Log Message:
py-json5: updated to 0.9.25

v0.9.25 (2024-04-12)

GitHub issue 81 Explicitly specify the directory to use for the package in \ 
pyproject.toml.

v0.9.24 (2024-03-16)

Update GitHub workflow config to remove unnecessary steps and run on pull \ 
requests as well as commits.
Added note about removing hypothesize in v0.9.23.
No code changes.

v0.9.23 (2024-03-16)

Lots of cleanup:
Removed old code needed for Python2 compatibility.
Removed tests using hypothesize. This ran model-based checks and didn't really \ 
add anything useful in terms of coverage to the test suite, and it introduced \ 
dependencies and slowed down the tests significantly. It was a good experiment \ 
but I think we're better off without it.
Got everything linting cleanly with pylint 3.1 and ruff check using ruff 0.3.3 \ 
(Note that commit message in 00d73a3 says pylint 3.11, which is a typo).
Code reformatted with ruff format
Added missing tests to bring coverage up to 100%.
Lots of minor code changes as the result of linting and coverage testing, but no \ 
intentional functional differences.

v0.9.22 (2024-03-06)

Attempt to fix the GitHub CI configuration now that setup.py is gone. Also, test \ 
on 3.12 instead of 3.11.
No code changes.

v0.9.21 (2024-03-06)

Moved the benchmarks/*.json data files' license information to //LICENSE to \ 
(hopefully) make the Google linter happy.

v0.9.20 (2024-03-03)

Added json5.__version__ in addition to json5.VERSION.
More packaging modernization (no more setup.{cfg,py} files).
Mark Python3.12 as supported in project.classifiers.
Updated the //run script to use python3.

v0.9.19 (2024-03-03)

Replaced the benchmarking data files that came from chromium.org with three \ 
files obtained from other datasets on GitHub. Since this repo is vendored into \ 
the chromium/src repo it was occasionally confusing people who thought the data \ 
was actually used for non-benchmarking purposes and thus updating it for \ 
whatever reason.
No code changes.

v0.9.18 (2024-02-29)

Add typing information to the module. This is kind of a big change, but there \ 
should be no functional differences.