Skip to content

AGENTS.md

Latest commit

 

History

History
438 lines (339 loc) · 11.5 KB

AGENTS.md

File metadata and controls

438 lines (339 loc) · 11.5 KB

CLAUDE.md

This file provides guidance to Claude Code (claude.ai/code) when working with code in this repository.

CRITICAL REQUIREMENTS

Test Success

  • ALL tests MUST pass for code to be considered complete and working
  • Never describe code as "working as expected" if there are ANY failing tests
  • Even if specific feature tests pass, failing tests elsewhere indicate broken functionality
  • Changes that break existing tests must be fixed before considering implementation complete
  • A successful implementation must pass linting, type checking, AND all existing tests

Project Overview

libvcs is a lite, typed Python tool for:

  • Detecting and parsing URLs for Git, Mercurial, and Subversion repositories
  • Providing command abstractions for git, hg, and svn
  • Synchronizing repositories locally
  • Creating pytest fixtures for testing with temporary repositories

The library powers vcspull, a tool for managing and synchronizing multiple git, svn, and mercurial repositories.

Development Environment

This project uses:

  • Python 3.9+
  • uv for dependency management
  • ruff for linting and formatting
  • mypy for type checking
  • pytest for testing

Common Commands

Setting Up Environment

# Install dependencies
uv pip install --editable .
uv pip sync

# Install with development dependencies
uv pip install --editable . -G dev

Running Tests

# Run all tests
just test
# or directly with pytest
uv run pytest

# Run a single test file
uv run pytest tests/sync/test_git.py

# Run a specific test
uv run pytest tests/sync/test_git.py::test_remotes

# Run tests with test watcher
just start
# or
uv run ptw .

Linting and Type Checking

# Run ruff for linting
just ruff
# or directly
uv run ruff check .

# Format code with ruff
just ruff-format
# or directly
uv run ruff format .

# Run ruff linting with auto-fixes
uv run ruff check . --fix --show-fixes

# Run mypy for type checking
just mypy
# or directly
uv run mypy src tests

# Watch mode for linting (using entr)
just watch-ruff
just watch-mypy

Development Workflow

Follow this workflow for code changes:

  1. Format First: uv run ruff format .
  2. Run Tests: uv run pytest
  3. Run Linting: uv run ruff check . --fix --show-fixes
  4. Check Types: uv run mypy
  5. Verify Tests Again: uv run pytest

Documentation

# Build documentation
just build-docs

# Start documentation server with auto-reload
just start-docs

# Update documentation CSS/JS
just design-docs

Code Architecture

libvcs is organized into three main modules:

  1. URL Detection and Parsing (libvcs.url)

    • Base URL classes in url/base.py
    • VCS-specific implementations in url/git.py, url/hg.py, and url/svn.py
    • URL registry in url/registry.py
    • Constants in url/constants.py

  2. Command Abstraction (libvcs.cmd)

    • Command classes for git, hg, and svn in cmd/git.py, cmd/hg.py, and cmd/svn.py
    • Built on top of Python's subprocess module (via _internal/subprocess.py)

  3. Repository Synchronization (libvcs.sync)

    • Base sync classes in sync/base.py
    • VCS-specific sync implementations in sync/git.py, sync/hg.py, and sync/svn.py

  4. Internal Utilities (libvcs._internal)

    • Subprocess wrappers in _internal/subprocess.py
    • Data structures in _internal/dataclasses.py and _internal/query_list.py
    • Runtime helpers in _internal/run.py and _internal/shortcuts.py

  5. pytest Plugin (libvcs.pytest_plugin)

    • Provides fixtures for creating temporary repositories for testing

Testing Strategy

libvcs uses pytest for testing with many custom fixtures. The pytest plugin (pytest_plugin.py) defines fixtures for creating temporary repositories for testing. These include:

  • create_git_remote_repo: Creates a git repository for testing
  • create_hg_remote_repo: Creates a Mercurial repository for testing
  • create_svn_remote_repo: Creates a Subversion repository for testing
  • git_repo, svn_repo, hg_repo: Pre-made repository instances
  • set_home, gitconfig, hgconfig, git_commit_envvars: Environment fixtures

These fixtures handle setup and teardown automatically, creating isolated test environments.

For running tests with actual VCS commands, tests will be skipped if the corresponding VCS binary is not installed.

Example Fixture Usage

def test_repo_sync(git_repo):
    # git_repo is already a GitSync instance with a clean repository
    # Use it directly in your tests
    assert git_repo.get_revision() == "initial"

Parameterized Tests

Use typing.NamedTuple for parameterized tests:

class RepoFixture(t.NamedTuple):
    test_id: str  # For test naming
    repo_args: dict[str, t.Any]
    expected_result: str

@pytest.mark.parametrize(
    list(RepoFixture._fields),
    REPO_FIXTURES,
    ids=[test.test_id for test in REPO_FIXTURES],
)
def test_sync(
    # Parameters and fixtures...
):
    # Test implementation

Coding Standards

Imports

  • Use namespace imports: import enum instead of from enum import Enum
  • For typing, use import typing as t and access via namespace: t.NamedTuple, etc.
  • Use from __future__ import annotations at the top of all Python files

Docstrings

Follow NumPy docstring style for all functions and methods:

"""Short description of the function or class.

Detailed description using reStructuredText format.

Parameters
----------
param1 : type
    Description of param1
param2 : type
    Description of param2

Returns
-------
type
    Description of return value
"""

Doctests

All functions and methods MUST have working doctests. Doctests serve as both documentation and tests.

CRITICAL RULES:

  • Doctests MUST actually execute - never comment out asyncio.run() or similar calls
  • Doctests MUST NOT be converted to .. code-block:: as a workaround (code-blocks don't run)
  • If you cannot create a working doctest, STOP and ask for help

Available tools for doctests:

  • doctest_namespace fixtures: tmp_path, asyncio, create_git_remote_repo, create_hg_remote_repo, create_svn_remote_repo, example_git_repo
  • Ellipsis for variable output: # doctest: +ELLIPSIS
  • Update pytest_plugin.py to add new fixtures to doctest_namespace

# doctest: +SKIP is NOT permitted - it's just another workaround that doesn't test anything. If a VCS binary might not be installed, pytest already handles skipping via skip_if_binaries_missing. Use the fixtures properly.

Async doctest pattern:

>>> async def example():
...     result = await some_async_function()
...     return result
>>> asyncio.run(example())
'expected output'

Using fixtures in doctests:

>>> git = Git(path=tmp_path)  # tmp_path from doctest_namespace
>>> git.run(['status'])
'...'

When output varies, use ellipsis:

>>> git.clone(url=f'file://{create_git_remote_repo()}')  # doctest: +ELLIPSIS
'Cloning into ...'

Git Commit Standards

Format commit messages as:

Component/File(commit-type[Subcomponent/method]): Concise description

why: Explanation of necessity or impact.
what:
- Specific technical changes made
- Focused on a single topic

Common commit types:

  • feat: New features or enhancements
  • fix: Bug fixes
  • refactor: Code restructuring without functional change
  • docs: Documentation updates
  • chore: Maintenance (dependencies, tooling, config)
  • test: Test-related updates
  • style: Code style and formatting

Example:

url/git(feat[GitURL]): Add support for custom SSH port syntax

why: Enable parsing of Git URLs with custom SSH ports
what:
- Add port capture to SCP_REGEX pattern
- Update GitURL.to_url() to include port if specified
- Add tests for the new functionality

For multi-line commits, use heredoc to preserve formatting:

git commit -m "$(cat <<'EOF'
feat(Component[method]) add feature description

why: Explanation of the change.
what:
- First change
- Second change
EOF
)"

Asyncio Development

Architecture

libvcs async support is organized in _async/ subpackages:

libvcs/
├── _internal/
│   ├── subprocess.py         # Sync subprocess wrapper
│   └── async_subprocess.py   # Async subprocess wrapper
├── cmd/
│   ├── git.py                # Git (sync)
│   └── _async/git.py         # AsyncGit
├── sync/
│   ├── git.py                # GitSync (sync)
│   └── _async/git.py         # AsyncGitSync

Async Subprocess Patterns

Always use communicate() for subprocess I/O:

proc = await asyncio.create_subprocess_shell(...)
stdout, stderr = await proc.communicate()  # Prevents deadlocks

Use asyncio.timeout() for timeouts:

async with asyncio.timeout(300):
    stdout, stderr = await proc.communicate()

Handle BrokenPipeError gracefully:

try:
    proc.stdin.write(data)
    await proc.stdin.drain()
except BrokenPipeError:
    pass  # Process already exited - expected behavior

Async API Conventions

  • Class naming: Use Async prefix: AsyncGit, AsyncGitSync
  • Callbacks: Async APIs accept only async callbacks (no union types)
  • Shared logic: Extract argument-building to sync functions, share with async
# Shared argument building (sync)
def build_clone_args(url: str, depth: int | None = None) -> list[str]:
    args = ["clone", url]
    if depth:
        args.extend(["--depth", str(depth)])
    return args

# Async method uses shared logic
async def clone(self, url: str, depth: int | None = None) -> str:
    args = build_clone_args(url, depth)
    return await self.run(args)

Async Testing

pytest configuration:

[tool.pytest.ini_options]
asyncio_mode = "strict"
asyncio_default_fixture_loop_scope = "function"

Async fixture pattern:

@pytest_asyncio.fixture(loop_scope="function")
async def async_git_repo(tmp_path: Path) -> t.AsyncGenerator[AsyncGitSync, None]:
    repo = AsyncGitSync(url="...", path=tmp_path / "repo")
    await repo.obtain()
    yield repo

Parametrized async tests:

class CloneFixture(t.NamedTuple):
    test_id: str
    clone_kwargs: dict[str, t.Any]
    expected: list[str]

CLONE_FIXTURES = [
    CloneFixture("basic", {}, [".git"]),
    CloneFixture("shallow", {"depth": 1}, [".git"]),
]

@pytest.mark.parametrize(
    list(CloneFixture._fields),
    CLONE_FIXTURES,
    ids=[f.test_id for f in CLONE_FIXTURES],
)
@pytest.mark.asyncio
async def test_clone(test_id: str, clone_kwargs: dict, expected: list) -> None:
    ...

Async Anti-Patterns

DON'T poll returncode:

# WRONG
while proc.returncode is None:
    await asyncio.sleep(0.1)

# RIGHT
await proc.wait()

DON'T mix blocking calls in async code:

# WRONG
async def bad():
    subprocess.run(["git", "clone", url])  # Blocks event loop!

# RIGHT
async def good():
    proc = await asyncio.create_subprocess_shell(...)
    await proc.wait()

DON'T close the event loop in tests:

# WRONG - breaks pytest-asyncio cleanup
loop = asyncio.get_running_loop()
loop.close()

Debugging Tips

When stuck in debugging loops:

  1. Pause and acknowledge the loop
  2. Minimize to MVP: Remove all debugging cruft and experimental code
  3. Document the issue comprehensively for a fresh approach
  4. Format for portability (using quadruple backticks)