Skip to content

CONTRIBUTING.md

Latest commit

 

History

History
executable file
·
146 lines (107 loc) · 4.21 KB

CONTRIBUTING.md

File metadata and controls

executable file
·
146 lines (107 loc) · 4.21 KB
Raw

Contributing

Contributions are welcome! Here's how to get going.

Setup

brew install mise
git clone https://github.com/RevenueCat/cli && cd cli
mise install          # installs Go 1.25 per mise.toml
go mod tidy
make install-hooks    # installs git hooks via lefthook (fmt + vet + test on commit)
make check            # verify everything passes clean

Dev loop

go run ./cmd/rc --help
go run ./cmd/rc commands --json | jq    # full surface
go run ./cmd/rc schema customers grant   # per-command schema

Build a local binary:

go build -o /tmp/rc ./cmd/rc
/tmp/rc --version

Running against the real API

Use env vars and an isolated config dir so dev runs don't touch your real profile:

export RC_API_KEY="sk_..."
export RC_PROJECT_ID="proj_..."
export RC_CONFIG_DIR="$(mktemp -d)"

go run ./cmd/rc projects list
go run ./cmd/rc customers show <id>

When done: unset RC_API_KEY and revoke the key in the dashboard.

Running tests

make test       # go test -race ./...
make cover      # coverage report
make check      # fmt + vet + test (run before pushing)

Tests use scrubbed fixtures from internal/api/testdata/v2/ — no API key needed.

Capturing new fixtures

When the API surface changes, capture and scrub new responses:

# 1. Capture with curl
mkdir /tmp/rc-fixtures
curl -s -H "Authorization: Bearer $RC_API_KEY" \
  https://api.revenuecat.com/v2/projects/$PROJ/<path> \
  > /tmp/rc-fixtures/<name>.json

# 2. Scrub (replaces real IDs, emails, IPs with deterministic fakes)
go run ./internal/api/testdata/scrub \
  -in /tmp/rc-fixtures \
  -out internal/api/testdata/v2

# 3. Verify no leaks
git diff internal/api/testdata/v2/
grep -r "$REAL_ID" internal/api/testdata/v2/ && echo "LEAK"

Make targets

Target What it does
make build go build ./...
make test go test -race ./...
make cover Coverage report
make fmt gofmt -w
make fmt-check Fails if anything's unformatted (CI parity)
make vet go vet
make lint staticcheck
make check fmt-check + vet + test
make tidy go mod tidy

Cutting a release

Releases are automated — maintainers only need to push a tag:

# Make sure main is green, then:
git tag v0.1.0 && git push --tags

GitHub Actions picks up the tag, runs GoReleaser, publishes binaries to the GitHub release, and updates the Homebrew formula in RevenueCat/homebrew-tap automatically.

One-time setup: add a HOMEBREW_TAP_GITHUB_TOKEN repo secret with write access to RevenueCat/homebrew-tap.

Use semver: v0.x.y while pre-1.0, v1.x.y after stable release.

Commit messages

This repo uses Conventional Commits. The pre-commit hook enforces it automatically after make install-hooks.

<type>(<scope>): <description>

Types: feat  fix  docs  refactor  perf  test  chore  ci  build  revert

Examples:

feat(auth): add OAuth browser login
fix(charts): filter future cohort periods
docs: update README quick start

Release notes are auto-generated from commit history — feat and fix commits show up in the GitHub release, others are filtered out.

Sending a PR

  1. Check AGENTS.md — it documents the architecture and conventions.
  2. Check docs/command-surface.md before adding or renaming a command.
  3. Open an issue first if you're unsure whether something's in scope.
  4. Make sure make check passes before pushing.
  5. Use conventional commit messages — the hook will tell you if the format is wrong.
  6. Open a PR with a clear description of what changed and why.

Architecture overview

cmd/rc/          entry point
internal/
  api/           typed REST client — one file per resource, no CLI concepts
  cli/           user-intent commands — composes api/ calls into UX
  config/        profile + env layering (~/.config/revenuecat/*.json)
  output/        TTY pretty-print vs --json renderer
  tui/           interactive prompts (huh) and chart viewer (BubbleTea)

The key rule: internal/api/ knows nothing about the CLI. internal/cli/ owns all UX. See AGENTS.md for the full conventions.