Job summary shows truncated message

[tgehrs]

We have a pull request that exceeds the max_files number and the python-coverage-comment-action displays this message at the bottom of the coverage comment.

When I navigate to the workflow summary page, it shows a similar message about truncating:

https://github.com/<org>/<repo>/actions/runs/<run_id>/attempts/<attempt_num>#summary-<summary_id>

From what I can tell, the add_job_summary function is passed the same comment as was generated for the pull request, which includes the default max_files:

python-coverage-comment-action/coverage_comment/main.py

Lines 167 to 184 in e1188d5

	comment = template.get_comment_markdown(
	coverage=coverage,
	diff_coverage=diff_coverage,
	previous_coverage=previous_coverage,
	previous_coverage_rate=previous_coverage_rate,
	files=files_info,
	count_files=count_files,
	max_files=config.MAX_FILES_IN_COMMENT,
	minimum_green=config.MINIMUM_GREEN,
	minimum_orange=config.MINIMUM_ORANGE,
	repo_name=config.GITHUB_REPOSITORY,
	pr_number=config.GITHUB_PR_NUMBER,
	base_template=template.read_template_file("comment.md.j2"),
	custom_template=config.COMMENT_TEMPLATE,
	pr_targets_default_branch=pr_targets_default_branch,
	marker=marker,
	subproject_id=config.SUBPROJECT_ID,
	)

python-coverage-comment-action/coverage_comment/main.py

Lines 202 to 204 in e1188d5

	github.add_job_summary(
	content=comment, github_step_summary=config.GITHUB_STEP_SUMMARY
	)

I would be happy to create a pull request to resolve this if needed, I would propose to either re-generate the markdown without the max_files parameter or conditionally re-generate it if the number of files exceeds the max_files setting. Is there a preference from the authors/maintainers?

Also, it might be nifty to add a link to the workflow summary from the error message, I could try to take a crack at that too as part of this fix.

[ewjoachim]

Ah oops, I believe I completely forgot to generate the full report, you're right.

I would propose to either re-generate the markdown without the max_files parameter or conditionally re-generate it if the number of files exceeds the max_files setting. Is there a preference from the authors/maintainers?

I believe the generation is likely a few milliseconds, so either way, but likely generate it twice will lead to the clearest code.

Also, it might be nifty to add a link to the workflow summary from the error message, I could try to take a crack at that too as part of this fix.

Good idea!

And thank you for your proposal of doing the PR :) It will be a pleasure!

[tgehrs]

Closed with #444