Skip to content

Add script to print GitHub workflow run errors #1761

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom
Draft
wants to merge 12 commits into
base: main
Choose a base branch
from
Draft

Add script to print GitHub workflow run errors #1761

wants to merge 12 commits into from
+589 −0

Conversation

jonsimantov
Copy link
Contributor

This script allows users to specify a workflow name and branch to find the latest run. It then identifies any failed jobs within that run and prints the last N lines of logs for each failed step.

Key features:

  • Fetches the most recent workflow run for a given workflow and branch.
  • Identifies jobs within the run that have a 'failure' conclusion.
  • For each failed job, attempts to identify failed steps and extracts their logs.
  • Prints the last 500 lines (configurable) of the log for each failed step.
  • Handles various scenarios including successful runs, running workflows, and missing data.
  • Supports GitHub token via CLI arg, env var, or ~/.github_token.
  • Auto-detects repository from git remote, or accepts via CLI args.

Description

Provide details of the change, and generalize the change in the PR title above.

Testing

Describe how you've tested these changes. Link any manually triggered Integration tests or CPP binary SDK Packaging Github Action workflows, if applicable.

Type of Change

Place an x the applicable box:

  • Bug fix. Add the issue # below if applicable.
  • New feature. A non-breaking change which adds functionality.
  • Other, such as a build process or documentation change.

Notes

  • Bug fixes and feature changes require an update to the Release Notes section of release_build_files/readme.md.
  • Read the contribution guidelines CONTRIBUTING.md.
  • Changes to the public API require an internal API review. If you'd like to help us make Firebase APIs better, please propose your change in a feature request so that we can discuss it together.

google-labs-jules bot added 12 commits June 27, 2025 20:59
@google-labs-jules
Add script to print GitHub workflow run errors
33bd1f3 
This script allows users to specify a workflow name and branch to find
the latest run. It then identifies any failed jobs within that run and
prints the last N lines of logs for each failed step.

Key features:
- Fetches the most recent workflow run for a given workflow and branch.
- Identifies jobs within the run that have a 'failure' conclusion.
- For each failed job, attempts to identify failed steps and extracts their logs.
- Prints the last 500 lines (configurable) of the log for each failed step.
- Handles various scenarios including successful runs, running workflows,
  and missing data.
- Supports GitHub token via CLI arg, env var, or ~/.github_token.
- Auto-detects repository from git remote, or accepts via CLI args.
@google-labs-jules
Enhance workflow error script with options and defaults
bcdf292 
This commit updates the `print_workflow_run_errors.py` script:

- Workflow name and branch are now optional arguments:
    - `--workflow` (or `--workflow-name`) defaults to "integration_test.yml".
    - `--branch` defaults to the current Git branch.
- Changed default log lines printed from 500 to 100 (`--log-lines`).
- Added `--all-failed-steps` flag:
    - If false (default), only logs for the first failed step in a job are printed.
    - If true, logs for all failed steps in a job are printed.

These changes provide more flexibility and sensible defaults for common use cases.
@google-labs-jules
Add grep functionality to workflow error script
6ba2558 
This commit introduces a grep-like feature to the
`print_workflow_run_errors.py` script.

New features:
- Added `--grep-pattern` (`-g`) argument to specify an Extended
  Regular Expression (ERE) for searching within fetched logs.
- Added `--grep-context` (`-C`) argument to specify the number of
  lines of context to show around matches (default is 5).

Behavior:
- If a grep pattern is provided, the script will use the system `grep`
  command to filter the logs of failed steps (or the full job log
  if a specific step's log cannot be isolated).
- Output clearly indicates when grep results are shown, the pattern used,
  and the context lines.
- Handles cases where `grep` finds no matches or if the `grep` command
  itself fails (e.g., not found, bad pattern).
- If no grep pattern is provided, the script defaults to its previous
  behavior of printing the last N lines of the log.
@google-labs-jules
Update grep defaults in workflow error script
2f0aab6 
This commit refines the grep functionality in the
`print_workflow_run_errors.py` script by updating default values:

- The default `--grep-pattern` is now `"[Ee]rror[: ]"`.
  Logs will be automatically filtered for this pattern if no other
  pattern is specified.
- The default `--grep-context` is now 10 lines.

If an empty string is explicitly passed to `--grep-pattern`
(e.g., `--grep-pattern ""`), grep functionality will be disabled,
and the script will fall back to printing the last N lines of the log.
@google-labs-jules
Add job name pattern filtering to workflow error script
7319def 
This commit introduces a new feature to filter jobs by name using a
regular expression in the `print_workflow_run_errors.py` script.

New features:
- Added `--job-pattern` argument, which accepts a regex string.
  Only jobs whose names match this pattern will be processed for failures.
- The default value for `--job-pattern` is `'^build.*'`, meaning by
  default, the script will only look at jobs starting with 'build'.

Behavior:
- Job name filtering is applied before checking for job failures.
- If an invalid regex is provided for `--job-pattern`, the script prints
  an error and exits gracefully.
- This new filter works in conjunction with existing log processing
  options (like `--grep-pattern` and `--all-failed-steps`), which are
  applied to the jobs that pass the name pattern filter and are failed.
@google-labs-jules
Refine default grep pattern in workflow error script
c20edf6 
This commit updates the default regular expression for the
`--grep-pattern` argument in the `print_workflow_run_errors.py` script.

The default pattern is changed from `"[Ee]rror[: ]"` to
`"[Ee][Rr][Rr][Oo][Rr][: ]"` for more specific matching of "Error"
(case-insensitive) followed by a colon or space.
@google-labs-jules
Format workflow error script output as Markdown
bcf6245 
This commit updates the `print_workflow_run_errors.py` script to format
its standard output using Markdown for improved readability when viewed
in Markdown-aware environments.

Changes include:
- Job and step headings are now formatted as Markdown headers (H1-H4).
- Workflow run and job URLs are presented as clickable Markdown links.
- Actual log content (both grep results and last N lines) is enclosed
  in ```log ... ``` fenced code blocks.
- Horizontal rules (`---`) are used to better separate sections.
- Minor textual adjustments for clarity within the Markdown structure.

The stderr output remains plain text for informational messages.
@google-labs-jules
Enhance script with log download status and timestamp stripping
4c2fb08 
This commit adds two main enhancements to the
`print_workflow_run_errors.py` script:

1.  **Log Download Status (stderr)**:
    - Prints messages to stderr indicating the progress of log downloads
      (e.g., "INFO: Downloading log X/Y for job 'job_name'...").
    - Prints a summary to stderr after all jobs are processed, showing how
      many logs were successfully fetched and processed (e.g.,
      "INFO: Processed logs for S/T targeted failed jobs.").
    - Includes a warning on stderr if a specific job's log fails to download.

2.  **Timestamp Stripping (stdout)**:
    - Implemented a function `strip_initial_timestamp` that uses a regex
      to remove ISO 8601-like timestamps from the beginning of log lines.
    - This stripping is applied to all log lines (from specific step
      segments or full job log fallbacks) before they are further
      processed by `grep` or printed as the 'last N lines'.
      This makes the logs cleaner and potentially easier for other tools
      or users to parse.
@google-labs-jules
Implement prioritized job pattern checking
0ffb5a6 
This commit introduces a prioritized, cascading job pattern checking
feature to the `print_workflow_run_errors.py` script.

Key changes:
- The `--job-pattern` argument now uses `action='append'`, allowing users
  to specify multiple patterns. These are checked in the order provided.
- If no `--job-pattern` is specified by the user, a default prioritized
  sequence is used: `['^build.*', '^test.*', '.*']`.
- The script iterates through the determined list of patterns:
    - For the first pattern that matches jobs AND has failures among those
      matched jobs, the script processes and displays logs for those
      failures.
    - It then stops checking subsequent patterns.
- If a pattern results in no matching jobs, or if all matching jobs
  succeeded, the script moves to the next pattern in the sequence.
- Informative messages are printed to stderr indicating which pattern is
  being checked, the outcome for that pattern, and if subsequent
  patterns are skipped.
- The main log processing loop has been refactored into a helper function
  `_process_and_display_logs_for_failed_jobs` for better organization.
- Handles invalid regular expressions within the pattern list by warning
  the user and skipping the invalid pattern.
@google-labs-jules
Revert default grep context to 5
e1ce17a 
This commit changes the default value for the `--grep-context`
argument in the `print_workflow_run_errors.py` script back to 5
(from 10).
@google-labs-jules
Add --run-id option to process specific workflow run
d7cea5c 
This commit enhances `print_workflow_run_errors.py` by adding a
`--run-id` command-line option. This allows users to specify a
particular workflow run ID for processing, bypassing the default
behavior of searching for the latest run by workflow name and branch.

Key changes:
- Added `--run-id <ID>` optional argument.
- If `--run-id` is provided, the script fetches details for that specific
  run using a new helper function `get_workflow_run_details_by_id`.
  The `--workflow` and `--branch` arguments are ignored in this mode.
- If `--run-id` is not provided, the script retains its existing behavior
  of using `--workflow` and `--branch` to find the latest run.
- The new helper function includes error handling for invalid or
  non-existent run IDs.
- Standard error messages have been updated to reflect whether the script
  is processing a run by specified ID or by search criteria.
@google-labs-jules
Fix NameError for run_details variable
15531c5 
Corrects a NameError in the `main` function when calling
`_process_and_display_logs_for_failed_jobs`. The script was
attempting to use `run.get('html_url')` where it should have been
`run_details.get('html_url')` after a previous refactor.

This fix ensures the correct variable is used, allowing the script
to properly pass the workflow run's HTML URL to the log processing
function.
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers
No reviews
Assignees
No one assigned
Labels
None yet
Projects
None yet
Milestone
No milestone
Development

Successfully merging this pull request may close these issues.
1 participant
@jonsimantov