/generate

/qt-suite:generate — AI Test Generation

Generate unit tests for this Qt project. When an argument is provided, generate tests for that specific file or class. When no argument is provided, scan the project to identify the files most in need of tests.

Step 1: Identify Targets

If an argument was provided (e.g., /qt-suite:generate src/calculator.py):

• Use the argument as the target file or class name
• Find the file with Glob if only a filename was given

If no argument was provided:

1. Read .qt-test.json if present to determine project_type and test_dir
2. Detect project type by checking for CMakeLists.txt (C++) or pyproject.toml/setup.cfg (Python)
3. Use Glob to find all source files:
   • Python: **/*.py (excluding tests/, __init__.py, migration files)
   • C++: **/*.cpp and **/*.h (excluding tests/, moc_*, ui_*, build/)
4. For each source file, check if a corresponding test file exists:
   • Python: tests/test_<module>.py for src/<module>.py
   • C++: tests/<name>_test.cpp or tests/test_<name>.cpp for src/<name>.cpp
5. Prioritize files with no test file at all, then files with low complexity coverage (look for classes with multiple public methods but few or no tests)
6. Present the top 3 candidates and select the best target. Prefer business logic classes over utility/helper files.

Step 2: Read and Analyze the Source File

Read the target source file completely. Identify:

• Class names and their public interface
• Methods and their signatures, parameters, return types
• Side effects: file I/O, signals emitted, state mutations
• Edge cases visible from the code: null checks, boundary conditions, exception paths
• Qt-specific elements: signals, slots, widget interactions, model overrides

Step 3: Check Existing Tests

If a test file already exists for this source:

• Read it to understand existing coverage
• Focus generated tests on uncovered methods and edge cases
• Do not duplicate tests that already exist

Step 4: Determine Test File Location and Framework

Python projects:

• Test file: tests/test_<module_name>.py
• Framework: pytest + pytest-qt (for GUI classes), plain pytest (for non-GUI)
• Detect if class inherits from QWidget/QDialog/etc. → use qtbot fixture

C++ projects:

• Test file: tests/<ClassName>Test.cpp (or tests/test_<name>.cpp)
• Framework: QTest
• Each test class is a QObject subclass with Q_OBJECT macro
• End file with QTEST_MAIN(TestClassName) and #include "<file>.moc"

QML:

• Test file: tests/tst_<ComponentName>.qml
• Framework: QtQuickTest TestCase item

Step 5: Generate Tests

Write comprehensive tests targeting:

1. Happy path — each public method with valid inputs
2. Edge cases — empty input, null/None, boundary values, max values
3. Error paths — invalid input, exceptions thrown, signals emitted on error
4. State transitions — object state before and after operations
5. Signal behavior — signals emitted with correct arguments (use QSignalSpy in C++, waitSignal/waitSignals in pytest-qt)

Reference the qtest-patterns skill for correct syntax and patterns.

For pytest-qt tests, always use qtbot.addWidget() for any widget created in a test. For QTest C++ tests, use init() / cleanup() slots for per-test setup/teardown.

Step 6: Write the Test File

Write the test file to the determined location. If the file already exists, append new test classes/functions — never overwrite existing tests.

Step 7: Report

Report concisely:

• File generated: tests/test_calculator.py
• Tests written: N test functions / N test cases
• Coverage targets: list the methods/paths covered
• To run: pytest tests/test_calculator.py -v (Python) or cmake --build build && ctest -R TestCalculator (C++)
• Note any methods that couldn't be tested without further context (e.g., methods that require a running database)