Fix #233: Loop mode treats all files as unit test files

[Serhan-Asad]

Fixes the issue #233 where pdd fix --loop incorrectly treats the last file as ERROR_FILE instead of a unit test file.

Change

• Fix for loop mode file parsing in pdd/commands/fix.py
• Validation requiring --verification-program for loop mode
• Updated documentation in README.md and example files
• Unit tests verifying the fix

Test Files

• tests/test_commands_fix.py - Added 2 new tests

Tests

• test_fix_loop_mode_treats_all_files_as_unit_tests - PASSES (verifies all files treated as unit tests)
• test_fix_requires_verification_program_in_loop_mode - PASSES (verifies validation)
• All existing tests - PASS

Cause

The bug was in pdd/commands/fix.py. Click's argument parsing with nargs=-1 followed by a required positional argument causes the last file to be assigned to error_file, even in loop mode where all files should be unit test files.

The fix recombines error_file back into unit_test_files when in loop mode, then sets error_file = "" (auto-generated by verification program)

Also added that for --loop verification-program is needed

[Copilot]

Pull request overview

This PR fixes issue #233 where pdd fix --loop incorrectly treated the last file as an error file instead of a unit test file. The fix addresses Click's argument parsing behavior with nargs=-1 by recombining the error_file back into unit_test_files when in loop mode. Additionally, it enforces that --verification-program is required for loop mode.

Key changes:

• Fixed loop mode file parsing logic in pdd/commands/fix.py
• Added validation requiring --verification-program for loop mode
• Updated documentation to clarify loop vs non-loop mode behavior

Reviewed changes

Copilot reviewed 4 out of 4 changed files in this pull request and generated 3 comments.

File	Description
pdd/commands/fix.py	Added argument validation and recombination logic for loop mode; enforces verification-program requirement
tests/test_commands_fix.py	Added two new tests verifying loop mode behavior and validation
context/commands/fix_example.py	Updated documentation to clarify loop mode requirements and remove error_file from loop example
README.md	Enhanced documentation with separate examples for loop and non-loop modes

---

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

[Copilot]

The docstring incorrectly states "ERROR_FILE is treated as an additional unit test file" in loop mode. According to the code logic (lines 108-111), ERROR_FILE is recombined into unit_test_files and then set to empty string - it's not "treated as" a unit test file, it was incorrectly parsed as ERROR_FILE and is being corrected. Consider rephrasing to: "Loop mode: All arguments are unit test files. ERROR_FILE is recombined into UNIT_TEST_FILES (correcting Click's parsing behavior)."

Suggested change

	Loop mode: ERROR_FILE is treated as an additional unit test file. All files
	are test files. Errors are generated by running the verification program.
	Non-loop mode: UNIT_TEST_FILES are test files, ERROR_FILE contains static errors.
	Loop mode: All arguments are unit test files. ERROR_FILE is recombined into
	UNIT_TEST_FILES (correcting Click's parsing behavior), and errors are
	generated by running the verification program.
	Non-loop mode: UNIT_TEST_FILES are test files, and ERROR_FILE contains
	static errors.

[Copilot]

The syntax [ERROR_FILE] suggests this argument is optional in all cases, but according to lines 1470-1471, ERROR_FILE is required in non-loop mode. The syntax should indicate this requirement more clearly, perhaps with a note or different formatting to distinguish between loop and non-loop mode requirements.

[Copilot]

While this explanation is helpful, it could be confusing because the command line shown on line 1500 doesn't make it obvious which file is ERROR_FILE versus a unit test file. Consider explicitly labeling the arguments in the command example or adding a comment showing which file serves which purpose.

Suggested change

	pdd [GLOBAL OPTIONS] fix --output-code src/factorial_fixed.py --output-results results/fix_results.log \
	factorial_python.prompt src/factorial.py tests/test_factorial.py tests/test_edge_cases.py errors.log
	pdd [GLOBAL OPTIONS] fix \
	--output-code src/factorial_fixed.py \ # OUTPUT_CODE (fixed code file)
	--output-results results/fix_results.log \ # OUTPUT_RESULTS (LLM + test results log)
	factorial_python.prompt \ # PROMPT_FILE
	src/factorial.py \ # CODE_FILE
	tests/test_factorial.py tests/test_edge_cases.py \# TEST_FILE(S)
	errors.log # ERROR_FILE (static error messages)