Add test suite for ddss/output.py

[Copilot]

The output.py module lacked test coverage. It continuously polls the database for Facts and Ideas entries, formats them with unparse() (e.g., a\n----\nb → a => b), and prints them. Testing was non-trivial due to the infinite loop.

Changes

• Test structure: Created tests/ directory with test_output.py using pytest and pytest-asyncio
• Test coverage:
  • Facts and Ideas formatting verification
  • Multiple entry handling
  • Graceful cancellation without hanging
• Test approach: Uses pytest features including async fixtures (@pytest_asyncio.fixture), async test support (@pytest.mark.asyncio), and the capsys fixture for stdout capture. Temporary databases and asyncio.Task.cancel() safely test the infinite loop.

# Example test pattern with pytest
@pytest.mark.asyncio
async def test_output_formats_facts_correctly(temp_db, capsys):
    addr, engine, session = temp_db
    async with session() as sess:
        sess.add(Facts(data="a\n----\nb"))
        await sess.commit()
    
    task = asyncio.create_task(main(addr, engine, session))
    await asyncio.sleep(0.2)
    task.cancel()
    
    captured = capsys.readouterr()
    assert "fact: a => b" in captured.out

Original prompt

---

This section details on the original issue you should resolve

<issue_title>给ddss/output.py写一个测试</issue_title>
<issue_description>其他文件的测试在后续的pr里，先不管。

比如在sql中写入一个

a

b

这样的东西

output会输出

a => b

这样的东西。

需要注意output是一个死循环，平时需要ctrl c之类的东西才会中断，写测试的时候注意别死循环了。最好是看python -m ddss.output sqlite+aiosqlite:///./data.db 这样的命令的输出</issue_description>

Comments on the Issue (you are @copilot in this section)

• Fixes 给ddss/output.py写一个测试 #11

---

💡 You can make Copilot smarter by setting up custom instructions, customizing its development environment and configuring Model Context Protocol (MCP) servers. Learn more Copilot coding agent tips in the docs.

[hzhangxyz]

改用pytest吧，如果可以用pytest的特性就用其高级特性，不行就直接用pytest

[Copilot]

改用pytest吧，如果可以用pytest的特性就用其高级特性，不行就直接用pytest

Converted tests to pytest with pytest-asyncio. The changes leverage:

• @pytest_asyncio.fixture for async database setup
• @pytest.mark.asyncio for async test functions
• capsys fixture for stdout capture (replacing manual StringIO)
• Native async/await support

Commit: a07d2fe

[Copilot]

Pull request overview

This PR adds a comprehensive test suite for the ddss/output.py module, which continuously polls the database for Facts and Ideas entries and outputs them in a formatted way. The testing approach successfully handles the challenge of testing an infinite loop by using asyncio task cancellation.

Key Changes

• Created tests/ directory with test_output.py containing 4 test cases covering Facts/Ideas formatting, multiple entries, and graceful cancellation
• Added development dependencies (pytest, pytest-asyncio, pytest-cov, ruff) to pyproject.toml and uv.lock
• Tests use async fixtures and task cancellation to safely test the infinite polling loop without hanging

Reviewed changes

Copilot reviewed 2 out of 3 changed files in this pull request and generated 6 comments.

File	Description
tests/test_output.py	New test file with 4 test cases for output formatting and cancellation behavior
pyproject.toml	Added dev optional dependencies for testing and linting tools
uv.lock	Updated lock file with new test dependencies (pytest, pytest-asyncio, pytest-cov, ruff, colorama, coverage, iniconfig, packaging, pluggy, pygments, tomli)

---

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

[Copilot]

The test fixture disposes the engine after yielding, but the main() function being tested also disposes the engine in its finally block (line 37 of output.py). This could lead to a "double dispose" situation where the engine is disposed twice, potentially causing warnings or errors.

Consider either:

1. Not disposing the engine in the fixture since main() will handle it
2. Adding a flag to main() to skip engine disposal when running in test mode
3. Catching and ignoring any disposal errors in the fixture cleanup

Suggested change

	await engine.dispose()
	try:
	await engine.dispose()
	except Exception:
	# Ignore errors if the engine was already disposed (e.g., by main()).
	pass

[Copilot]

The tests rely on timing with asyncio.sleep(0.2) to allow the output loop to process entries. This makes tests non-deterministic and potentially flaky, especially on slower systems or under load. The tests could fail if the system is too slow to process within 200ms, or could waste time if processing completes faster.

Consider using a more reliable synchronization mechanism, such as:

1. Adding a callback or event mechanism to main() for testing
2. Using a mock/spy on the print function to detect when output occurs
3. Polling the output buffer until expected content appears (with a timeout)

[Copilot]

Missing pytest-asyncio configuration in pyproject.toml. pytest-asyncio requires explicit configuration to handle async fixtures and tests properly. Without configuration, you may encounter warnings or unexpected behavior.

Add a [tool.pytest.ini_options] section with asyncio_mode configuration, for example:

[tool.pytest.ini_options]
asyncio_mode = "auto"

or

asyncio_default_fixture_loop_scope = "function"

This ensures pytest-asyncio operates in the expected mode.

[Copilot]

'except' clause does nothing but pass and there is no explanatory comment.

[Copilot]

'except' clause does nothing but pass and there is no explanatory comment.

[Copilot]

'except' clause does nothing but pass and there is no explanatory comment.