feat: auto-generate Pydantic models from JSON schema & add CI + linting

[cka-y]

Summary:

Automates diff output model generation and adds CI infrastructure.

• Add generate_models.sh script to auto-generate Pydantic v2 models from the GTFS Diff JSON Schema using datamodel-code-generator
• Add schema.conf to pin the schema version (v2-rc1)
• Add CI workflow with path-filtered jobs: tests (Python 3.10/3.12/3.14), lint check, and models-freshness guard
• Add Ruff linting & formatting config with lint.sh / lint-fix.sh scripts
• Update engine logic to align with the v2-rc1 schema
• Apply lint formatting fixes across cli.py, gtfs_definitions.py, and engine.py

Closes #8, closes #9

[cka-y]

This file is used to automatically generate the Pydantic models

[cka-y]

All changes are only lint related

[cka-y]

Changes here are lint related + updating some of the logic to match the newly updated v2-rc1 schema. No additional computation logic was added.

[cka-y]

Lint updates exclusively

[Copilot]

Pull request overview

This PR modernizes the project’s GTFS Diff v2 output handling by switching to schema-driven (auto-generated) Pydantic v2 models, aligning engine output with the v2-rc1 schema, and adding CI + Ruff-based lint/format tooling to keep models and style consistent.

Changes:

• Add scripts/generate_models.sh + schema.conf to generate and pin src/gtfs_diff/models.py from the upstream JSON Schema.
• Update engine + tests to match the updated schema structure (notably moving per-file counts into file_diffs[].stats).
• Add Ruff configuration/scripts and a path-filtered GitHub Actions workflow (tests, lint, models freshness).

Reviewed changes

Copilot reviewed 14 out of 14 changed files in this pull request and generated 3 comments.

Show a summary per file

File	Description
src/gtfs_diff/models.py	Replace hand-written models with auto-generated schema-derived Pydantic v2 models.
scripts/generate_models.sh	Add a generator script to fetch schema + run datamodel-codegen and post-process the output.
schema.conf	Pin the schema version used for generation (v2-rc1).
src/gtfs_diff/engine.py	Align output to new schema (introduce FileStats, compute totals from stats, read schema version).
src/gtfs_diff/cli.py	Ruff-driven formatting and CLI output formatting tweaks.
src/gtfs_diff/gtfs_definitions.py	Formatting-only changes and wrapped docstring.
tests/test_engine.py	Update assertions to use file_diffs[].stats and apply formatting.
tests/test_models.py	Update/extend model tests for new schema fields (FileStats, files_not_compared_count, etc.).
tests/test_cli.py	Formatting updates for fixture zips and imports.
tests/conftest.py	Minor whitespace cleanup.
scripts/lint.sh / scripts/lint-fix.sh	Add lint/format check and auto-fix scripts via Ruff.
pyproject.toml	Add Ruff + datamodel-code-generator (dev deps) and configure Ruff.
.github/workflows/ci.yml	Add CI jobs for lint, tests (matrix), and model freshness enforcement.

Comments suppressed due to low confidence (1)

src/gtfs_diff/engine.py:480

• This comment is now inaccurate: per-file counts are no longer surfaced in summary.files (FileSummary only contains file_name/status). The counts are now in file_diffs[].stats.

    # Determine row-changes output based on cap.
    # cap=0 means "summary counts only" — row_changes is omitted from the output
    # entirely (serialized as null, excluded by exclude_none=True in the CLI).
    # True counts are still computed and surfaced in summary.files.
    base_header_set = set(base_headers)

---

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

[Copilot]

Pull request overview

Copilot reviewed 14 out of 14 changed files in this pull request and generated 2 comments.

[jcpitre]

LGTM!