[jbachorik]

What does this PR do?

Replaces NUL-delimited string cache identities with an immutable structural key containing the
source pattern, flags, and enabled ReggieOption values.

The structural key is used consistently by the level-1 matcher cache and all runtime
engine/factory caches. Annotation-stub backtracker entries now have a separate string-keyed cache,
matching the existing PikeVM stub/runtime separation.

Regression coverage exercises the formerly colliding identities in both call orders through:

• the shared level-1 matcher cache;
• mutable hybrid/factory reconstruction;
• the runtime backtracking cache, including JDK capture-span parity;
• explicit cached(key, pattern, options) validation.

Motivation

NUL is a supported pattern literal, so concatenating option names to the pattern with a NUL
separator is not injective. For example, these two identities previously produced the same engine
key:

pattern="a", options=CAPTURE_NAMED_ONLY
pattern="a\0CAPTURE_NAMED_ONLY", options=DEFAULT

That could return a matcher compiled for a different pattern instead of compiling independently or
throwing for reuse of an explicit user key.

This follows up the finding on #149:
#149 (comment)

Knowledge constraint consumed: kb/find-nul-pattern-truncation — pattern-side NUL remains valid
data and must not be treated as a sentinel or separator.

Related Issue(s)

Follow-up to #149.

Change Type

• Bug fix
• New feature
• Performance improvement
• Refactoring (no functional change)
• Documentation
• Test improvement
• Build/CI change

Checklist

• I have read the CONTRIBUTING.md guidelines
• All existing tests pass (./gradlew build)
• I have added tests for my changes
• I have updated documentation (if applicable)
• My commits are signed

Performance Impact

No matching-path impact is expected. Compilation-cache lookup now hashes a small immutable value
object instead of a delimiter-encoded string. Default and non-default option identities use the
same structural representation.

Additional Notes

Verification completed:

• focused cache-key, user-cache, cache-hygiene, diagnostic, and shared-DFA tests;
• full :reggie-runtime:test suite (3,667 tests);
• ./gradlew build;
• :reggie-integration-tests:test;
• spotlessApply / spotlessCheck;
• git diff --check.

[chatgpt-codex-connector]

Codex Review Summary

This comment shows the latest Codex review activity on this pull request.

Review	Status	Commit	Review trigger
📝 Code Review	✅ Completed 2026-10-19T19:59:25.275571Z	35724ba	PR opened

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

• Open a pull request for review
• Mark a draft as ready
• Comment "@codex review" or "@codex security review".

Codex reacts with 👀 while any review is running, comments if it has suggestions, and reacts with 👍 once all reviews finish with no findings.

[datadog-prod-us1-4]

More details

The structural keys keep literal-NUL patterns distinct from option-bearing compilations across runtime caches, including explicit user-key validation and mutable matcher reconstruction. The completed static review identified no reportable regression.

Was this helpful? React 👍 or 👎

_{Open Bits AI session}

_{🤖 Bits Code Review · Commit 35724ba}

[codecov-commenter]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 85.0%. Comparing base (e37366f) to head (35724ba).

Additional details and impacted files

@@           Coverage Diff           @@
##              main    #152   +/-   ##
=======================================
  Coverage     85.0%   85.0%           
  Complexity       1       1           
=======================================
  Files          168     168           
  Lines        49878   49882    +4     
  Branches      7594    7592    -2     
=======================================
+ Hits         42409   42417    +8     
+ Misses        5466    5464    -2     
+ Partials      2003    2001    -2     

Files with missing lines	Coverage Δ
.../com/datadoghq/reggie/runtime/RuntimeCompiler.java	82.9% <100.0%> (+0.2%)	⬆️

... and 3 files with indirect coverage changes

---

Continue to review full report in Codecov by Harness.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update e37366f...35724ba. Read the comment docs.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.