Skip to main content

🧪 Local Testing & Quality Assurance

TGDCFS maintains high test coverage and strict type safety. All code changes must pass unit tests, type checks, and linter rules before being merged.

Testing Pipeline Overview​

Commands to run before submitting any code:

# 1. Run unit tests
poetry run pytest

# 2. Run static type checker
poetry run mypy .

# 3. Run linter and code style checks
poetry run ruff check .

1. Running Unit Tests with pytest​

The test suite contains over 57 test modules covering all subsystems (tests/).

Basic Test Invocations​

Run the complete test suite:

poetry run pytest

Run a specific test file:

poetry run pytest tests/test_cache.py

Run a specific test function with verbose output:

poetry run pytest tests/test_replication.py -k "test_queue_retry" -v

Generate test coverage report:

poetry run pytest --cov=tgdcfs --cov-report=term-missing

2. Testing with In-Memory Fake Stores (tests/fakes/store.py)​

You do not need live Telegram or Discord API tokens to write or run unit tests!

TGDCFS provides a feature-rich in-memory fake store implementation in tests/fakes/store.py:

  • FakeStore: Implements IStore entirely in memory.
  • Configurable capabilities (max_part_bytes, max_text_chars, supports_server_copy).
  • Fault injection: Simulate network timeouts, FloodWait rate limits, transient channel errors, or dead stores.

Example unit test using FakeStore:

import pytest
from tests.fakes.store import FakeStore
from tgdcfs.backends.base import StoreCapabilities

@pytest.mark.asyncio
async def test_fake_store_upload_and_download():
    # Construct a fake store simulating a Discord channel (10 MB parts)
    caps = StoreCapabilities(
        max_part_bytes=10_000_000,
        max_text_chars=2000,
        supports_server_copy=False,
        supports_edit_media=True,
        supports_bulk_delete=True,
        delete_age_limit_days=14,
    )
    store = FakeStore(key="dc:123456", caps=caps)

    # Test uploading text and binary data
    msg_id = await store.send_text("Hello TGDCFS")
    retrieved_text = await store.get_text(msg_id)
    assert retrieved_text == "Hello TGDCFS"

3. Static Typechecking with mypy​

TGDCFS uses strict mypy configuration for full static type coverage:

poetry run mypy .
# or use the Makefile shortcut
make mypy

Ensure no type errors (Success: no issues found in XX source files) remain.

4. Linting & Code Style with ruff​

TGDCFS uses Ruff for fast linting and code formatting:

# Check code for lint errors
poetry run ruff check .

# Auto-fix fixable lint errors
poetry run ruff check --fix .

# Makefile shortcut
make ruff

5. Docker Smoke Check Script (scripts/docker_smoke_check.py)​

To verify that a built Docker image or container starts properly and serves SFTP/WebDAV requests:

docker run --rm -v "$PWD/scripts:/app/scripts:ro" xyvran/tgdcfs:latest \
  python scripts/docker_smoke_check.py