ROB: Do not crash on a non-array destination

[eeshsaxena]

_build_destination already degrades a null, string, short-array or missing destination to a null destination, but a destination that is some other single object slips past that check and hits page, typ, *array = array, raising TypeError: cannot unpack non-iterable ... object. I hit this reading a PDF whose /Names destination tree had a bare number where a destination array was expected, which crashes reader.named_destinations. Tightened the guard to require an array of at least two elements so anything else falls back the same way the other invalid cases already do.

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 98.00%. Comparing base (29f389c) to head (bbfbf0a).
⚠️ Report is 5 commits behind head on main.

Additional details and impacted files

@@           Coverage Diff           @@
##             main    #3976   +/-   ##
=======================================
  Coverage   98.00%   98.00%           
=======================================
  Files          57       57           
  Lines       11120    11128    +8     
  Branches     2084     2087    +3     
=======================================
+ Hits        10898    10906    +8     
  Misses        124      124           
  Partials       98       98           

☔ View full report in Codecov by Harness.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

[eeshsaxena]

Hi! Gentle nudge on this one whenever you have some bandwidth. It's a small, self-contained fix (ROB: Do not crash on a non-array destination), and it's currently mergeable with no conflicts. No urgency at all, and I'm happy to make any changes you'd like. Thanks for maintaining pypdf!

[stefan6419846]

Please see the inline remark for some possible changes.

Gentle nudge on this one whenever you have some bandwidth. It's a small, self-contained fix [...] and it's currently mergeable with no conflicts. No urgency at all, and I'm happy to make any changes you'd like.

Please note that the first part of the message does not sound like there is no urgency. Unless there really has been merge activity, but I missed your changes for some time, you can gently ping us. Otherwise, like after less than a day in this case, this just moves your PR down the queue.

[eeshsaxena]

Fair point on the ping, apologies for that. I'll address the inline remark and push the change, no rush needed on your side.

[stefan6419846]

Please review the code style issues.

[eeshsaxena]

Good catch, that was ruff D209 on the two new test docstrings. Moved the closing quotes to their own line and the code style check passes now. Happy to fold those inline comments into the docstrings too if you'd prefer that.

[eeshsaxena]

Thanks, addressed all three:

• Switched the check to isinstance(array, ArrayObject) instead of list. That matches how the rest of _doc_common.py types destination arrays (e.g. the isinstance(dest, ArrayObject) right above the call site), and let me drop the long inline union type hint along with the now-unused NumberObject import.
• Shortened the guard comment.
• Moved the short_array test note out of inline comments and into a trimmed docstring.

ruff check is clean, mypy is happy with the narrower type, and the build_destination / named_destinations tests pass locally.

[stefan6419846]

mypy is reporting an issue. Could you please review it?

[eeshsaxena]

Fixed. After rebasing on main, #3988 now types the destination tree, so Fit(fit_args=array) type-checks on its own and the # type: ignore[arg-type] was flagged as unused. Removed it. mypy . and ruff are clean and the destination tests pass.

And fair point on the earlier nudge, that was a bad habit on my part and the "no urgency" line did read as contradictory. I will leave it with you and not bump it.