build(deps): bump unstructured and unstructured-inference (#210)

Author: awalker4

.github/workflows/ci.yml

@@ -7,7 +7,7 @@ on:
     branches: [ main ]
 
 env:
-  PYTHON_VERSION: "3.8"
+  PYTHON_VERSION: "3.10"
   PIPELINE_FAMILY: "general"
 
 jobs:

.github/workflows/docker-publish.yml

@@ -11,7 +11,7 @@ env:
   PACKAGE: "unstructured-api"
   PIPELINE_FAMILY: "general"
   PIP_VERSION: "22.2.1"
-  PYTHON_VERSION: "3.8"
+  PYTHON_VERSION: "3.10"
 
 jobs:
   setup:

CHANGELOG.md

@@ -1,3 +1,8 @@
+## 0.0.41
+
+* Bump unstructured library to 0.10.8
+* Bump unstructured-inference to 0.5.17
+
 ## 0.0.40
 
 * Reject traffic when overloaded via `UNSTRUCTURED_MEMORY_FREE_MINIMUM_MB`

pipeline-notebooks/pipeline-general.ipynb

@@ -680,15 +680,15 @@
     "\n",
     "    return elements\n",
     "\n",
-    "def partition_pdf_splits(request, pdf_pages, file, file_filename, content_type, coordinates, **partition_kwargs):\n",
+    "def partition_pdf_splits(request, pdf_pages, file, metadata_filename, content_type, coordinates, **partition_kwargs):\n",
     "    '''\n",
     "    Split a pdf into chunks and process in parallel with more api calls, or partition\n",
     "    locally if the chunk is small enough. As soon as any remote call fails, bubble up\n",
     "    the error.\n",
     "    \n",
     "    Arguments:\n",
     "    request is used to forward relevant headers to the api calls\n",
-    "    file, file_filename and content_type are passed on in the file argument to requests.post\n",
+    "    file, metadata_filename and content_type are passed on in the file argument to requests.post\n",
     "    coordinates is passed on to the api calls, but cannot be used in the local partition case\n",
     "    partition_kwargs holds any others parameters that will be forwarded, or passed to partition\n",
     "    '''    \n",
@@ -698,7 +698,7 @@
     "    if len(pdf_pages) <= pages_per_pdf:\n",
     "        return partition(\n",
     "            file=file,\n",
-    "            file_filename=file_filename,\n",
+    "            metadata_filename=metadata_filename,\n",
     "            content_type=content_type,\n",
     "            **partition_kwargs\n",
     "       )\n",
@@ -709,7 +709,7 @@
     "    partition_func = partial(\n",
     "        partition_file_via_api,\n",
     "        request=request,\n",
-    "        filename=file_filename,\n",
+    "        filename=metadata_filename,\n",
     "        content_type=content_type,\n",
     "        coordinates=coordinates,\n",
     "        **partition_kwargs\n",
@@ -864,7 +864,7 @@
     "                request,\n",
     "                pdf_pages=pdf.pages,\n",
     "                file=file,\n",
-    "                file_filename=filename,\n",
+    "                metadata_filename=filename,\n",
     "                content_type=file_content_type,\n",
     "                coordinates=show_coordinates,\n",
     "                # partition_kwargs\n",
@@ -880,7 +880,7 @@
     "        else:\n",
     "            elements = partition(\n",
     "                file=file,\n",
-    "                file_filename=filename,\n",
+    "                metadata_filename=filename,\n",
     "                content_type=file_content_type,\n",
     "                # partition_kwargs\n",
     "                encoding=encoding,\n",

prepline_general/api/general.py

@@ -153,7 +153,7 @@ def partition_file_via_api(file_tuple, request, filename, content_type, **partit
 
 
 def partition_pdf_splits(
-    request, pdf_pages, file, file_filename, content_type, coordinates, **partition_kwargs
+    request, pdf_pages, file, metadata_filename, content_type, coordinates, **partition_kwargs
 ):
     """
     Split a pdf into chunks and process in parallel with more api calls, or partition
@@ -162,7 +162,7 @@ def partition_pdf_splits(
 
     Arguments:
     request is used to forward relevant headers to the api calls
-    file, file_filename and content_type are passed on in the file argument to requests.post
+    file, metadata_filename and content_type are passed on in the file argument to requests.post
     coordinates is passed on to the api calls, but cannot be used in the local partition case
     partition_kwargs holds any others parameters that will be forwarded, or passed to partition
     """
@@ -171,7 +171,10 @@ def partition_pdf_splits(
     # If it's small enough, just process locally
     if len(pdf_pages) <= pages_per_pdf:
         return partition(
-            file=file, file_filename=file_filename, content_type=content_type, **partition_kwargs
+            file=file,
+            metadata_filename=metadata_filename,
+            content_type=content_type,
+            **partition_kwargs,
         )
 
     results = []
@@ -180,7 +183,7 @@ def partition_pdf_splits(
     partition_func = partial(
         partition_file_via_api,
         request=request,
-        filename=file_filename,
+        filename=metadata_filename,
         content_type=content_type,
         coordinates=coordinates,
         **partition_kwargs,
@@ -340,7 +343,7 @@ def pipeline_api(
                 request,
                 pdf_pages=pdf.pages,
                 file=file,
-                file_filename=filename,
+                metadata_filename=filename,
                 content_type=file_content_type,
                 coordinates=show_coordinates,
                 # partition_kwargs
@@ -356,7 +359,7 @@ def pipeline_api(
         else:
             elements = partition(
                 file=file,
-                file_filename=filename,
+                metadata_filename=filename,
                 content_type=file_content_type,
                 # partition_kwargs
                 encoding=encoding,
@@ -502,7 +505,7 @@ def return_content_type(filename):
 
 
 @router.post("/general/v0/general")
-@router.post("/general/v0.0.40/general")
+@router.post("/general/v0.0.41/general")
 def pipeline_1(
     request: Request,
     gz_uncompressed_content_type: Optional[str] = Form(default=None),

preprocessing-pipeline-family.yaml

@@ -1,2 +1,2 @@
 name: general
-version: 0.0.40
+version: 0.0.41

requirements/base.txt

@@ -1,5 +1,5 @@
 #
-# This file is autogenerated by pip-compile with Python 3.8
+# This file is autogenerated by pip-compile with Python 3.10
 # by the following command:
 #
 #    pip-compile requirements/base.in
@@ -58,11 +58,11 @@ et-xmlfile==1.1.0
     # via openpyxl
 exceptiongroup==1.1.3
     # via anyio
-fastapi==0.101.1
+fastapi==0.103.0
     # via unstructured-api-tools
 fastjsonschema==2.18.0
     # via nbformat
-filelock==3.12.2
+filelock==3.12.3
     # via
     #   huggingface-hub
     #   torch
@@ -90,16 +90,6 @@ idna==3.4
     # via
     #   anyio
     #   requests
-importlib-metadata==6.8.0
-    # via
-    #   jupyter-client
-    #   markdown
-    #   nbconvert
-importlib-resources==6.0.1
-    # via
-    #   jsonschema
-    #   jsonschema-specifications
-    #   matplotlib
 iopath==0.1.10
     # via layoutparser
 jinja2==3.1.2
@@ -113,7 +103,7 @@ jsonschema==4.19.0
     # via nbformat
 jsonschema-specifications==2023.7.1
     # via jsonschema
-jupyter-client==8.3.0
+jupyter-client==8.3.1
     # via nbclient
 jupyter-core==5.3.1
     # via
@@ -153,7 +143,7 @@ mypy-extensions==1.0.0
     # via mypy
 nbclient==0.8.0
     # via nbconvert
-nbconvert==7.7.4
+nbconvert==7.8.0
     # via unstructured-api-tools
 nbformat==5.9.2
     # via
@@ -163,7 +153,7 @@ networkx==3.1
     # via torch
 nltk==3.8.1
     # via unstructured
-numpy==1.24.4
+numpy==1.25.2
     # via
     #   contourpy
     #   layoutparser
@@ -222,13 +212,11 @@ pillow==9.5.0
     #   python-pptx
     #   torchvision
     #   unstructured
-pkgutil-resolve-name==1.3.10
-    # via jsonschema
 platformdirs==3.10.0
     # via jupyter-core
 portalocker==2.7.0
     # via iopath
-protobuf==4.24.1
+protobuf==4.24.2
     # via onnxruntime
 psutil==5.9.5
     # via -r requirements/base.in
@@ -250,9 +238,9 @@ pypandoc==1.11
     # via unstructured
 pyparsing==3.0.9
     # via matplotlib
-pypdf==3.15.2
+pypdf==3.15.4
     # via -r requirements/base.in
-pypdfium2==4.18.0
+pypdfium2==4.19.0
     # via pdfplumber
 pytesseract==0.3.10
     # via layoutparser
@@ -271,7 +259,7 @@ python-multipart==0.0.6
     # via
     #   unstructured-api-tools
     #   unstructured-inference
-python-pptx==0.6.21
+python-pptx==0.6.22
     # via unstructured
 pytz==2023.3
     # via pandas
@@ -302,7 +290,7 @@ requests==2.31.0
     #   torchvision
     #   transformers
     #   unstructured
-rpds-py==0.9.2
+rpds-py==0.10.0
     # via
     #   jsonschema
     #   referencing
@@ -311,7 +299,7 @@ safetensors==0.3.2
     #   -c requirements/constraints.in
     #   timm
     #   transformers
-scipy==1.10.1
+scipy==1.11.2
     # via layoutparser
 six==1.16.0
     # via
@@ -366,7 +354,7 @@ traitlets==5.9.0
     #   nbclient
     #   nbconvert
     #   nbformat
-transformers==4.32.0
+transformers==4.32.1
     # via unstructured-inference
 types-requests==2.31.0.2
     # via unstructured-api-tools
@@ -377,21 +365,21 @@ types-urllib3==1.26.25.14
 typing-extensions==4.7.1
     # via
     #   fastapi
+    #   filelock
     #   huggingface-hub
     #   iopath
     #   mypy
     #   pydantic
-    #   pypdf
-    #   starlette
+    #   pydantic-core
     #   torch
     #   uvicorn
 tzdata==2023.3
     # via pandas
-unstructured[local-inference]==0.10.5
+unstructured[local-inference]==0.10.8
     # via -r requirements/base.in
 unstructured-api-tools==0.10.11
     # via -r requirements/base.in
-unstructured-inference==0.5.16
+unstructured-inference==0.5.17
     # via unstructured
 urllib3==2.0.4
     # via requests
@@ -411,7 +399,3 @@ xlrd==2.0.1
     # via unstructured
 xlsxwriter==3.1.2
     # via python-pptx
-zipp==3.16.2
-    # via
-    #   importlib-metadata
-    #   importlib-resources