Contributing to TAPA

Pull Request Process

TAPA uses GitHub Actions for continuous integration. The CI pipeline:

Builds binary distributions on Ubuntu 18.04 self-hosted runners.
Performs code quality checks using pre-commit hooks on every commit.
Runs functional and integration tests via staging workflows across a matrix of platforms and Vitis versions for every main branch push.

Update the documentation in the docs/src/ directory for any new features or changes.
Use Markdown format for documentation files.
Preview your changes locally with a live-reloading server. Bazel fetches mdBook for you, so nothing needs to be installed first:
```
bazel run //docs:serve
```
To produce the static HTML site instead (output: bazel-bin/docs/book.tar.gz):
```
bazel build //docs:build
```

Add appropriate unit tests for new features or bug fixes.
Ensure all existing tests pass before submitting your changes.
Run the full test suite using the following command:
```
bazel test //...
```
If you change the task-graph schema, note that it has two implementations that must agree: tapacc/tapacc.cpp emits the JSON, and the tapa-ir Rust crate parses it with deny_unknown_fields. The fixtures under tapa-core/testdata/ are hand-written and cannot catch the two drifting apart. //tapa-core:tapacc_conformance_test is what does: it runs the real tapacc on tests/apps/vadd/vadd.cpp and strict-parses its verbatim stdout. tapacc is a Clang tool that only builds on Linux, so the test is gated on @platforms//os:linux and is skipped everywhere else:
```
bazel test //tapa-core:tapacc_conformance_test  # Linux only
```