perf: parse w3c traceparent using a custom parser

[dmehala]

Description

std::regex is notoriously known to be slow. This custom parser implementation offers significantly better performance.

Benchmark

Bench: https://quick-bench.com/q/XJw86Zh_SBDpQ3xhI2nrY-k-Xbc
Result:

TODO

• fuzz me!

[zacharycmontoya]

LGTM

[pablomartinezbernardo]

That benchmark 😮

Apart from fuzzing, it would be cool to get some unit tests for those invalid values as well as a valid one.

[pr-commenter]

Benchmarks

Benchmark execution time: 2025-01-27 09:59:08

Comparing candidate commit 07b7518 in PR branch dmehala/faster-w3c-parsing with baseline commit 6971362 in branch main.

Found 0 performance improvements and 0 performance regressions! Performance is the same for 1 metrics, 0 unstable metrics.

[pablomartinezbernardo]

That benchmark 😮

Apart from fuzzing, it would be cool to get some unit tests for those invalid values as well as a valid one.

Oh, we did have unit tests, sorry 😓

[codecov-commenter]

Codecov Report

Attention: Patch coverage is 92.30769% with 4 lines in your changes missing coverage. Please review.

Project coverage is 93.81%. Comparing base (6971362) to head (07b7518).
Report is 1 commits behind head on main.

Files with missing lines	Patch %	Lines
src/datadog/w3c_propagation.cpp	92.30%	4 Missing ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##             main     #178      +/-   ##
==========================================
- Coverage   93.91%   93.81%   -0.11%     
==========================================
  Files          73       73              
  Lines        4176     4202      +26     
==========================================
+ Hits         3922     3942      +20     
- Misses        254      260       +6     

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

[dmehala]

Benchmarks

Benchmark execution time: 2025-01-27 09:59:08

Comparing candidate commit 07b7518 in PR branch dmehala/faster-w3c-parsing with baseline commit 6971362 in branch main.

Found 0 performance improvements and 0 performance regressions! Performance is the same for 1 metrics, 0 unstable metrics.

Don't be fooled by the bot, this is expected. Our macro benchmark don't measure context propagation... which is something we probably should ;)