Fix Global Heap over-read on sub-header trailing padding

[ajelenak]

Description

GlobalHeap.objects looped while offset < len(self.heap_data), so it entered the loop with fewer than a full 16-byte GLOBAL_HEAP_OBJECT header remaining and then over-read the buffer. A collection's trailing free space is marked by an object with index 0, but libhdf5 writes that marker only when at least one 16-byte object header fits. When the remainder is smaller it is undefined padding.

Add tests/test_global_heap.py, which builds raw GCOL collections in memory and reads them through GlobalHeap, deterministically reproducing the crash (a <16-byte trailing pad) and covering the index-0 terminator and exactly-full cases.

Closes #259.

Checklist

• This pull request has a descriptive title and labels
• This pull request has a minimal description (most was discussed in the issue, but a two-liner description is still desirable)
• Unit tests have been added (if codecov test fails)
• All tests pass

[davidhassell]

Hi Aleksandar,

A discussed on the closed #260, this is the PR to close #259, and it all looks good to me.

It's better that #260 in that it has unit tests; and calculates, rather than assumes, the GLOBAL_HEAP_OBJECT_SIZE.

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 78.62%. Comparing base (667c468) to head (6b673fc).
⚠️ Report is 12 commits behind head on main.

Additional details and impacted files

@@            Coverage Diff             @@
##             main     #261      +/-   ##
==========================================
+ Coverage   78.48%   78.62%   +0.13%     
==========================================
  Files          15       15              
  Lines        3416     3485      +69     
  Branches      546      555       +9     
==========================================
+ Hits         2681     2740      +59     
- Misses        593      602       +9     
- Partials      142      143       +1     

☔ 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.

[ajelenak]

Thanks @davidhassell @bnlawrence!