Unittest Operation

Unittest executes Python unit-test functions through the pytest adapter. It records function-level results and generates reports without Local LLM analysis.

Framework structure and result contracts are documented in Unittest Framework.

Operation Summary

Item	Unittest operation
Test unit	Test function or `unittest.TestCase` method
Primary identifier	Execution ID
TEST ID and Fixture ID	Not used
Mock/HIL mode	Not used
Result capture	`test_envs/tests/unittest/conftest.py`
Local LLM	Not used
Coverage	`pytest-cov` or VS Code Testing Coverage

Execution Sequence

Collect unittest tests through pytest
        ↓
Execute setup, call, and teardown phases
        ↓
Capture function status, duration, and failure detail
        ↓
Aggregate every function at session finish
        ↓
Write Result JSON and Result Log

Step	Operation
1	The pytest adapter collects tests under `test_envs/tests/unittest`
2	The result hook observes setup, call, and teardown reports
3	Each function becomes `PASS`, `FAIL`, `ERROR`, or `SKIP`
4	`pytest_sessionfinish` aggregates function results into one execution
5	The framework stores one Result JSON and Result Log per Execution ID

Setup or teardown failures become ERROR. Assertion failures during the call phase become FAIL.

Run Unittest

Run the complete Unittest area:

.\.venv\Scripts\python.exe -m pytest -p no:cacheprovider test_envs/tests/unittest

Run one test file:

.\.venv\Scripts\python.exe -m pytest -p no:cacheprovider test_envs/tests/unittest/ct_framework/python/test_configuration.py

VS Code Testing also discovers these tests through the pytest adapter. Native VS Code unittest discovery remains disabled to prevent duplicate nodes.

Test Coverage

Run Unittest and print missing Python lines:

.\.venv\Scripts\python.exe -m pytest -p no:cacheprovider test_envs/tests/unittest --cov=test_envs --cov-report=term-missing

pytest-cov stores coverage data in .coverage. Add --cov-report=html to generate an HTML report under htmlcov/.

In VS Code Testing, select the Unittest tree or a test node and use Run Test with Coverage.

Result and Report

Unittest execution and Report generation are separate operations. The shared Report pipeline detects category: unit and skips Local LLM analysis and escalation.

Unittest execution
        ↓
Result JSON + Result Log
        ↓
test_envs.tools.test_result
        ↓
Local LLM and escalation: skipped
        ↓
Canonical Markdown
        ├── MkDocs Results
        └── Pandoc Report

Output	Path
Result JSON	`test_reports/results/unittest/<execution-id>_result.json`
Result log	`test_reports/results/unittest/<execution-id>_result.log`
Local LLM log	Not generated
Canonical Markdown	`test_reports/markdown/unittest/<execution-id>_result.md`
MkDocs result	`docs/tests/unittest/`
Pandoc output	`test_reports/pandocs/unittest/`
DOCX reference	`test_reports/pandocs/reference.docx`

Generate every pending Markdown report and publish MkDocs result pages:

.\.venv\Scripts\python.exe -m test_envs.tools.test_result --pending --docs

The command is shared with Pytest, but Unittest produces its Markdown directly from the aggregated function results without contacting Ollama.

See Unittest Results for published results.