.10...2.0.11) (2022-01-30)\n\n### Added\n- Explicit support for Python 3.11 (PR #164)\n\n### Changed\n- The logging behavior have been completely reviewed, now using only TRACE and DEBUG levels (PR #163 #165)\n\n## [2.0.10](https://github.com/Ousret/charset_normalizer/compare/2.0.9...2.0.10) (2022-01-04)\n\n### Fixed\n- Fallback match entries might lead to UnicodeDecodeError for large bytes sequence (PR #154)\n\n### Changed\n- Skipping the language-detection (CD) on ASCII (PR #155)\n\n## [2.0.9](https://github.com/Ousret/charset_normalizer/compare/2.0.8...2.0.9) (2021-12-03)\n\n### Changed\n- Moderating the logging impact (since 2.0.8) for specific environments (PR #147)\n\n### Fixed\n- Wrong logging level applied when setting kwarg `explain` to True (PR #146)\n\n## [2.0.8](https://github.com/Ousret/charset_normalizer/compare/2.0.7...2.0.8) (2021-11-24)\n### Changed\n- Improvement over Vietnamese detection (PR #126)\n- MD improvement on trailing data and long foreign (non-pure latin) data (PR #124)\n- Efficiency improvements in cd/alphabet_languages from [@adbar](https://github.com/adbar) (PR #122)\n- call sum() without an intermediary list following PEP 289 recommendations from [@adbar](https://github.com/adbar) (PR #129)\n- Code style as refactored by Sourcery-AI (PR #131)\n- Minor adjustment on the MD around european words (PR #133)\n- Remove and replace SRTs from assets / tests (PR #139)\n- Initialize the library logger with a `NullHandler` by default from [@nmaynes](https://github.com/nmaynes) (PR #135)\n- Setting kwarg `explain` to True will add provisionally (bounded to function lifespan) a specific stream handler (PR #135)\n\n### Fixed\n- Fix large (misleading) sequence giving UnicodeDecodeError (PR #137)\n- Avoid using too insignificant chunk (PR #137)\n\n### Added\n- Add and expose function `set_logging_handler` to configure a specific StreamHandler from [@nmaynes](https://github.com/nmaynes) (PR #135)\n- Add `CHANGELOG.md` entries, format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/) (PR #141)\n\n## [2.0.7](https://github.com/Ousret/charset_normalizer/compare/2.0.6...2.0.7) (2021-10-11)\n### Added\n- Add support for Kazakh (Cyrillic) language detection (PR #109)\n\n### Changed\n- Further, improve inferring the language from a given single-byte code page (PR #112)\n- Vainly trying to leverage PEP263 when PEP3120 is not supported (PR #116)\n- Refactoring for potential performance improvements in loops from [@adbar](https://github.com/adbar) (PR #113)\n- Various detection improvement (MD+CD) (PR #117)\n\n### Removed\n- Remove redundant logging entry about detected language(s) (PR #115)\n\n### Fixed\n- Fix a minor inconsistency between Python 3.5 and other versions regarding language detection (PR #117 #102)\n\n## [2.0.6](https://github.com/Ousret/charset_normalizer/compare/2.0.5...2.0.6) (2021-09-18)\n### Fixed\n- Unforeseen regression with the loss of the backward-compatibility with some older minor of Python 3.5.x (PR #100)\n- Fix CLI crash when using --minimal output in certain cases (PR #103)\n\n### Changed\n- Minor improvement to the detection efficiency (less than 1%) (PR #106 #101)\n\n## [2.0.5](https://github.com/Ousret/charset_normalizer/compare/2.0.4...2.0.5) (2021-09-14)\n### Changed\n- The project now comply with: flake8, mypy, isort and black to ensure a better overall quality (PR #81)\n- The BC-support with v1.x was improved, the old staticmethods are restored (PR #82)\n- The Unicode detection is slightly improved (PR #93)\n- Add syntax sugar \\_\\_bool\\_\\_ for results CharsetMatches list-container (PR #91)\n\n### Removed\n- The project no longer raise warning on tiny content given for detection, will be simply logged as warning instead (PR #92)\n\n### Fixed\n- In some rare case, the chunks extractor could cut in the middle of a multi-byte character and could mislead the mess detection (PR #95)\n- Some rare 'space' characters could trip up the UnprintablePlugin/Mess detection (PR #96)\n- The MANIFEST.in was not exhaustive (PR #78)\n\n## [2.0.4](https://github.com/Ousret/charset_normalizer/compare/2.0.3...2.0.4) (2021-07-30)\n### Fixed\n- The CLI no longer raise an unexpected exception when no encoding has been found (PR #70)\n- Fix accessing the 'alphabets' property when the payload contains surrogate characters (PR #68)\n- The logger could mislead (explain=True) on detected languages and the impact of one MBCS match (PR #72)\n- Submatch factoring could be wrong in rare edge cases (PR #72)\n- Multiple files given to the CLI were ignored when publishing results to STDOUT. (After the first path) (PR #72)\n- Fix line endings from CRLF to LF for certain project files (PR #67)\n\n### Changed\n- Adjust the MD to lower the sensitivity, thus improving the global detection reliability (PR #69 #76)\n- Allow fallback on specified encoding if any (PR #71)\n\n## [2.0.3](https://github.com/Ousret/charset_normalizer/compare/2.0.2...2.0.3) (2021-07-16)\n### Changed\n- Part of the detection mechanism has been improved to be less sensitive, resulting in more accurate detection results. Especially ASCII. (PR #63)\n- According to the community wishes, the detection will fall back on ASCII or UTF-8 in a last-resort case. (PR #64)\n\n## [2.0.2](https://github.com/Ousret/charset_normalizer/compare/2.0.1...2.0.2) (2021-07-15)\n### Fixed\n- Empty/Too small JSON payload miss-detection fixed. Report from [@tseaver](https://github.com/tseaver) (PR #59)\n\n### Changed\n- Don't inject unicodedata2 into sys.modules from [@akx](https://github.com/akx) (PR #57)\n\n## [2.0.1](https://github.com/Ousret/charset_normalizer/compare/2.0.0...2.0.1) (2021-07-13)\n### Fixed\n- Make it work where there isn't a filesystem available, dropping assets frequencies.json. Report from [@sethmlarson](https://github.com/sethmlarson). (PR #55)\n- Using explain=False permanently disable the verbose output in the current runtime (PR #47)\n- One log entry (language target preemptive) was not show in logs when using explain=True (PR #47)\n- Fix undesired exception (ValueError) on getitem of instance CharsetMatches (PR #52)\n\n### Changed\n- Public function normalize default args values were not aligned with from_bytes (PR #53)\n\n### Added\n- You may now use charset aliases in cp_isolation and cp_exclusion arguments (PR #47)\n\n## [2.0.0](https://github.com/Ousret/charset_normalizer/compare/1.4.1...2.0.0) (2021-07-02)\n### Changed\n- 4x to 5 times faster than the previous 1.4.0 release. At least 2x faster than Chardet.\n- Accent has been made on UTF-8 detection, should perform rather instantaneous.\n- The backward compatibility with Chardet has been greatly improved. The legacy detect function returns an identical charset name whenever possible.\n- The detection mechanism has been slightly improved, now Turkish content is detected correctly (most of the time)\n- The program has been rewritten to ease the readability and maintainability. (+Using static typing)+\n- utf_7 detection has been reinstated.\n\n### Removed\n- This package no longer require anything when used with Python 3.5 (Dropped cached_property)\n- Removed support for these languages: Catalan, Esperanto, Kazakh, Baque, Volap\u00fck, Azeri, Galician, Nynorsk, Macedonian, and Serbocroatian.\n- The exception hook on UnicodeDecodeError has been removed.\n\n### Deprecated\n- Methods coherence_non_latin, w_counter, chaos_secondary_pass of the class CharsetMatch are now deprecated and scheduled for removal in v3.0\n\n### Fixed\n- The CLI output used the relative path of the file(s). Should be absolute.\n\n## [1.4.1](https://github.com/Ousret/charset_normalizer/compare/1.4.0...1.4.1) (2021-05-28)\n### Fixed\n- Logger configuration/usage no longer conflict with others (PR #44)\n\n## [1.4.0](https://github.com/Ousret/charset_normalizer/compare/1.3.9...1.4.0) (2021-05-21)\n### Removed\n- Using standard logging instead of using the package loguru.\n- Dropping nose test framework in favor of the maintained pytest.\n- Choose to not use dragonmapper package to help with gibberish Chinese/CJK text.\n- Require cached_property only for Python 3.5 due to constraint. Dropping for every other interpreter version.\n- Stop support for UTF-7 that does not contain a SIG.\n- Dropping PrettyTable, replaced with pure JSON output in CLI.\n\n### Fixed\n- BOM marker in a CharsetNormalizerMatch instance could be False in rare cases even if obviously present. Due to the sub-match factoring process.\n- Not searching properly for the BOM when trying utf32/16 parent codec.\n\n### Changed\n- Improving the package final size by compressing frequencies.json.\n- Huge improvement over the larges payload.\n\n### Added\n- CLI now produces JSON consumable output.\n- Return ASCII if given sequences fit. Given reasonable confidence.\n\n## [1.3.9](https://github.com/Ousret/charset_normalizer/compare/1.3.8...1.3.9) (2021-05-13)\n\n### Fixed\n- In some very rare cases, you may end up getting encode/decode errors due to a bad bytes payload (PR #40)\n\n## [1.3.8](https://github.com/Ousret/charset_normalizer/compare/1.3.7...1.3.8) (2021-05-12)\n\n### Fixed\n- Empty given payload for detection may cause an exception if trying to access the `alphabets` property. (PR #39)\n\n## [1.3.7](https://github.com/Ousret/charset_normalizer/compare/1.3.6...1.3.7) (2021-05-12)\n\n### Fixed\n- The legacy detect function should return UTF-8-SIG if sig is present in the payload. (PR #38)\n\n## [1.3.6](https://github.com/Ousret/charset_normalizer/compare/1.3.5...1.3.6) (2021-02-09)\n\n### Changed\n- Amend the previous release to allow prettytable 2.0 (PR #35)\n\n## [1.3.5](https://github.com/Ousret/charset_normalizer/compare/1.3.4...1.3.5) (2021-02-08)\n\n### Fixed\n- Fix error while using the package with a python pre-release interpreter (PR #33)\n\n### Changed\n- Dependencies refactoring, constraints revised.\n\n### Added\n- Add python 3.9 and 3.10 to the supported interpreters\n\nMIT License\n\nCopyright (c) 2025 TAHRI Ahmed R.\n\nPermission is hereby granted, free of charge, to any person obtaining a copy\nof this software and associated documentation files (the \"Software\"), to deal\nin the Software without restriction, including without limitation the rights\nto use, copy, modify, merge, publish, distribute, sublicense, and/or sell\ncopies of the Software, and to permit persons to whom the Software is\nfurnished to do so, subject to the following conditions:\n\nThe above copyright notice and this permission notice shall be included in all\ncopies or substantial portions of the Software.\n\nTHE SOFTWARE IS PROVIDED \"AS IS\", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR\nIMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,\nFITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE\nAUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER\nLIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,\nOUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE\nSOFTWARE.", - "release_date": "2024-12-24T18:12:06", + "description": "The Real First Universal Charset Detector. Open, modern and actively maintained alternative to Chardet.\n

Charset Detection, for Everyone \ud83d\udc4b

\n\n

\n The Real First Universal Charset Detector
\n \n \n \n \n \"Download\n \n \n \n \n

\n

\n Featured Packages
\n \n \"Static\n \n \n \"Static\n \n

\n

\n In other language (unofficial port - by the community)
\n \n \"Static\n \n

\n\n> A library that helps you read text from an unknown charset encoding.
Motivated by `chardet`,\n> I'm trying to resolve the issue by taking a new approach.\n> All IANA character set names for which the Python core library provides codecs are supported.\n\n

\n >>>>> \ud83d\udc49 Try Me Online Now, Then Adopt Me \ud83d\udc48 <<<<<\n

\n\nThis project offers you an alternative to **Universal Charset Encoding Detector**, also known as **Chardet**.\n\n| Feature | [Chardet](https://github.com/chardet/chardet) | Charset Normalizer | [cChardet](https://github.com/PyYoshi/cChardet) |\n|--------------------------------------------------|:---------------------------------------------:|:--------------------------------------------------------------------------------------------------:|:-----------------------------------------------:|\n| `Fast` | \u274c | \u2705 | \u2705 |\n| `Universal**` | \u274c | \u2705 | \u274c |\n| `Reliable` **without** distinguishable standards | \u274c | \u2705 | \u2705 |\n| `Reliable` **with** distinguishable standards | \u2705 | \u2705 | \u2705 |\n| `License` | LGPL-2.1
_restrictive_ | MIT | MPL-1.1
_restrictive_ |\n| `Native Python` | \u2705 | \u2705 | \u274c |\n| `Detect spoken language` | \u274c | \u2705 | N/A |\n| `UnicodeDecodeError Safety` | \u274c | \u2705 | \u274c |\n| `Whl Size (min)` | 193.6 kB | 42 kB | ~200 kB |\n| `Supported Encoding` | 33 | \ud83c\udf89 [99](https://charset-normalizer.readthedocs.io/en/latest/user/support.html#supported-encodings) | 40 |\n\n

\n\"Reading\"Cat\n

\n\n*\\*\\* : They are clearly using specific code for a specific encoding even if covering most of used one*
\n\n## \u26a1 Performance\n\nThis package offer better performance than its counterpart Chardet. Here are some numbers.\n\n| Package | Accuracy | Mean per file (ms) | File per sec (est) |\n|-----------------------------------------------|:--------:|:------------------:|:------------------:|\n| [chardet](https://github.com/chardet/chardet) | 86 % | 63 ms | 16 file/sec |\n| charset-normalizer | **98 %** | **10 ms** | 100 file/sec |\n\n| Package | 99th percentile | 95th percentile | 50th percentile |\n|-----------------------------------------------|:---------------:|:---------------:|:---------------:|\n| [chardet](https://github.com/chardet/chardet) | 265 ms | 71 ms | 7 ms |\n| charset-normalizer | 100 ms | 50 ms | 5 ms |\n\n_updated as of december 2024 using CPython 3.12_\n\nChardet's performance on larger file (1MB+) are very poor. Expect huge difference on large payload.\n\n> Stats are generated using 400+ files using default parameters. More details on used files, see GHA workflows.\n> And yes, these results might change at any time. The dataset can be updated to include more files.\n> The actual delays heavily depends on your CPU capabilities. The factors should remain the same.\n> Keep in mind that the stats are generous and that Chardet accuracy vs our is measured using Chardet initial capability\n> (e.g. Supported Encoding) Challenge-them if you want.\n\n## \u2728 Installation\n\nUsing pip:\n\n```sh\npip install charset-normalizer -U\n```\n\n## \ud83d\ude80 Basic Usage\n\n### CLI\nThis package comes with a CLI.\n\n```\nusage: normalizer [-h] [-v] [-a] [-n] [-m] [-r] [-f] [-t THRESHOLD]\n file [file ...]\n\nThe Real First Universal Charset Detector. Discover originating encoding used\non text file. Normalize text to unicode.\n\npositional arguments:\n files File(s) to be analysed\n\noptional arguments:\n -h, --help show this help message and exit\n -v, --verbose Display complementary information about file if any.\n Stdout will contain logs about the detection process.\n -a, --with-alternative\n Output complementary possibilities if any. Top-level\n JSON WILL be a list.\n -n, --normalize Permit to normalize input file. If not set, program\n does not write anything.\n -m, --minimal Only output the charset detected to STDOUT. Disabling\n JSON output.\n -r, --replace Replace file when trying to normalize it instead of\n creating a new one.\n -f, --force Replace file without asking if you are sure, use this\n flag with caution.\n -t THRESHOLD, --threshold THRESHOLD\n Define a custom maximum amount of chaos allowed in\n decoded content. 0. <= chaos <= 1.\n --version Show version information and exit.\n```\n\n```bash\nnormalizer ./data/sample.1.fr.srt\n```\n\nor\n\n```bash\npython -m charset_normalizer ./data/sample.1.fr.srt\n```\n\n\ud83c\udf89 Since version 1.4.0 the CLI produce easily usable stdout result in JSON format.\n\n```json\n{\n \"path\": \"/home/default/projects/charset_normalizer/data/sample.1.fr.srt\",\n \"encoding\": \"cp1252\",\n \"encoding_aliases\": [\n \"1252\",\n \"windows_1252\"\n ],\n \"alternative_encodings\": [\n \"cp1254\",\n \"cp1256\",\n \"cp1258\",\n \"iso8859_14\",\n \"iso8859_15\",\n \"iso8859_16\",\n \"iso8859_3\",\n \"iso8859_9\",\n \"latin_1\",\n \"mbcs\"\n ],\n \"language\": \"French\",\n \"alphabets\": [\n \"Basic Latin\",\n \"Latin-1 Supplement\"\n ],\n \"has_sig_or_bom\": false,\n \"chaos\": 0.149,\n \"coherence\": 97.152,\n \"unicode_path\": null,\n \"is_preferred\": true\n}\n```\n\n### Python\n*Just print out normalized text*\n```python\nfrom charset_normalizer import from_path\n\nresults = from_path('./my_subtitle.srt')\n\nprint(str(results.best()))\n```\n\n*Upgrade your code without effort*\n```python\nfrom charset_normalizer import detect\n```\n\nThe above code will behave the same as **chardet**. We ensure that we offer the best (reasonable) BC result possible.\n\nSee the docs for advanced usage : [readthedocs.io](https://charset-normalizer.readthedocs.io/en/latest/)\n\n## \ud83d\ude07 Why\n\nWhen I started using Chardet, I noticed that it was not suited to my expectations, and I wanted to propose a\nreliable alternative using a completely different method. Also! I never back down on a good challenge!\n\nI **don't care** about the **originating charset** encoding, because **two different tables** can\nproduce **two identical rendered string.**\nWhat I want is to get readable text, the best I can.\n\nIn a way, **I'm brute forcing text decoding.** How cool is that ? \ud83d\ude0e\n\nDon't confuse package **ftfy** with charset-normalizer or chardet. ftfy goal is to repair Unicode string whereas charset-normalizer to convert raw file in unknown encoding to unicode.\n\n## \ud83c\udf70 How\n\n - Discard all charset encoding table that could not fit the binary content.\n - Measure noise, or the mess once opened (by chunks) with a corresponding charset encoding.\n - Extract matches with the lowest mess detected.\n - Additionally, we measure coherence / probe for a language.\n\n**Wait a minute**, what is noise/mess and coherence according to **YOU ?**\n\n*Noise :* I opened hundred of text files, **written by humans**, with the wrong encoding table. **I observed**, then\n**I established** some ground rules about **what is obvious** when **it seems like** a mess (aka. defining noise in rendered text).\n I know that my interpretation of what is noise is probably incomplete, feel free to contribute in order to\n improve or rewrite it.\n\n*Coherence :* For each language there is on earth, we have computed ranked letter appearance occurrences (the best we can). So I thought\nthat intel is worth something here. So I use those records against decoded text to check if I can detect intelligent design.\n\n## \u26a1 Known limitations\n\n - Language detection is unreliable when text contains two or more languages sharing identical letters. (eg. HTML (english tags) + Turkish content (Sharing Latin characters))\n - Every charset detector heavily depends on sufficient content. In common cases, do not bother run detection on very tiny content.\n\n## \u26a0\ufe0f About Python EOLs\n\n**If you are running:**\n\n- Python >=2.7,<3.5: Unsupported\n- Python 3.5: charset-normalizer < 2.1\n- Python 3.6: charset-normalizer < 3.1\n- Python 3.7: charset-normalizer < 4.0\n\nUpgrade your Python interpreter as soon as possible.\n\n## \ud83d\udc64 Contributing\n\nContributions, issues and feature requests are very much welcome.
\nFeel free to check [issues page](https://github.com/ousret/charset_normalizer/issues) if you want to contribute.\n\n## \ud83d\udcdd License\n\nCopyright \u00a9 [Ahmed TAHRI @Ousret](https://github.com/Ousret).
\nThis project is [MIT](https://github.com/Ousret/charset_normalizer/blob/master/LICENSE) licensed.\n\nCharacters frequencies used in this project \u00a9 2012 [Denny Vrande\u010di\u0107](http://simia.net/letters/)\n\n## \ud83d\udcbc For Enterprise\n\nProfessional support for charset-normalizer is available as part of the [Tidelift\nSubscription][1]. Tidelift gives software development teams a single source for\npurchasing and maintaining their software, with professional grade assurances\nfrom the experts who know it best, while seamlessly integrating with existing\ntools.\n\n[1]: https://tidelift.com/subscription/pkg/pypi-charset-normalizer?utm_source=pypi-charset-normalizer&utm_medium=readme\n\n[![OpenSSF Best Practices](https://www.bestpractices.dev/projects/7297/badge)](https://www.bestpractices.dev/projects/7297)\n\n# Changelog\nAll notable changes to charset-normalizer will be documented in this file. This project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html).\nThe format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/).\n\n## [3.4.2](https://github.com/Ousret/charset_normalizer/compare/3.4.1...3.4.2) (2025-05-02)\n\n### Fixed\n- Addressed the DeprecationWarning in our CLI regarding `argparse.FileType` by backporting the target class into the package. (#591)\n- Improved the overall reliability of the detector with CJK Ideographs. (#605) (#587)\n\n### Changed\n- Optional mypyc compilation upgraded to version 1.15 for Python >= 3.8\n\n## [3.4.1](https://github.com/Ousret/charset_normalizer/compare/3.4.0...3.4.1) (2024-12-24)\n\n### Changed\n- Project metadata are now stored using `pyproject.toml` instead of `setup.cfg` using setuptools as the build backend.\n- Enforce annotation delayed loading for a simpler and consistent types in the project.\n- Optional mypyc compilation upgraded to version 1.14 for Python >= 3.8\n\n### Added\n- pre-commit configuration.\n- noxfile.\n\n### Removed\n- `build-requirements.txt` as per using `pyproject.toml` native build configuration.\n- `bin/integration.py` and `bin/serve.py` in favor of downstream integration test (see noxfile).\n- `setup.cfg` in favor of `pyproject.toml` metadata configuration.\n- Unused `utils.range_scan` function.\n\n### Fixed\n- Converting content to Unicode bytes may insert `utf_8` instead of preferred `utf-8`. (#572)\n- Deprecation warning \"'count' is passed as positional argument\" when converting to Unicode bytes on Python 3.13+\n\n## [3.4.0](https://github.com/Ousret/charset_normalizer/compare/3.3.2...3.4.0) (2024-10-08)\n\n### Added\n- Argument `--no-preemptive` in the CLI to prevent the detector to search for hints.\n- Support for Python 3.13 (#512)\n\n### Fixed\n- Relax the TypeError exception thrown when trying to compare a CharsetMatch with anything else than a CharsetMatch.\n- Improved the general reliability of the detector based on user feedbacks. (#520) (#509) (#498) (#407) (#537)\n- Declared charset in content (preemptive detection) not changed when converting to utf-8 bytes. (#381)\n\n## [3.3.2](https://github.com/Ousret/charset_normalizer/compare/3.3.1...3.3.2) (2023-10-31)\n\n### Fixed\n- Unintentional memory usage regression when using large payload that match several encoding (#376)\n- Regression on some detection case showcased in the documentation (#371)\n\n### Added\n- Noise (md) probe that identify malformed arabic representation due to the presence of letters in isolated form (credit to my wife)\n\n## [3.3.1](https://github.com/Ousret/charset_normalizer/compare/3.3.0...3.3.1) (2023-10-22)\n\n### Changed\n- Optional mypyc compilation upgraded to version 1.6.1 for Python >= 3.8\n- Improved the general detection reliability based on reports from the community\n\n## [3.3.0](https://github.com/Ousret/charset_normalizer/compare/3.2.0...3.3.0) (2023-09-30)\n\n### Added\n- Allow to execute the CLI (e.g. normalizer) through `python -m charset_normalizer.cli` or `python -m charset_normalizer`\n- Support for 9 forgotten encoding that are supported by Python but unlisted in `encoding.aliases` as they have no alias (#323)\n\n### Removed\n- (internal) Redundant utils.is_ascii function and unused function is_private_use_only\n- (internal) charset_normalizer.assets is moved inside charset_normalizer.constant\n\n### Changed\n- (internal) Unicode code blocks in constants are updated using the latest v15.0.0 definition to improve detection\n- Optional mypyc compilation upgraded to version 1.5.1 for Python >= 3.8\n\n### Fixed\n- Unable to properly sort CharsetMatch when both chaos/noise and coherence were close due to an unreachable condition in \\_\\_lt\\_\\_ (#350)\n\n## [3.2.0](https://github.com/Ousret/charset_normalizer/compare/3.1.0...3.2.0) (2023-06-07)\n\n### Changed\n- Typehint for function `from_path` no longer enforce `PathLike` as its first argument\n- Minor improvement over the global detection reliability\n\n### Added\n- Introduce function `is_binary` that relies on main capabilities, and optimized to detect binaries\n- Propagate `enable_fallback` argument throughout `from_bytes`, `from_path`, and `from_fp` that allow a deeper control over the detection (default True)\n- Explicit support for Python 3.12\n\n### Fixed\n- Edge case detection failure where a file would contain 'very-long' camel cased word (Issue #289)\n\n## [3.1.0](https://github.com/Ousret/charset_normalizer/compare/3.0.1...3.1.0) (2023-03-06)\n\n### Added\n- Argument `should_rename_legacy` for legacy function `detect` and disregard any new arguments without errors (PR #262)\n\n### Removed\n- Support for Python 3.6 (PR #260)\n\n### Changed\n- Optional speedup provided by mypy/c 1.0.1\n\n## [3.0.1](https://github.com/Ousret/charset_normalizer/compare/3.0.0...3.0.1) (2022-11-18)\n\n### Fixed\n- Multi-bytes cutter/chunk generator did not always cut correctly (PR #233)\n\n### Changed\n- Speedup provided by mypy/c 0.990 on Python >= 3.7\n\n## [3.0.0](https://github.com/Ousret/charset_normalizer/compare/2.1.1...3.0.0) (2022-10-20)\n\n### Added\n- Extend the capability of explain=True when cp_isolation contains at most two entries (min one), will log in details of the Mess-detector results\n- Support for alternative language frequency set in charset_normalizer.assets.FREQUENCIES\n- Add parameter `language_threshold` in `from_bytes`, `from_path` and `from_fp` to adjust the minimum expected coherence ratio\n- `normalizer --version` now specify if current version provide extra speedup (meaning mypyc compilation whl)\n\n### Changed\n- Build with static metadata using 'build' frontend\n- Make the language detection stricter\n- Optional: Module `md.py` can be compiled using Mypyc to provide an extra speedup up to 4x faster than v2.1\n\n### Fixed\n- CLI with opt --normalize fail when using full path for files\n- TooManyAccentuatedPlugin induce false positive on the mess detection when too few alpha character have been fed to it\n- Sphinx warnings when generating the documentation\n\n### Removed\n- Coherence detector no longer return 'Simple English' instead return 'English'\n- Coherence detector no longer return 'Classical Chinese' instead return 'Chinese'\n- Breaking: Method `first()` and `best()` from CharsetMatch\n- UTF-7 will no longer appear as \"detected\" without a recognized SIG/mark (is unreliable/conflict with ASCII)\n- Breaking: Class aliases CharsetDetector, CharsetDoctor, CharsetNormalizerMatch and CharsetNormalizerMatches\n- Breaking: Top-level function `normalize`\n- Breaking: Properties `chaos_secondary_pass`, `coherence_non_latin` and `w_counter` from CharsetMatch\n- Support for the backport `unicodedata2`\n\n## [3.0.0rc1](https://github.com/Ousret/charset_normalizer/compare/3.0.0b2...3.0.0rc1) (2022-10-18)\n\n### Added\n- Extend the capability of explain=True when cp_isolation contains at most two entries (min one), will log in details of the Mess-detector results\n- Support for alternative language frequency set in charset_normalizer.assets.FREQUENCIES\n- Add parameter `language_threshold` in `from_bytes`, `from_path` and `from_fp` to adjust the minimum expected coherence ratio\n\n### Changed\n- Build with static metadata using 'build' frontend\n- Make the language detection stricter\n\n### Fixed\n- CLI with opt --normalize fail when using full path for files\n- TooManyAccentuatedPlugin induce false positive on the mess detection when too few alpha character have been fed to it\n\n### Removed\n- Coherence detector no longer return 'Simple English' instead return 'English'\n- Coherence detector no longer return 'Classical Chinese' instead return 'Chinese'\n\n## [3.0.0b2](https://github.com/Ousret/charset_normalizer/compare/3.0.0b1...3.0.0b2) (2022-08-21)\n\n### Added\n- `normalizer --version` now specify if current version provide extra speedup (meaning mypyc compilation whl)\n\n### Removed\n- Breaking: Method `first()` and `best()` from CharsetMatch\n- UTF-7 will no longer appear as \"detected\" without a recognized SIG/mark (is unreliable/conflict with ASCII)\n\n### Fixed\n- Sphinx warnings when generating the documentation\n\n## [3.0.0b1](https://github.com/Ousret/charset_normalizer/compare/2.1.0...3.0.0b1) (2022-08-15)\n\n### Changed\n- Optional: Module `md.py` can be compiled using Mypyc to provide an extra speedup up to 4x faster than v2.1\n\n### Removed\n- Breaking: Class aliases CharsetDetector, CharsetDoctor, CharsetNormalizerMatch and CharsetNormalizerMatches\n- Breaking: Top-level function `normalize`\n- Breaking: Properties `chaos_secondary_pass`, `coherence_non_latin` and `w_counter` from CharsetMatch\n- Support for the backport `unicodedata2`\n\n## [2.1.1](https://github.com/Ousret/charset_normalizer/compare/2.1.0...2.1.1) (2022-08-19)\n\n### Deprecated\n- Function `normalize` scheduled for removal in 3.0\n\n### Changed\n- Removed useless call to decode in fn is_unprintable (#206)\n\n### Fixed\n- Third-party library (i18n xgettext) crashing not recognizing utf_8 (PEP 263) with underscore from [@aleksandernovikov](https://github.com/aleksandernovikov) (#204)\n\n## [2.1.0](https://github.com/Ousret/charset_normalizer/compare/2.0.12...2.1.0) (2022-06-19)\n\n### Added\n- Output the Unicode table version when running the CLI with `--version` (PR #194)\n\n### Changed\n- Re-use decoded buffer for single byte character sets from [@nijel](https://github.com/nijel) (PR #175)\n- Fixing some performance bottlenecks from [@deedy5](https://github.com/deedy5) (PR #183)\n\n### Fixed\n- Workaround potential bug in cpython with Zero Width No-Break Space located in Arabic Presentation Forms-B, Unicode 1.1 not acknowledged as space (PR #175)\n- CLI default threshold aligned with the API threshold from [@oleksandr-kuzmenko](https://github.com/oleksandr-kuzmenko) (PR #181)\n\n### Removed\n- Support for Python 3.5 (PR #192)\n\n### Deprecated\n- Use of backport unicodedata from `unicodedata2` as Python is quickly catching up, scheduled for removal in 3.0 (PR #194)\n\n## [2.0.12](https://github.com/Ousret/charset_normalizer/compare/2.0.11...2.0.12) (2022-02-12)\n\n### Fixed\n- ASCII miss-detection on rare cases (PR #170)\n\n## [2.0.11](https://github.com/Ousret/charset_normalizer/compare/2.0.10...2.0.11) (2022-01-30)\n\n### Added\n- Explicit support for Python 3.11 (PR #164)\n\n### Changed\n- The logging behavior have been completely reviewed, now using only TRACE and DEBUG levels (PR #163 #165)\n\n## [2.0.10](https://github.com/Ousret/charset_normalizer/compare/2.0.9...2.0.10) (2022-01-04)\n\n### Fixed\n- Fallback match entries might lead to UnicodeDecodeError for large bytes sequence (PR #154)\n\n### Changed\n- Skipping the language-detection (CD) on ASCII (PR #155)\n\n## [2.0.9](https://github.com/Ousret/charset_normalizer/compare/2.0.8...2.0.9) (2021-12-03)\n\n### Changed\n- Moderating the logging impact (since 2.0.8) for specific environments (PR #147)\n\n### Fixed\n- Wrong logging level applied when setting kwarg `explain` to True (PR #146)\n\n## [2.0.8](https://github.com/Ousret/charset_normalizer/compare/2.0.7...2.0.8) (2021-11-24)\n### Changed\n- Improvement over Vietnamese detection (PR #126)\n- MD improvement on trailing data and long foreign (non-pure latin) data (PR #124)\n- Efficiency improvements in cd/alphabet_languages from [@adbar](https://github.com/adbar) (PR #122)\n- call sum() without an intermediary list following PEP 289 recommendations from [@adbar](https://github.com/adbar) (PR #129)\n- Code style as refactored by Sourcery-AI (PR #131)\n- Minor adjustment on the MD around european words (PR #133)\n- Remove and replace SRTs from assets / tests (PR #139)\n- Initialize the library logger with a `NullHandler` by default from [@nmaynes](https://github.com/nmaynes) (PR #135)\n- Setting kwarg `explain` to True will add provisionally (bounded to function lifespan) a specific stream handler (PR #135)\n\n### Fixed\n- Fix large (misleading) sequence giving UnicodeDecodeError (PR #137)\n- Avoid using too insignificant chunk (PR #137)\n\n### Added\n- Add and expose function `set_logging_handler` to configure a specific StreamHandler from [@nmaynes](https://github.com/nmaynes) (PR #135)\n- Add `CHANGELOG.md` entries, format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/) (PR #141)\n\n## [2.0.7](https://github.com/Ousret/charset_normalizer/compare/2.0.6...2.0.7) (2021-10-11)\n### Added\n- Add support for Kazakh (Cyrillic) language detection (PR #109)\n\n### Changed\n- Further, improve inferring the language from a given single-byte code page (PR #112)\n- Vainly trying to leverage PEP263 when PEP3120 is not supported (PR #116)\n- Refactoring for potential performance improvements in loops from [@adbar](https://github.com/adbar) (PR #113)\n- Various detection improvement (MD+CD) (PR #117)\n\n### Removed\n- Remove redundant logging entry about detected language(s) (PR #115)\n\n### Fixed\n- Fix a minor inconsistency between Python 3.5 and other versions regarding language detection (PR #117 #102)\n\n## [2.0.6](https://github.com/Ousret/charset_normalizer/compare/2.0.5...2.0.6) (2021-09-18)\n### Fixed\n- Unforeseen regression with the loss of the backward-compatibility with some older minor of Python 3.5.x (PR #100)\n- Fix CLI crash when using --minimal output in certain cases (PR #103)\n\n### Changed\n- Minor improvement to the detection efficiency (less than 1%) (PR #106 #101)\n\n## [2.0.5](https://github.com/Ousret/charset_normalizer/compare/2.0.4...2.0.5) (2021-09-14)\n### Changed\n- The project now comply with: flake8, mypy, isort and black to ensure a better overall quality (PR #81)\n- The BC-support with v1.x was improved, the old staticmethods are restored (PR #82)\n- The Unicode detection is slightly improved (PR #93)\n- Add syntax sugar \\_\\_bool\\_\\_ for results CharsetMatches list-container (PR #91)\n\n### Removed\n- The project no longer raise warning on tiny content given for detection, will be simply logged as warning instead (PR #92)\n\n### Fixed\n- In some rare case, the chunks extractor could cut in the middle of a multi-byte character and could mislead the mess detection (PR #95)\n- Some rare 'space' characters could trip up the UnprintablePlugin/Mess detection (PR #96)\n- The MANIFEST.in was not exhaustive (PR #78)\n\n## [2.0.4](https://github.com/Ousret/charset_normalizer/compare/2.0.3...2.0.4) (2021-07-30)\n### Fixed\n- The CLI no longer raise an unexpected exception when no encoding has been found (PR #70)\n- Fix accessing the 'alphabets' property when the payload contains surrogate characters (PR #68)\n- The logger could mislead (explain=True) on detected languages and the impact of one MBCS match (PR #72)\n- Submatch factoring could be wrong in rare edge cases (PR #72)\n- Multiple files given to the CLI were ignored when publishing results to STDOUT. (After the first path) (PR #72)\n- Fix line endings from CRLF to LF for certain project files (PR #67)\n\n### Changed\n- Adjust the MD to lower the sensitivity, thus improving the global detection reliability (PR #69 #76)\n- Allow fallback on specified encoding if any (PR #71)\n\n## [2.0.3](https://github.com/Ousret/charset_normalizer/compare/2.0.2...2.0.3) (2021-07-16)\n### Changed\n- Part of the detection mechanism has been improved to be less sensitive, resulting in more accurate detection results. Especially ASCII. (PR #63)\n- According to the community wishes, the detection will fall back on ASCII or UTF-8 in a last-resort case. (PR #64)\n\n## [2.0.2](https://github.com/Ousret/charset_normalizer/compare/2.0.1...2.0.2) (2021-07-15)\n### Fixed\n- Empty/Too small JSON payload miss-detection fixed. Report from [@tseaver](https://github.com/tseaver) (PR #59)\n\n### Changed\n- Don't inject unicodedata2 into sys.modules from [@akx](https://github.com/akx) (PR #57)\n\n## [2.0.1](https://github.com/Ousret/charset_normalizer/compare/2.0.0...2.0.1) (2021-07-13)\n### Fixed\n- Make it work where there isn't a filesystem available, dropping assets frequencies.json. Report from [@sethmlarson](https://github.com/sethmlarson). (PR #55)\n- Using explain=False permanently disable the verbose output in the current runtime (PR #47)\n- One log entry (language target preemptive) was not show in logs when using explain=True (PR #47)\n- Fix undesired exception (ValueError) on getitem of instance CharsetMatches (PR #52)\n\n### Changed\n- Public function normalize default args values were not aligned with from_bytes (PR #53)\n\n### Added\n- You may now use charset aliases in cp_isolation and cp_exclusion arguments (PR #47)\n\n## [2.0.0](https://github.com/Ousret/charset_normalizer/compare/1.4.1...2.0.0) (2021-07-02)\n### Changed\n- 4x to 5 times faster than the previous 1.4.0 release. At least 2x faster than Chardet.\n- Accent has been made on UTF-8 detection, should perform rather instantaneous.\n- The backward compatibility with Chardet has been greatly improved. The legacy detect function returns an identical charset name whenever possible.\n- The detection mechanism has been slightly improved, now Turkish content is detected correctly (most of the time)\n- The program has been rewritten to ease the readability and maintainability. (+Using static typing)+\n- utf_7 detection has been reinstated.\n\n### Removed\n- This package no longer require anything when used with Python 3.5 (Dropped cached_property)\n- Removed support for these languages: Catalan, Esperanto, Kazakh, Baque, Volap\u00fck, Azeri, Galician, Nynorsk, Macedonian, and Serbocroatian.\n- The exception hook on UnicodeDecodeError has been removed.\n\n### Deprecated\n- Methods coherence_non_latin, w_counter, chaos_secondary_pass of the class CharsetMatch are now deprecated and scheduled for removal in v3.0\n\n### Fixed\n- The CLI output used the relative path of the file(s). Should be absolute.\n\n## [1.4.1](https://github.com/Ousret/charset_normalizer/compare/1.4.0...1.4.1) (2021-05-28)\n### Fixed\n- Logger configuration/usage no longer conflict with others (PR #44)\n\n## [1.4.0](https://github.com/Ousret/charset_normalizer/compare/1.3.9...1.4.0) (2021-05-21)\n### Removed\n- Using standard logging instead of using the package loguru.\n- Dropping nose test framework in favor of the maintained pytest.\n- Choose to not use dragonmapper package to help with gibberish Chinese/CJK text.\n- Require cached_property only for Python 3.5 due to constraint. Dropping for every other interpreter version.\n- Stop support for UTF-7 that does not contain a SIG.\n- Dropping PrettyTable, replaced with pure JSON output in CLI.\n\n### Fixed\n- BOM marker in a CharsetNormalizerMatch instance could be False in rare cases even if obviously present. Due to the sub-match factoring process.\n- Not searching properly for the BOM when trying utf32/16 parent codec.\n\n### Changed\n- Improving the package final size by compressing frequencies.json.\n- Huge improvement over the larges payload.\n\n### Added\n- CLI now produces JSON consumable output.\n- Return ASCII if given sequences fit. Given reasonable confidence.\n\n## [1.3.9](https://github.com/Ousret/charset_normalizer/compare/1.3.8...1.3.9) (2021-05-13)\n\n### Fixed\n- In some very rare cases, you may end up getting encode/decode errors due to a bad bytes payload (PR #40)\n\n## [1.3.8](https://github.com/Ousret/charset_normalizer/compare/1.3.7...1.3.8) (2021-05-12)\n\n### Fixed\n- Empty given payload for detection may cause an exception if trying to access the `alphabets` property. (PR #39)\n\n## [1.3.7](https://github.com/Ousret/charset_normalizer/compare/1.3.6...1.3.7) (2021-05-12)\n\n### Fixed\n- The legacy detect function should return UTF-8-SIG if sig is present in the payload. (PR #38)\n\n## [1.3.6](https://github.com/Ousret/charset_normalizer/compare/1.3.5...1.3.6) (2021-02-09)\n\n### Changed\n- Amend the previous release to allow prettytable 2.0 (PR #35)\n\n## [1.3.5](https://github.com/Ousret/charset_normalizer/compare/1.3.4...1.3.5) (2021-02-08)\n\n### Fixed\n- Fix error while using the package with a python pre-release interpreter (PR #33)\n\n### Changed\n- Dependencies refactoring, constraints revised.\n\n### Added\n- Add python 3.9 and 3.10 to the supported interpreters\n\nMIT License\n\nCopyright (c) 2025 TAHRI Ahmed R.\n\nPermission is hereby granted, free of charge, to any person obtaining a copy\nof this software and associated documentation files (the \"Software\"), to deal\nin the Software without restriction, including without limitation the rights\nto use, copy, modify, merge, publish, distribute, sublicense, and/or sell\ncopies of the Software, and to permit persons to whom the Software is\nfurnished to do so, subject to the following conditions:\n\nThe above copyright notice and this permission notice shall be included in all\ncopies or substantial portions of the Software.\n\nTHE SOFTWARE IS PROVIDED \"AS IS\", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR\nIMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,\nFITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE\nAUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER\nLIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,\nOUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE\nSOFTWARE.", + "release_date": "2025-05-02T08:34:06", "parties": [ { "type": "person", @@ -492,11 +492,11 @@ "Typing :: Typed" ], "homepage_url": null, - "download_url": "https://files.pythonhosted.org/packages/e4/10/a78c0e91f487b4ad0ef7480ac765e15b774f83de2597f1b6ef0eaf7a2f99/charset_normalizer-3.4.1-cp38-cp38-musllinux_1_2_x86_64.whl", - "size": 145169, + "download_url": "https://files.pythonhosted.org/packages/ac/c6/80b93fabc626b75b1665ffe405e28c3cef0aae9237c5c05f15955af4edd8/charset_normalizer-3.4.2-cp38-cp38-musllinux_1_2_x86_64.whl", + "size": 148007, "sha1": null, - "md5": "fa24a7a84664c7914117414dfd45489f", - "sha256": "0907f11d019260cdc3f94fbdb23ff9125f6b5d1039b76003b5b0ac9d6a6c9d5b", + "md5": "190b711b0064216e7f5c1119dcabd5aa", + "sha256": "dc7039885fa1baf9be153a0626e337aa7ec8bf96b0128605fb0d77788ddc1681", "sha512": null, "bug_tracking_url": null, "code_view_url": "https://github.com/jawah/charset_normalizer", @@ -516,9 +516,9 @@ "dependencies": [], "repository_homepage_url": null, "repository_download_url": null, - "api_data_url": "https://pypi.org/pypi/charset-normalizer/3.4.1/json", + "api_data_url": "https://pypi.org/pypi/charset-normalizer/3.4.2/json", "datasource_id": null, - "purl": "pkg:pypi/charset-normalizer@3.4.1" + "purl": "pkg:pypi/charset-normalizer@3.4.2" }, { "type": "pypi", @@ -1356,7 +1356,7 @@ ] }, { - "package": "pkg:pypi/charset-normalizer@3.4.1", + "package": "pkg:pypi/charset-normalizer@3.4.2", "dependencies": [] }, { @@ -1406,7 +1406,7 @@ "package": "pkg:pypi/requests@2.32.3", "dependencies": [ "pkg:pypi/certifi@2025.4.26", - "pkg:pypi/charset-normalizer@3.4.1", + "pkg:pypi/charset-normalizer@3.4.2", "pkg:pypi/idna@3.10", "pkg:pypi/urllib3@2.2.3" ] diff --git a/tests/data/resolved_deps/autobahn-310-expected.json b/tests/data/resolved_deps/autobahn-310-expected.json new file mode 100644 index 0000000..0e4bbc2 --- /dev/null +++ b/tests/data/resolved_deps/autobahn-310-expected.json @@ -0,0 +1,57 @@ +[ + [ + { + "package": "pkg:pypi/autobahn@22.3.2", + "dependencies": [ + "pkg:pypi/cryptography@43.0.3", + "pkg:pypi/hyperlink@21.0.0", + "pkg:pypi/setuptools@80.1.0", + "pkg:pypi/txaio@23.1.1" + ] + }, + { + "package": "pkg:pypi/cffi@1.17.1", + "dependencies": [ + "pkg:pypi/pycparser@2.22" + ] + }, + { + "package": "pkg:pypi/cryptography@43.0.3", + "dependencies": [ + "pkg:pypi/cffi@1.17.1" + ] + }, + { + "package": "pkg:pypi/hyperlink@21.0.0", + "dependencies": [ + "pkg:pypi/idna@3.10" + ] + }, + { + "package": "pkg:pypi/idna@3.10", + "dependencies": [] + }, + { + "package": "pkg:pypi/pycparser@2.22", + "dependencies": [] + }, + { + "package": "pkg:pypi/setuptools@80.1.0", + "dependencies": [] + }, + { + "package": "pkg:pypi/txaio@23.1.1", + "dependencies": [] + } + ], + [ + "pkg:pypi/autobahn@22.3.2", + "pkg:pypi/cffi@1.17.1", + "pkg:pypi/cryptography@43.0.3", + "pkg:pypi/hyperlink@21.0.0", + "pkg:pypi/idna@3.10", + "pkg:pypi/pycparser@2.22", + "pkg:pypi/setuptools@80.1.0", + "pkg:pypi/txaio@23.1.1" + ] +] \ No newline at end of file diff --git a/tests/data/resolved_deps/flask-310-expected.json b/tests/data/resolved_deps/flask-310-expected.json new file mode 100644 index 0000000..7216782 --- /dev/null +++ b/tests/data/resolved_deps/flask-310-expected.json @@ -0,0 +1,45 @@ +[ + [ + { + "package": "pkg:pypi/click@8.1.8", + "dependencies": [] + }, + { + "package": "pkg:pypi/flask@2.1.2", + "dependencies": [ + "pkg:pypi/click@8.1.8", + "pkg:pypi/itsdangerous@2.2.0", + "pkg:pypi/jinja2@3.1.6", + "pkg:pypi/werkzeug@3.1.3" + ] + }, + { + "package": "pkg:pypi/itsdangerous@2.2.0", + "dependencies": [] + }, + { + "package": "pkg:pypi/jinja2@3.1.6", + "dependencies": [ + "pkg:pypi/markupsafe@3.0.2" + ] + }, + { + "package": "pkg:pypi/markupsafe@3.0.2", + "dependencies": [] + }, + { + "package": "pkg:pypi/werkzeug@3.1.3", + "dependencies": [ + "pkg:pypi/markupsafe@3.0.2" + ] + } + ], + [ + "pkg:pypi/click@8.1.8", + "pkg:pypi/flask@2.1.2", + "pkg:pypi/itsdangerous@2.2.0", + "pkg:pypi/jinja2@3.1.6", + "pkg:pypi/markupsafe@3.0.2", + "pkg:pypi/werkzeug@3.1.3" + ] +] \ No newline at end of file diff --git a/tests/data/resolved_deps/flask-310-win-expected.json b/tests/data/resolved_deps/flask-310-win-expected.json new file mode 100644 index 0000000..af954aa --- /dev/null +++ b/tests/data/resolved_deps/flask-310-win-expected.json @@ -0,0 +1,52 @@ +[ + [ + { + "package": "pkg:pypi/click@8.1.8", + "dependencies": [ + "pkg:pypi/colorama@0.4.6" + ] + }, + { + "package": "pkg:pypi/colorama@0.4.6", + "dependencies": [] + }, + { + "package": "pkg:pypi/flask@2.1.2", + "dependencies": [ + "pkg:pypi/click@8.1.8", + "pkg:pypi/itsdangerous@2.2.0", + "pkg:pypi/jinja2@3.1.6", + "pkg:pypi/werkzeug@3.1.3" + ] + }, + { + "package": "pkg:pypi/itsdangerous@2.2.0", + "dependencies": [] + }, + { + "package": "pkg:pypi/jinja2@3.1.6", + "dependencies": [ + "pkg:pypi/markupsafe@3.0.2" + ] + }, + { + "package": "pkg:pypi/markupsafe@3.0.2", + "dependencies": [] + }, + { + "package": "pkg:pypi/werkzeug@3.1.3", + "dependencies": [ + "pkg:pypi/markupsafe@3.0.2" + ] + } + ], + [ + "pkg:pypi/click@8.1.8", + "pkg:pypi/colorama@0.4.6", + "pkg:pypi/flask@2.1.2", + "pkg:pypi/itsdangerous@2.2.0", + "pkg:pypi/jinja2@3.1.6", + "pkg:pypi/markupsafe@3.0.2", + "pkg:pypi/werkzeug@3.1.3" + ] +] \ No newline at end of file diff --git a/tests/data/resolved_deps/flask-36-expected.json b/tests/data/resolved_deps/flask-36-expected.json new file mode 100644 index 0000000..1e1f807 --- /dev/null +++ b/tests/data/resolved_deps/flask-36-expected.json @@ -0,0 +1,70 @@ +[ + [ + { + "package": "pkg:pypi/click@8.0.4", + "dependencies": [ + "pkg:pypi/importlib-metadata@4.8.3" + ] + }, + { + "package": "pkg:pypi/dataclasses@0.8", + "dependencies": [] + }, + { + "package": "pkg:pypi/flask@2.0.3", + "dependencies": [ + "pkg:pypi/click@8.0.4", + "pkg:pypi/itsdangerous@2.0.1", + "pkg:pypi/jinja2@3.0.3", + "pkg:pypi/werkzeug@2.0.3" + ] + }, + { + "package": "pkg:pypi/importlib-metadata@4.8.3", + "dependencies": [ + "pkg:pypi/typing-extensions@4.1.1", + "pkg:pypi/zipp@3.6.0" + ] + }, + { + "package": "pkg:pypi/itsdangerous@2.0.1", + "dependencies": [] + }, + { + "package": "pkg:pypi/jinja2@3.0.3", + "dependencies": [ + "pkg:pypi/markupsafe@2.0.1" + ] + }, + { + "package": "pkg:pypi/markupsafe@2.0.1", + "dependencies": [] + }, + { + "package": "pkg:pypi/typing-extensions@4.1.1", + "dependencies": [] + }, + { + "package": "pkg:pypi/werkzeug@2.0.3", + "dependencies": [ + "pkg:pypi/dataclasses@0.8" + ] + }, + { + "package": "pkg:pypi/zipp@3.6.0", + "dependencies": [] + } + ], + [ + "pkg:pypi/click@8.0.4", + "pkg:pypi/dataclasses@0.8", + "pkg:pypi/flask@2.0.3", + "pkg:pypi/importlib-metadata@4.8.3", + "pkg:pypi/itsdangerous@2.0.1", + "pkg:pypi/jinja2@3.0.3", + "pkg:pypi/markupsafe@2.0.1", + "pkg:pypi/typing-extensions@4.1.1", + "pkg:pypi/werkzeug@2.0.3", + "pkg:pypi/zipp@3.6.0" + ] +] \ No newline at end of file diff --git a/tests/data/resolved_deps/flask-39-expected.json b/tests/data/resolved_deps/flask-39-expected.json new file mode 100644 index 0000000..96d5bf3 --- /dev/null +++ b/tests/data/resolved_deps/flask-39-expected.json @@ -0,0 +1,58 @@ +[ + [ + { + "package": "pkg:pypi/click@8.1.8", + "dependencies": [] + }, + { + "package": "pkg:pypi/flask@2.1.3", + "dependencies": [ + "pkg:pypi/click@8.1.8", + "pkg:pypi/importlib-metadata@8.7.0", + "pkg:pypi/itsdangerous@2.2.0", + "pkg:pypi/jinja2@3.1.6", + "pkg:pypi/werkzeug@3.1.3" + ] + }, + { + "package": "pkg:pypi/importlib-metadata@8.7.0", + "dependencies": [ + "pkg:pypi/zipp@3.21.0" + ] + }, + { + "package": "pkg:pypi/itsdangerous@2.2.0", + "dependencies": [] + }, + { + "package": "pkg:pypi/jinja2@3.1.6", + "dependencies": [ + "pkg:pypi/markupsafe@3.0.2" + ] + }, + { + "package": "pkg:pypi/markupsafe@3.0.2", + "dependencies": [] + }, + { + "package": "pkg:pypi/werkzeug@3.1.3", + "dependencies": [ + "pkg:pypi/markupsafe@3.0.2" + ] + }, + { + "package": "pkg:pypi/zipp@3.21.0", + "dependencies": [] + } + ], + [ + "pkg:pypi/click@8.1.8", + "pkg:pypi/flask@2.1.3", + "pkg:pypi/importlib-metadata@8.7.0", + "pkg:pypi/itsdangerous@2.2.0", + "pkg:pypi/jinja2@3.1.6", + "pkg:pypi/markupsafe@3.0.2", + "pkg:pypi/werkzeug@3.1.3", + "pkg:pypi/zipp@3.21.0" + ] +] \ No newline at end of file diff --git a/tests/data/resolved_deps/torch-312-expected.json b/tests/data/resolved_deps/torch-312-expected.json new file mode 100644 index 0000000..20e7c59 --- /dev/null +++ b/tests/data/resolved_deps/torch-312-expected.json @@ -0,0 +1,11 @@ +[ + [ + { + "package": "pkg:pypi/torchcodec@0.3.0%2Bcu126", + "dependencies": [] + } + ], + [ + "pkg:pypi/torchcodec@0.3.0%2Bcu126" + ] +] \ No newline at end of file diff --git a/tests/data/test-api-with-requirement-file.json b/tests/data/test-api-with-requirement-file.json index 25f8e0e..1c5c08d 100644 --- a/tests/data/test-api-with-requirement-file.json +++ b/tests/data/test-api-with-requirement-file.json @@ -3727,12 +3727,12 @@ "type": "pypi", "namespace": null, "name": "pip", - "version": "25.1", + "version": "25.1.1", "qualifiers": {}, "subpath": null, "primary_language": "Python", "description": "The PyPA recommended tool for installing Python packages.\npip - The Python Package Installer\n==================================\n\n.. |pypi-version| image:: https://img.shields.io/pypi/v/pip.svg\n :target: https://pypi.org/project/pip/\n :alt: PyPI\n\n.. |python-versions| image:: https://img.shields.io/pypi/pyversions/pip\n :target: https://pypi.org/project/pip\n :alt: PyPI - Python Version\n\n.. |docs-badge| image:: https://readthedocs.org/projects/pip/badge/?version=latest\n :target: https://pip.pypa.io/en/latest\n :alt: Documentation\n\n|pypi-version| |python-versions| |docs-badge|\n\npip is the `package installer`_ for Python. You can use pip to install packages from the `Python Package Index`_ and other indexes.\n\nPlease take a look at our documentation for how to install and use pip:\n\n* `Installation`_\n* `Usage`_\n\nWe release updates regularly, with a new version every 3 months. Find more details in our documentation:\n\n* `Release notes`_\n* `Release process`_\n\nIf you find bugs, need help, or want to talk to the developers, please use our mailing lists or chat rooms:\n\n* `Issue tracking`_\n* `Discourse channel`_\n* `User IRC`_\n\nIf you want to get involved head over to GitHub to get the source code, look at our development documentation and feel free to jump on the developer mailing lists and chat rooms:\n\n* `GitHub page`_\n* `Development documentation`_\n* `Development IRC`_\n\nCode of Conduct\n---------------\n\nEveryone interacting in the pip project's codebases, issue trackers, chat\nrooms, and mailing lists is expected to follow the `PSF Code of Conduct`_.\n\n.. _package installer: https://packaging.python.org/guides/tool-recommendations/\n.. _Python Package Index: https://pypi.org\n.. _Installation: https://pip.pypa.io/en/stable/installation/\n.. _Usage: https://pip.pypa.io/en/stable/\n.. _Release notes: https://pip.pypa.io/en/stable/news.html\n.. _Release process: https://pip.pypa.io/en/latest/development/release-process/\n.. _GitHub page: https://github.com/pypa/pip\n.. _Development documentation: https://pip.pypa.io/en/latest/development\n.. _Issue tracking: https://github.com/pypa/pip/issues\n.. _Discourse channel: https://discuss.python.org/c/packaging\n.. _User IRC: https://kiwiirc.com/nextclient/#ircs://irc.libera.chat:+6697/pypa\n.. _Development IRC: https://kiwiirc.com/nextclient/#ircs://irc.libera.chat:+6697/pypa-dev\n.. _PSF Code of Conduct: https://github.com/pypa/.github/blob/main/CODE_OF_CONDUCT.md", - "release_date": "2025-04-26T09:45:33", + "release_date": "2025-05-02T15:13:59", "parties": [ { "type": "person", @@ -3758,11 +3758,11 @@ "Topic :: Software Development :: Build Tools" ], "homepage_url": null, - "download_url": "https://files.pythonhosted.org/packages/e0/f0/8a2806114cd36e282823fd4d8e88e3b94dc943c2569c350d0c826a49db38/pip-25.1-py3-none-any.whl", - "size": 1824948, + "download_url": "https://files.pythonhosted.org/packages/29/a2/d40fb2460e883eca5199c62cfc2463fd261f760556ae6290f88488c362c0/pip-25.1.1-py3-none-any.whl", + "size": 1825227, "sha1": null, - "md5": "4eb2baef792e6841cfcd4c24c808421a", - "sha256": "13b4aa0aaad055020a11bec8a1c2a70a2b2d080e12d89b962266029fff0a16ba", + "md5": "6a01d861bf88ee075c6942419a04839d", + "sha256": "2913a38a2abf4ea6b64ab507bd9e967f3b53dc1ede74b01b0931e1ce548751af", "sha512": null, "bug_tracking_url": null, "code_view_url": "https://github.com/pypa/pip", @@ -3782,9 +3782,9 @@ "dependencies": [], "repository_homepage_url": null, "repository_download_url": null, - "api_data_url": "https://pypi.org/pypi/pip/25.1/json", + "api_data_url": "https://pypi.org/pypi/pip/25.1.1/json", "datasource_id": null, - "purl": "pkg:pypi/pip@25.1" + "purl": "pkg:pypi/pip@25.1.1" }, { "type": "pypi", @@ -6162,13 +6162,13 @@ ] }, { - "package": "pkg:pypi/pip@25.1", + "package": "pkg:pypi/pip@25.1.1", "dependencies": [] }, { "package": "pkg:pypi/pipdeptree@2.2.1", "dependencies": [ - "pkg:pypi/pip@25.1" + "pkg:pypi/pip@25.1.1" ] }, { diff --git a/tests/test_resolution.py b/tests/test_resolution.py index 0e84ef3..9ad953d 100644 --- a/tests/test_resolution.py +++ b/tests/test_resolution.py @@ -13,9 +13,10 @@ import packvers import pytest -from commoncode.system import on_mac +from commoncode.system import on_linux from commoncode.testcase import FileDrivenTesting from packvers.requirements import Requirement +from test_cli import check_data_results from _packagedcode import models from python_inspector.api import get_resolved_dependencies @@ -29,162 +30,146 @@ from python_inspector.utils_pypi import Environment from python_inspector.utils_pypi import PypiSimpleRepository +# Used for tests to regenerate fixtures with regen=True +REGEN_TEST_FIXTURES = os.getenv("PYINSP_REGEN_TEST_FIXTURES", False) + setup_test_env = FileDrivenTesting() setup_test_env.test_data_dir = os.path.join(os.path.dirname(__file__), "data") +def check_get_resolved_dependencies( + requirement: Requirement, + expected_file, + python_version, + operating_system, + repos=None, + as_tree=False, + regen=REGEN_TEST_FIXTURES, +): + env = Environment(python_version=python_version, operating_system=operating_system) + + results = list( + get_resolved_dependencies( + requirements=[requirement], + environment=env, + repos=repos or [PYPI_PUBLIC_REPO], + as_tree=as_tree, + ) + ) + check_data_results(results=results, expected_file=expected_file, regen=regen) + + @pytest.mark.online def test_get_resolved_dependencies_with_flask_and_python_310(): req = Requirement("flask==2.1.2") req.is_requirement_resolved = True - _, plist = get_resolved_dependencies( - requirements=[req], - environment=Environment( - python_version="310", - operating_system="linux", - ), - repos=[PYPI_PUBLIC_REPO], + + expected_file = setup_test_env.get_test_loc( + "resolved_deps/flask-310-expected.json", must_exist=False + ) + + check_get_resolved_dependencies( + req, + expected_file=expected_file, + python_version="310", + operating_system="linux", as_tree=False, ) - assert plist == [ - "pkg:pypi/click@8.1.8", - "pkg:pypi/flask@2.1.2", - "pkg:pypi/itsdangerous@2.2.0", - "pkg:pypi/jinja2@3.1.6", - "pkg:pypi/markupsafe@3.0.2", - "pkg:pypi/werkzeug@3.1.3", - ] @pytest.mark.online def test_get_resolved_dependencies_with_flask_and_python_310_windows(): req = Requirement("flask==2.1.2") req.is_requirement_resolved = True - _, plist = get_resolved_dependencies( - requirements=[req], - environment=Environment( - python_version="310", - operating_system="windows", - ), - repos=[PYPI_PUBLIC_REPO], + + expected_file = setup_test_env.get_test_loc( + "resolved_deps/flask-310-win-expected.json", must_exist=False + ) + + check_get_resolved_dependencies( + req, + expected_file=expected_file, + python_version="310", + operating_system="windows", as_tree=False, ) - assert plist == [ - "pkg:pypi/click@8.1.8", - "pkg:pypi/colorama@0.4.6", - "pkg:pypi/flask@2.1.2", - "pkg:pypi/itsdangerous@2.2.0", - "pkg:pypi/jinja2@3.1.6", - "pkg:pypi/markupsafe@3.0.2", - "pkg:pypi/werkzeug@3.1.3", - ] @pytest.mark.online def test_get_resolved_dependencies_with_flask_and_python_36(): req = Requirement("flask") req.is_requirement_resolved = False - _, plist = get_resolved_dependencies( - requirements=[req], - environment=Environment( - python_version="36", - operating_system="linux", - ), - repos=[PYPI_PUBLIC_REPO], - as_tree=False, + + expected_file = setup_test_env.get_test_loc( + "resolved_deps/flask-36-expected.json", must_exist=False ) - assert plist == [ - "pkg:pypi/click@8.0.4", - "pkg:pypi/dataclasses@0.8", - "pkg:pypi/flask@2.0.3", - "pkg:pypi/importlib-metadata@4.8.3", - "pkg:pypi/itsdangerous@2.0.1", - "pkg:pypi/jinja2@3.0.3", - "pkg:pypi/markupsafe@2.0.1", - "pkg:pypi/typing-extensions@4.1.1", - "pkg:pypi/werkzeug@2.0.3", - "pkg:pypi/zipp@3.6.0", - ] + check_get_resolved_dependencies( + req, + expected_file=expected_file, + python_version="36", + operating_system="linux", + as_tree=False, + ) @pytest.mark.online def test_get_resolved_dependencies_with_tilde_requirement_using_json_api(): req = Requirement("flask~=2.1.2") req.is_requirement_resolved = False - _, plist = get_resolved_dependencies( - requirements=[req], + + expected_file = setup_test_env.get_test_loc( + "resolved_deps/flask-39-expected.json", must_exist=False + ) + + check_get_resolved_dependencies( + req, + expected_file=expected_file, + python_version="39", + operating_system="linux", as_tree=False, - environment=Environment( - python_version="38", - operating_system="linux", - ), ) - assert plist == [ - "pkg:pypi/click@8.1.8", - "pkg:pypi/flask@2.1.3", - "pkg:pypi/importlib-metadata@8.7.0", - "pkg:pypi/itsdangerous@2.2.0", - "pkg:pypi/jinja2@3.1.6", - "pkg:pypi/markupsafe@3.0.2", - "pkg:pypi/werkzeug@3.1.3", - "pkg:pypi/zipp@3.21.0", - ] + + +# fbgemm_gpu-1.2.0+cu118-cp312-cp312-manylinux_2_28_x86_64.whl @pytest.mark.online -@pytest.mark.skipif(on_mac, reason="torch is only available for linux and windows.") +@pytest.mark.skipif(not on_linux, reason="torch is only available for linux.") def test_get_resolved_dependencies_for_version_containing_local_version_identifier(): - req = Requirement("torch==2.0.0+cpu") + req = Requirement("torchcodec==0.3.0+cu126") req.is_requirement_resolved = True - _, plist = get_resolved_dependencies( - requirements=[req], - environment=Environment( - python_version="310", - operating_system="linux", - ), - repos=[ - PypiSimpleRepository(index_url="https://download.pytorch.org/whl/cpu", credentials=None) - ], - as_tree=False, + + repos = [PypiSimpleRepository(index_url="https://download.pytorch.org/whl")] + expected_file = setup_test_env.get_test_loc( + "resolved_deps/torch-312-expected.json", must_exist=False ) - assert plist == [ - "pkg:pypi/filelock@3.13.1", - "pkg:pypi/jinja2@3.1.4", - "pkg:pypi/markupsafe@2.1.5", - "pkg:pypi/mpmath@1.3.0", - "pkg:pypi/networkx@3.3", - "pkg:pypi/sympy@1.13.3", - "pkg:pypi/torch@2.0.0%2Bcpu", - "pkg:pypi/typing-extensions@4.12.2", - ] + check_get_resolved_dependencies( + req, + expected_file=expected_file, + python_version="312", + operating_system="linux", + repos=repos, + as_tree=False, + ) @pytest.mark.online def test_without_supported_wheels(): req = Requirement("autobahn==22.3.2") req.is_requirement_resolved = True - _, plist = get_resolved_dependencies( - requirements=[req], - as_tree=False, - repos=[PYPI_PUBLIC_REPO], - environment=Environment( - python_version="38", - operating_system="linux", - ), + expected_file = setup_test_env.get_test_loc( + "resolved_deps/autobahn-310-expected.json", must_exist=False ) - assert plist == [ - "pkg:pypi/autobahn@22.3.2", - "pkg:pypi/cffi@1.17.1", - "pkg:pypi/cryptography@44.0.2", - "pkg:pypi/hyperlink@21.0.0", - "pkg:pypi/idna@3.10", - "pkg:pypi/pycparser@2.22", - "pkg:pypi/setuptools@75.3.2", - "pkg:pypi/txaio@23.1.1", - ] + check_get_resolved_dependencies( + req, + expected_file=expected_file, + python_version="39", + operating_system="linux", + as_tree=False, + ) def test_is_valid_version():