pgschema / run-tests
Install for your project team
Run this command in your project directory to install the skill for your entire team:
mkdir -p .claude/skills/run-tests && curl -L -o skill.zip "https://fastmcp.me/Skills/Download/3574" && unzip -o skill.zip -d .claude/skills/run-tests && rm skill.zip
Project Skills
This skill will be saved in .claude/skills/run-tests/ and checked into git. All team members will have access to it automatically.
Important: Please verify the skill by reviewing its instructions before using it.
Run pgschema automated tests (go test) to validate diff logic, plan generation, and dump functionality using test fixtures
0 views
0 installs
Skill Content
--- name: run_tests description: Run pgschema automated tests (go test) to validate diff logic, plan generation, and dump functionality using test fixtures. Use this skill whenever you need to run tests, debug test failures, add new test cases, regenerate expected outputs, or validate changes across PostgreSQL versions 14-18. --- # Run Tests Run pgschema tests for validating implementation changes. ## Test Types | Type | Command | Speed | What it tests | |------|---------|-------|---------------| | Diff | `go test -v ./internal/diff -run TestDiffFromFiles` | ~2s | Schema comparison logic (no DB) | | Plan/Apply | `go test -v ./cmd -run TestPlanAndApply` | ~30-60s | Full workflow with embedded PostgreSQL | | Dump | `go test -v ./cmd/dump -run TestDumpCommand` | ~10-20s | Schema extraction from test databases | ## Filtering Tests Use `PGSCHEMA_TEST_FILTER` to scope to specific cases: ```bash # Specific test case PGSCHEMA_TEST_FILTER="create_trigger/add_trigger_when_distinct" go test -v ./internal/diff -run TestDiffFromFiles # All tests in a category PGSCHEMA_TEST_FILTER="create_trigger/" go test -v ./cmd -run TestPlanAndApply ``` **Categories** (in `testdata/diff/`): `comment/`, `create_domain/`, `create_function/`, `create_index/`, `create_materialized_view/`, `create_policy/`, `create_procedure/`, `create_sequence/`, `create_table/`, `create_trigger/`, `create_type/`, `create_view/`, `default_privilege/`, `privilege/`, `dependency/`, `online/`, `migrate/` ## Regenerating Expected Output When implementation intentionally changes generated DDL: ```bash PGSCHEMA_TEST_FILTER="category/test_name" go test -v ./cmd -run TestPlanAndApply --generate ``` This overwrites `diff.sql`, `plan.json`, `plan.sql`, `plan.txt`. Always review with `git diff` afterward and re-run without `--generate` to confirm. ## Testing Across PostgreSQL Versions ```bash PGSCHEMA_POSTGRES_VERSION=14 go test -v ./cmd/dump -run TestDumpCommand_Employee PGSCHEMA_POSTGRES_VERSION=17 PGSCHEMA_TEST_FILTER="create_trigger/" go test -v ./cmd -run TestPlanAndApply ``` Supported: 14, 15, 16, 17, 18. ## Adding a New Test Case 1. Create directory: `mkdir -p testdata/diff/<category>/<test_name>` 2. Create `old.sql` (starting state) and `new.sql` (desired state) 3. Generate expected files: `PGSCHEMA_TEST_FILTER="<category>/<test_name>" go test -v ./cmd -run TestPlanAndApply --generate` 4. Review generated `diff.sql` and plan files 5. Run without `--generate` to verify For dump tests, create `testdata/dump/<name>/` with `manifest.json`, `raw.sql`, `pgdump.sql`, `pgschema.sql` and register in `cmd/dump/dump_integration_test.go`. ## Test Data Structure ``` testdata/diff/<category>/<test_name>/ ├── old.sql # Starting schema ├── new.sql # Desired schema ├── diff.sql # Expected migration DDL ├── plan.json # Plan in JSON ├── plan.sql # Plan as SQL └── plan.txt # Plan as text ``` ## Recommended Testing Order 1. **Fast iteration**: diff tests first (no DB startup) 2. **Full validation**: integration tests for the same category 3. **Before commit**: `go test -v ./...` (or scoped to affected areas if full suite times out) ## Environment Variables | Variable | Purpose | Example | |----------|---------|---------| | `PGSCHEMA_TEST_FILTER` | Run specific test cases | `"create_trigger/"` | | `PGSCHEMA_POSTGRES_VERSION` | Test specific PG version | `14`, `17` |