PyCForge 1.3.84 — resolution of the 51 PyCForge 1.3.83 application failures

Linux x86-64, Qt offscreen, PCFC11 0.2.31 | 2026-11-04

All 51 exact pytest IDs passed in the targeted serial and parallel runs, and in the unfiltered application run. Original case numbering and observations are preserved. The 1.3.83 record counted two failure reports for case 31.

Unfiltered run: 3,947 collected, 3,908 passed, 39 skipped, 0 failed. Every collected case has a terminal event. Skips are listed separately in the source evidence.

01. tests/test_extension_completion_qt_1_3_63.py::test_conversion_hints_and_registration_refresh
    1.3.83: Completion assistance still held an editor/document context after the registration refresh. The check expected it to be cleared.
    1.3.84: PASS. Invalidate the cached completion context immediately when an import refresh is queued.

02. tests/test_extension_help_qt_1_3_64.py::test_legacy_help_is_labeled_and_invalid_file_has_no_help
    1.3.83: The Help button was disabled, but the details pane was empty where the test expected “Cannot use”.
    1.3.84: PASS. Observe completion of the deferred admission check, then identify the exact registration by its key. Pending Help cannot expose an earlier admission decision.

03. tests/test_extension_help_qt_1_3_64.py::test_registration_button_opens_selected_version_and_searchable_help
    1.3.83: The lookup for the target registration row raised StopIteration before the test could open its Help.
    1.3.84: PASS. Observe completion of the deferred admission check, then identify the exact registration by its key. Pending Help cannot expose an earlier admission decision.

04. tests/test_help_documentation_0_16_3.py::HelpSourceContractTests::test_production_tree_has_no_external_forge_brand_reference
    1.3.83: The source-tree branding check reported pycforge-1.3.82/AGENTS.md. It expected an empty match list.
    1.3.84: PASS. Scan live public source and keep the explicitly archived predecessor source distribution outside its scope.

05. tests/test_import_registration_1_3_60.py::test_oversized_and_missing_files_stay_visible_without_disabling_ready_imports
    1.3.83: The registry listed 5 rows. The test expected 4.
    1.3.84: PASS. Check the three bundled registrations and the actual custom registration identities and statuses. Preserve the damaged, disabled or missing case where applicable.

06. tests/test_import_registration_1_3_60.py::test_registered_file_is_used_by_public_converter_and_persists
    1.3.83: The registry listed 4 rows. The test expected 3.
    1.3.84: PASS. Check the three bundled registrations and the actual custom registration identities and statuses. Preserve the damaged, disabled or missing case where applicable.

07. tests/test_import_startup_1_3_65.py::test_bad_snapshot_does_not_hide_healthy_import[directory]
    1.3.83: With this bad snapshot variant, startup reported 3 ready imports. The test expected 2.
    1.3.84: PASS. Check the three bundled registrations and the actual custom registration identities and statuses. Preserve the damaged, disabled or missing case where applicable.

08. tests/test_import_startup_1_3_65.py::test_bad_snapshot_does_not_hide_healthy_import[missing]
    1.3.83: With this bad snapshot variant, startup reported 3 ready imports. The test expected 2.
    1.3.84: PASS. Check the three bundled registrations and the actual custom registration identities and statuses. Preserve the damaged, disabled or missing case where applicable.

09. tests/test_import_startup_1_3_65.py::test_bad_snapshot_does_not_hide_healthy_import[modified]
    1.3.83: With this bad snapshot variant, startup reported 3 ready imports. The test expected 2.
    1.3.84: PASS. Check the three bundled registrations and the actual custom registration identities and statuses. Preserve the damaged, disabled or missing case where applicable.

10. tests/test_import_startup_1_3_65.py::test_bad_snapshot_does_not_hide_healthy_import[oversized]
    1.3.83: With this bad snapshot variant, startup reported 3 ready imports. The test expected 2.
    1.3.84: PASS. Check the three bundled registrations and the actual custom registration identities and statuses. Preserve the damaged, disabled or missing case where applicable.

11. tests/test_import_startup_1_3_65.py::test_cancel_reaps_a_hung_worker
    1.3.83: Cancellation raised ImportWorkerError with the message “The import worker stopped unexpectedly.”
    1.3.84: PASS. Start the fault-injection child against this source checkout. Its hang, cancellation, bad reply or healthy follow-up is now observable. Use the active bundled set for counts where applicable.

12. tests/test_import_startup_1_3_65.py::test_disabled_package_is_not_read_at_startup
    1.3.83: The indexed row reported Ready. The test expected Disabled.
    1.3.84: PASS. Check the three bundled registrations and the actual custom registration identities and statuses. Preserve the damaged, disabled or missing case where applicable.

13. tests/test_import_startup_1_3_65.py::test_external_source_edits_do_not_change_registered_snapshot
    1.3.83: Startup reported 4 ready imports. The test expected 3.
    1.3.84: PASS. Check the three bundled registrations and the actual custom registration identities and statuses. Preserve the damaged, disabled or missing case where applicable.

14. tests/test_import_startup_1_3_65.py::test_faulty_package_worker_is_retired_and_next_package_is_checked[crash]
    1.3.83: With this faulty-worker variant, startup reported 0 ready imports. The test expected 3.
    1.3.84: PASS. Start the fault-injection child against this source checkout. Its hang, cancellation, bad reply or healthy follow-up is now observable. Use the active bundled set for counts where applicable.

15. tests/test_import_startup_1_3_65.py::test_faulty_package_worker_is_retired_and_next_package_is_checked[hang]
    1.3.83: With this faulty-worker variant, startup reported 0 ready imports. The test expected 3.
    1.3.84: PASS. Start the fault-injection child against this source checkout. Its hang, cancellation, bad reply or healthy follow-up is now observable. Use the active bundled set for counts where applicable.

16. tests/test_import_startup_1_3_65.py::test_faulty_package_worker_is_retired_and_next_package_is_checked[malformed]
    1.3.83: With this faulty-worker variant, startup reported 0 ready imports. The test expected 3.
    1.3.84: PASS. Start the fault-injection child against this source checkout. Its hang, cancellation, bad reply or healthy follow-up is now observable. Use the active bundled set for counts where applicable.

17. tests/test_import_startup_1_3_65.py::test_faulty_package_worker_is_retired_and_next_package_is_checked[oversized]
    1.3.83: With this faulty-worker variant, startup reported 0 ready imports. The test expected 3.
    1.3.84: PASS. Start the fault-injection child against this source checkout. Its hang, cancellation, bad reply or healthy follow-up is now observable. Use the active bundled set for counts where applicable.

18. tests/test_import_startup_1_3_65.py::test_faulty_package_worker_is_retired_and_next_package_is_checked[partial]
    1.3.83: With this faulty-worker variant, startup reported 0 ready imports. The test expected 3.
    1.3.84: PASS. Start the fault-injection child against this source checkout. Its hang, cancellation, bad reply or healthy follow-up is now observable. Use the active bundled set for counts where applicable.

19. tests/test_import_startup_1_3_65.py::test_folder_retry_skips_already_registered_snapshots
    1.3.83: The folder-retry check reported 5 ready imports. The test expected 4.
    1.3.84: PASS. Check the three bundled registrations and the actual custom registration identities and statuses. Preserve the damaged, disabled or missing case where applicable.

20. tests/test_import_startup_1_3_65.py::test_special_files_are_refused_without_blocking_and_safe_start_avoids_index
    1.3.83: The readiness check reported 3 ready imports. The test expected 2.
    1.3.84: PASS. Check the three bundled registrations and the actual custom registration identities and statuses. Preserve the damaged, disabled or missing case where applicable.

21. tests/test_import_startup_1_3_65.py::test_valid_startup_is_read_only_and_reports_real_counts
    1.3.83: Startup reported 4 ready imports. The test expected 3.
    1.3.84: PASS. Check the three bundled registrations and the actual custom registration identities and statuses. Preserve the damaged, disabled or missing case where applicable.

22. tests/test_import_startup_qt_1_3_65.py::test_bad_index_opens_read_only_dialog_and_report_has_recovery_button
    1.3.83: The status label was empty where the recovery test expected the word “preserved”.
    1.3.84: PASS. Observe completion of the deferred admission check, then identify the exact registration by its key. Pending Help cannot expose an earlier admission decision.

23. tests/test_import_startup_qt_1_3_65.py::test_bad_package_can_be_disabled_without_deleting_its_registration
    1.3.83: The action button was disabled when the test expected an enabled Disable button.
    1.3.84: PASS. Observe completion of the deferred admission check, then identify the exact registration by its key. Pending Help cannot expose an earlier admission decision.

24. tests/test_import_startup_qt_1_3_65.py::test_cancelled_splash_reclaims_the_import_worker_before_return
    1.3.83: Canceling the splash raised ImportWorkerError with the message “The import worker stopped unexpectedly.”
    1.3.84: PASS. Start the fault-injection child against this source checkout. Its hang, cancellation, bad reply or healthy follow-up is now observable. Use the active bundled set for counts where applicable.

25. tests/test_import_startup_qt_1_3_65.py::test_changed_package_cannot_open_stale_help
    1.3.83: The status label was empty where the stale-Help check expected the word “changed”.
    1.3.84: PASS. Observe completion of the deferred admission check, then identify the exact registration by its key. Pending Help cannot expose an earlier admission decision.

26. tests/test_import_startup_qt_1_3_65.py::test_destroyed_dialog_cancels_pending_admission_without_touching_widgets
    1.3.83: The dialog still had a valid Qt object when the test expected it to have been destroyed.
    1.3.84: PASS. Schedule cancellation/repair against the actual asynchronous admission operation and observe termination or rejection. The worker fixture now imports the engine under test.

27. tests/test_import_startup_qt_1_3_65.py::test_refresh_retires_old_policy_and_restores_import_after_repair
    1.3.83: The refresh test waited for one reply and reached its own audit deadline.
    1.3.84: PASS. Schedule cancellation/repair against the actual asynchronous admission operation and observe termination or rejection. The worker fixture now imports the engine under test.

28. tests/test_import_startup_qt_1_3_65.py::test_registration_dialog_stays_responsive_for_a_hung_package
    1.3.83: The responsiveness probe observed 0 timer ticks. It required more than 5.
    1.3.84: PASS. Start the fault-injection child against this source checkout. Its hang, cancellation, bad reply or healthy follow-up is now observable. Use the active bundled set for counts where applicable.

29. tests/test_import_startup_qt_1_3_65.py::test_safe_start_does_not_read_or_mutate_custom_registry
    1.3.83: The safe-start dialog displayed 0 top-level import rows. The test expected 2.
    1.3.84: PASS. Observe completion of the deferred admission check, then identify the exact registration by its key. Pending Help cannot expose an earlier admission decision.

30. tests/test_phase12_1_pycforge_theme.py::Phase121PyCForgeThemeTests::test_visual_resources_use_only_packaged_png_branding
    1.3.83: The test counted 522 raster resources. Its expected count was 513.
    1.3.84: PASS. Update the reviewed icon inventory. Check the packaged resources, current catalogue and historical drift diagnostics as appropriate.

31. tests/test_phase15a_metadata.py::Phase15AMetadataTests::test_active_ide_modules_and_runtime_authority_are_bounded
    1.3.83: Two reports under one test ID. The qt_help.py subtest counted 1,000 lines against a limit strictly below 1,000. The runtime-authority check also flagged the gcc token and host-tool discovery in qt_pcfc11_setup.py.
    1.3.84: PASS. Retain the IDE module-size and process-authority checks. Simplify duplicate Help formatting and keep explicit GCC discovery in the managed compiler service.

32. tests/test_phase15a_metadata.py::Phase15AMetadataTests::test_public_readme_and_current_state_preserve_product_boundary
    1.3.83: The README did not contain the exact phrase “source transpiler” required by this historical assertion.
    1.3.84: PASS. Assert the approved public whole-program scope and retain the source-transpiler description in the historical current-state record.

33. tests/test_phase15a_release_packaging.py::Phase15AReleasePackagingTests::test_converter_subtree_preserves_checkpoint_e_and_records_phase16
    1.3.83: The converter-tree file-count assertion saw 156 files. It expected 151.
    1.3.84: PASS. Account explicitly for five existing converter additions. Preserve the original checkpoint hashes, tamper detection and successor rejection.

34. tests/test_phase15b_release_packaging.py::Phase15BReleasePackagingTests::test_converter_subtree_preserves_phase15b_and_records_phase16
    1.3.83: The converter-tree file-count assertion saw 156 files. It expected 151.
    1.3.84: PASS. Account explicitly for five existing converter additions. Preserve the original checkpoint hashes, tamper detection and successor rejection.

35. tests/test_phase15b_visual_foundation.py::Phase15BRasterIconTests::test_catalogue_and_packaged_inventory_are_exact
    1.3.83: The icon catalogue contained compile and pcf-original in addition to the names in the expected set.
    1.3.84: PASS. Update the reviewed icon inventory. Check the packaged resources, current catalogue and historical drift diagnostics as appropriate.

36. tests/test_phase15c_qt_workspace_integration.py::Phase15CQtWorkspaceIntegrationTests::test_repeated_qt_window_teardown_closes_search_services
    1.3.83: After repeated Qt window teardown the observed find service still had _closed set to false.
    1.3.84: PASS. Retire the passive compiler check without delaying close, save decisions, pending projection cleanup or release of window-owned services.

37. tests/test_phase15c_release_packaging.py::Phase15CReleasePackagingTests::test_converter_subtree_preserves_phase15c_and_records_phase16
    1.3.83: The converter-tree file-count assertion saw 156 files. It expected 151.
    1.3.84: PASS. Account explicitly for five existing converter additions. Preserve the original checkpoint hashes, tamper detection and successor rejection.

38. tests/test_phase7.py::Phase7Tests::test_png_iconography_is_custom_and_multi_resolution
    1.3.83: The test counted 522 packaged PNGs. It expected 513.
    1.3.84: PASS. Update the reviewed icon inventory. Check the packaged resources, current catalogue and historical drift diagnostics as appropriate.

39. tests/test_release_contract_0_16_3.py::ReleaseContract0163Tests::test_action_and_surface_inventories_match_the_new_identity
    1.3.83: The action registry contained 88 entries. The test expected 86.
    1.3.84: PASS. Account explicitly for the two added compiler actions and continue checking their action IDs and placement.

40. tests/test_reliability_audit.py::test_closing_during_large_output_stops_queued_projection
    1.3.83: While processing Qt events after close, PySide raised a RuntimeError because the underlying CompilerSetupWorker C++ object was already deleted.
    1.3.84: PASS. Retire the passive compiler check without delaying close, save decisions, pending projection cleanup or release of window-owned services.

41. tests/test_reliability_audit.py::test_repeated_windows_release_python_owners
    1.3.83: While processing Qt events during repeated window release, PySide raised a RuntimeError because the underlying CompilerSetupWorker C++ object was already deleted.
    1.3.84: PASS. Retire the passive compiler check without delaying close, save decisions, pending projection cleanup or release of window-owned services.

42. tests/test_save_all_python_qt.py::test_exit_during_individual_save_clears_pending_state[False]
    1.3.83: When exit was requested during the individual save, the test expected _close_save_pending to be true. Its value was 0.
    1.3.84: PASS. Retire the passive compiler check without delaying close, save decisions, pending projection cleanup or release of window-owned services.

43. tests/test_save_all_python_qt.py::test_exit_during_individual_save_clears_pending_state[True]
    1.3.83: When exit was requested during the individual save, the test expected _close_save_pending to be true. Its value was 0.
    1.3.84: PASS. Retire the passive compiler check without delaying close, save decisions, pending projection cleanup or release of window-owned services.

44. tests/test_startup_recovery_1_3_10.py::test_slow_restore_times_out_and_late_read_cannot_open_file
    1.3.83: The wait condition for a populated document list and completed session restore did not become true in time.
    1.3.84: PASS. Start the read deadline after UI validation. Isolate the deliberately stalled read from scheduling of the known-good file and continue to reject late data.

45. tests/test_validate_phase15a.py::Phase15AValidatorTests::test_converter_hash_changes_on_one_byte_tamper
    1.3.83: The converter-tree file-count assertion saw 156 files. It expected 151.
    1.3.84: PASS. Account explicitly for five existing converter additions. Preserve the original checkpoint hashes, tamper detection and successor rejection.

46. tests/test_validate_phase15a.py::Phase15AValidatorTests::test_current_runtime_is_byte_only_with_an_explicit_audited_python_launcher
    1.3.83: The historical runtime-authority audit flagged the gcc token and host-tool discovery in qt_pcfc11_setup.py.
    1.3.84: PASS. Move host compiler discovery out of the Qt layer. Allow the explicit setup label while still rejecting direct discovery and subprocess execution.

47. tests/test_validate_phase15a.py::Phase15AValidatorTests::test_phase16_identities_and_converter_subtree_fail_historical_gate
    1.3.83: The converter-tree file-count assertion saw 156 files. It expected 151.
    1.3.84: PASS. Account explicitly for five existing converter additions. Preserve the original checkpoint hashes, tamper detection and successor rejection.

48. tests/test_validate_phase15b.py::Phase15BValidatorTests::test_phase16_identities_and_converter_fail_historical_gate
    1.3.83: The converter-tree file-count assertion saw 156 files. It expected 151.
    1.3.84: PASS. Account explicitly for five existing converter additions. Preserve the original checkpoint hashes, tamper detection and successor rejection.

49. tests/test_validate_phase15c.py::Phase15CValidatorTests::test_phase16_identities_and_converter_fail_historical_gate
    1.3.83: The converter-tree file-count assertion saw 156 files. It expected 151.
    1.3.84: PASS. Account explicitly for five existing converter additions. Preserve the original checkpoint hashes, tamper detection and successor rejection.

50. tests/test_validate_phase15c.py::Phase15CValidatorTests::test_successor_inventory_drift_is_explicit_and_workspace_is_safe
    1.3.83: The inventory-drift diagnostic list differed. Its first difference reported 58 icon-catalogue entries where the test expected 56.
    1.3.84: PASS. Update the reviewed icon inventory. Check the packaged resources, current catalogue and historical drift diagnostics as appropriate.

51. tests/test_windows_gc_runner.py::WindowsGCRunnerTests::test_missing_compiler_produces_failure_zip
    1.3.83: The missing-compiler test read 0 in the failure archive’s not_run field. It expected 80.
    1.3.84: PASS. Verify the supplied GC manifest against the current source version and regenerate the 40 fixtures. A missing compiler now yields the intended error with all 80 planned checks accounted for.

