[SAY-5]

read_po (and PoFileParser) now take an optional shared_msgids dict. Identical message IDs within one file always share a string object, and passing the same dict to several read_po calls shares them between catalogs too, which is what #571 is after, without sys.intern or any free-threading check.

Refs #571.

[akx]

Instead of using sys.intern, what about just having a temporary dict of these strings within the parser? That would work for free-threaded builds too.

[SAY-5]

Switched to a dict on the parser, dropped sys.intern and the build check, and cut the tests down to a single case in test_pofile_read.py. It only dedups within one read_po call now (e.g. the same msgid under different contexts), so separately loaded catalogs no longer share IDs.

[akx]

Hmmmn, sorry, I had forgotten to actually read the context of #571 before suggesting the per-parser dict. I'd just rather not like to have to probe around for whether we're doing free-threading...

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 93.51%. Comparing base (9a3ef47) to head (4da5393).
⚠️ Report is 6 commits behind head on master.

Additional details and impacted files

@@           Coverage Diff           @@
##           master    #1395   +/-   ##
=======================================
  Coverage   93.51%   93.51%           
=======================================
  Files          27       27           
  Lines        4932     4936    +4     
=======================================
+ Hits         4612     4616    +4     
  Misses        320      320           

Flag	Coverage Δ
macos-14-3.10	92.60% <100.00%> (+<0.01%)	⬆️
macos-14-3.11	92.54% <100.00%> (+<0.01%)	⬆️
macos-14-3.12	92.74% <100.00%> (+<0.01%)	⬆️
macos-14-3.13	92.74% <100.00%> (+<0.01%)	⬆️
macos-14-3.14	92.72% <100.00%> (+<0.01%)	⬆️
macos-14-3.15	92.72% <100.00%> (+<0.01%)	⬆️
macos-14-3.8	92.47% <100.00%> (+<0.01%)	⬆️
macos-14-3.9	92.54% <100.00%> (+<0.01%)	⬆️
macos-14-pypy3.10	92.60% <100.00%> (+<0.01%)	⬆️
ubuntu-24.04-3.10	92.62% <100.00%> (+<0.01%)	⬆️
ubuntu-24.04-3.11	92.56% <100.00%> (+<0.01%)	⬆️
ubuntu-24.04-3.12	92.76% <100.00%> (+<0.01%)	⬆️
ubuntu-24.04-3.13	92.76% <100.00%> (+<0.01%)	⬆️
ubuntu-24.04-3.14	92.74% <100.00%> (+<0.01%)	⬆️
ubuntu-24.04-3.15	92.74% <100.00%> (+<0.01%)	⬆️
ubuntu-24.04-3.8	92.49% <100.00%> (+<0.01%)	⬆️
ubuntu-24.04-3.9	92.56% <100.00%> (+<0.01%)	⬆️
ubuntu-24.04-pypy3.10	92.62% <100.00%> (+<0.01%)	⬆️
windows-2022-3.10	92.61% <100.00%> (+<0.01%)	⬆️
windows-2022-3.11	92.55% <100.00%> (+<0.01%)	⬆️
windows-2022-3.12	92.76% <100.00%> (+<0.01%)	⬆️
windows-2022-3.13	92.76% <100.00%> (+<0.01%)	⬆️
windows-2022-3.14	92.73% <100.00%> (+<0.01%)	⬆️
windows-2022-3.15	92.73% <100.00%> (+<0.01%)	⬆️
windows-2022-3.8	92.59% <100.00%> (+<0.01%)	⬆️
windows-2022-3.9	92.55% <100.00%> (+<0.01%)	⬆️
windows-2022-pypy3.10	92.61% <100.00%> (+<0.01%)	⬆️

Flags with carried forward coverage won't be shown. Click here to find out more.

☔ 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.
• 📦 JS Bundle Analysis: Save yourself from yourself by tracking and limiting bundle sizes in JS merges.

[codspeed]

Merging this PR will not alter performance

✅ 88 untouched benchmarks

---

_{Comparing SAY-5:fix/po-message-id-sharing-20261018 (79b1b5a) with master (3072a70)¹}

Footnotes

1. No successful run was found on master (19d490d) during the generation of this report, so 3072a70 was used instead as the comparison base. There might be some changes unrelated to this pull request in this report. ↩

[SAY-5]

No worries. read_po now takes an optional shared_msgids dict (the parser still uses its own one by default), so passing the same dict for every catalog shares the IDs across them like the snippet in #571, with no sys.intern or free-threading check.