Support for parsing parial JSON strings in Python

[samuelcolvin]

As requested by @jxnl, add allow_partial argument to python parsing.

[codecov]

Codecov Report

Merging #66 (c3f6b6b) into main (4dd0c35) will increase coverage by 0.60%.
The diff coverage is 94.59%.

Additional details and impacted files

@@            Coverage Diff             @@
##             main      #66      +/-   ##
==========================================
+ Coverage   95.74%   96.35%   +0.60%     
==========================================
  Files           8        8              
  Lines        1199     1234      +35     
==========================================
+ Hits         1148     1189      +41     
+ Misses         51       45       -6     

Files	Coverage Δ
src/python.rs	95.93% <94.59%> (+0.80%)	⬆️

... and 6 files with indirect coverage changes

---

Continue to review full report in Codecov by Sentry.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update 4dd0c35...c3f6b6b. Read the comment docs.

[codspeed-hq]

CodSpeed Performance Report

Merging #66 will not alter performance

_{Comparing partial-parsing (c3f6b6b) with main (4dd0c35)}

Summary

✅ 56 untouched benchmarks

[jxnl]

Thanks, with this we'll be able to stream a little bit cleaner.

[samuelcolvin]

I need to update the docstring above.

[davidhewitt]

Nice, overall looks good! Couple of questions.

I also wonder, just to check I understand. The idea is that users will keep the buffer of their partial JSON, so when the next chunk arrives they concatenate those and get jiter to re-parse the lot?

I.e. it sees to me that this doesn't allow the parser to "save" state and resume with the next chunk of bytes, which might be nice for a true streaming implementation.

[samuelcolvin]

I.e. it sees to me that this doesn't allow the parser to "save" state and resume with the next chunk of bytes, which might be nice for a true streaming implementation.

That's correct. We could perhaps add support for saving the state, but that's a much bigger issue and should be done later.

[davidhewitt]

Happy overall subject to any docstring updates / refactoring you see fit to do before merge.