[ankit2235]

A ZIP made with macOS Finder ("Compress") also holds __MACOSX/…/._<name>
AppleDouble files and .DS_Store files. ZIP input converted each of them as a
document:

• .DS_Store failed, so convert_all() raised ConversionError by default
  and stopped the batch.
• __MACOSX/docs/._notes.md was converted "successfully" from binary
  metadata, which added a junk document to the results.

This change skips members under a top-level __MACOSX/ folder, members named
._*, and .DS_Store. Other unsupported members still give a failure result,
as before.

Tests: added test_macos_archive_metadata_is_skipped. It fails on main and
passes with this change. The existing ZIP tests pass unchanged.

[github-actions]

✅ DCO Check Passed

Thanks @ankit2235, all your commits are properly signed off. 🎉

[mergify]

Merge Protections

🟢 Merge protection satisfied — ready to merge.

Show 1 satisfied protection

🟢 Enforce conventional commit

Make sure that we follow https://www.conventionalcommits.org/en/v1.0.0/

• title ~= ^(fix|feat|docs|style|refactor|perf|test|build|ci|chore|revert)(?:\(.+\))?(!)?:

[github-actions]

AI triage (advisory)

Checked commit d43a6da680. The analysis read the diff only. It did not run code from this pull request.

The PR makes ZIP input skip macOS metadata members: files under a top-level __MACOSX/ folder, files named ._*, and .DS_Store. It adds tests for this and for a member with an empty name.

CI safety: no concern found. A maintainer must still approve CI.

_{The ai:* labels are advisory. They do not approve CI and they do not add tests:full. A new push runs this check again.}

[github-actions]

AI first review (advisory): blocker found

This review covers the full diff at 0df00701d9. It read the code and did not run it.

The main risk is an unhandled IndexError when a ZIP member has an empty filename and empty path parts. Check that path parts exist before index access, or use a membership check.

_{A maintainer makes the merge decision. This review never approves the pull request.}

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.

📢 Thoughts on this report? Let us know!

[dolfim-ibm]

Thanks for the fix, we like this change: macOS-made ZIPs are a common input, and the new test reproduces the problem well.

One blocker before merge, plus two small follow-ups.

Blocker: a ZIP member named "" or "." now aborts the whole batch. For these names PurePosixPath(...).parts is (), so path.parts[0] raises IndexError. The error is raised outside the try/except around the member read, so it escapes convert_all() even with raises_on_error=False. On main, the same archive yields its other members plus a failure result for the bad entry. Details and a suggested fix are in the inline comment.

Follow-ups (non-blocking):

• A regression case for a member with an empty name would protect the fix.
• ._* is matched at any depth, so a real file such as docs/._real.md is now dropped silently. That is fine for AppleDouble files, but a _log.debug line per skipped member would make it traceable.

[dolfim-ibm]

path.parts is empty for a member named "" or ".", so this raises IndexError: tuple index out of range and aborts convert_all() even with raises_on_error=False.

The bot's "__MACOSX" in path.parts would also skip a legitimate nested a/__MACOSX/ folder. A slice keeps the top-level-only scope and is safe on empty parts:

Suggested change

	path.parts[0] == "__MACOSX"
	path.parts[:1] == ("__MACOSX",)

[ankit2235]

Thanks @dolfim-ibm! Fixed in the latest commit:

• Used path.parts[:1] == ("__MACOSX",), so a member named "" or "." no longer raises, and only the top-level __MACOSX/ folder is skipped.
• Added test_member_with_an_empty_name_does_not_abort_the_batch.
• Each skipped member is now logged at debug level.