[Copilot]

This PR introduces a reproducible performance baseline for parser APIs using existing syntax fixtures, with machine-readable output for historical comparison. Perf coverage is isolated from regular test/release pipelines via a dedicated pytest marker and workflow split.

• Perf benchmark module (tests/test_perf.py)

  • Adds a dedicated @pytest.mark.perf test that reuses valid_syntax.json / invalid_syntax.json.
  • Benchmarks explicit public parser entry points:
    • parse
    • is_valid_syntax
    • make_syntax_validator
    • RFC3987_SYNTAX_TERM_VALIDATORS[term]
    • exported is_valid_syntax_<term> validators (for fixture-backed terms)
  • Separates build/init timing from runtime parse timing.
  • Captures deterministic runtime metrics plus tracemalloc peak memory.
  • Emits structured JSON results (RFC3987_PERF_OUTPUT_JSON, default .pytest_perf/perf-results.json).

• Deterministic loop controls

  • Adds env-driven loop/warmup configuration:
    • RFC3987_PERF_LOOPS (default 1 for CI safety)
    • RFC3987_PERF_WARMUP_LOOPS (default 1)
  • No random input generation; corpus and execution path are fixed.

• Pytest marker registration

  • Registers perf marker in pyproject.toml to avoid unknown-marker warnings and enable clean selector-based execution.

• CI pipeline separation

  • Adds dedicated workflow: .github/workflows/perf.yml
    • Triggers: workflow_dispatch + schedule (weekly)
    • Uploads benchmark artifacts (artifacts/perf/...) for baseline retention/comparison
  • Updates existing workflows to exclude perf tests:
    • python-tests.yml: pytest -m "not perf" ...
    • python-publish.yml: pytest -m "not perf" ...
  • Keeps perf runs out of release/build/deploy flows.

• Docs / contributor guidance

  • Updates CONTRIBUTING.md with local perf execution, loop tuning, and JSON output path usage.

# Local perf run (default deterministic settings)
python -m pytest -m perf tests/test_perf.py

# Higher-volume baseline run with explicit output
RFC3987_PERF_LOOPS=1000 \
RFC3987_PERF_WARMUP_LOOPS=1 \
RFC3987_PERF_OUTPUT_JSON=.pytest_perf/perf-results.json \
python -m pytest -m perf tests/test_perf.py

[Copilot]

Pull request overview

Adds a deterministic, fixture-backed performance baseline suite for the RFC3987 parser/validator APIs, and isolates it from the normal CI/release pipelines via a dedicated perf pytest marker and a separate scheduled workflow that publishes machine-readable artifacts.

Changes:

• Introduces tests/test_perf.py to benchmark parser and validator entry points with deterministic loop controls, build/init vs runtime split, and JSON output (including tracemalloc peak memory).
• Registers the perf pytest marker and updates standard CI/test + publish workflows to exclude perf tests (-m "not perf").
• Adds a dedicated .github/workflows/perf.yml workflow (manual + weekly schedule) to run perf tests and upload benchmark artifacts; updates contributor docs and ignores local perf output directory.

Reviewed changes

Copilot reviewed 6 out of 7 changed files in this pull request and generated 1 comment.

Show a summary per file

File	Description
tests/test_perf.py	Adds deterministic perf benchmark test producing structured JSON metrics for parser/validator APIs.
pyproject.toml	Registers the perf pytest marker.
CONTRIBUTING.md	Documents how to run perf benchmarks locally and clarifies default test invocation excludes perf.
.gitignore	Ignores .pytest_perf/ benchmark output directory.
.github/workflows/python-tests.yml	Excludes perf tests from the main CI test runs.
.github/workflows/python-publish.yml	Excludes perf tests from the publish pipeline’s test step.
.github/workflows/perf.yml	Adds an isolated, scheduled/manual perf workflow that uploads benchmark artifacts.

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.