All automated tests persist their generated plots under test/output/<suite>/.
If images, PDFs, or text dumps appear in the repository root, remove them and
rerun the tests so the artifacts land in their dedicated subdirectories. The
make clean target also prunes stray files outside test/output/.
# Run comprehensive validation framework tests
make test ARGS="--target test_mpeg_comprehensive_validation_framework"
# Run false positive detection tests
make test ARGS="--target test_mpeg_false_positive_detection_comprehensive"
# Run all MPEG validation tests
find test/ -name "*mpeg*.f90" | wc -l # Shows 21 test files
make test | grep mpeg
21 Test Files covering: - Comprehensive validation framework - False positive detection (Issue #32) - Size validation with frame/resolution correlation - Header validation for MP4 format signatures - FFprobe external tool integration - Edge cases and stress testing - Performance validation - Media player compatibility
Successful Output:
=== COMPREHENSIVE VALIDATION FRAMEWORK TESTS ===
Layer 1 - Basic validation: T
Layer 2 - Size validation: T
Layer 3 - Header validation: T
Layer 4 - Semantic validation: T
Layer 5 - External tool validation: T
Overall framework validation: T
Failure Analysis: - Layer 1 Failure: File missing or corrupted - Layer 2 Failure: File size inadequate for content - Layer 3 Failure: Invalid MP4 format headers - Layer 5 Failure: External tools detect format errors
Control warning output during testing and development:
# Suppress all warnings (useful for CI and automated testing)
export FORTPLOT_SUPPRESS_WARNINGS=1
make test
# Force warnings even in CI environments
export FORTPLOT_FORCE_WARNINGS=1
make test
# Default behavior (warnings visible during development)
unset FORTPLOT_SUPPRESS_WARNINGS
unset FORTPLOT_FORCE_WARNINGS
make test
Supported Values:
- FORTPLOT_SUPPRESS_WARNINGS: 1, true, yes, on (case-insensitive)
- FORTPLOT_FORCE_WARNINGS: 1, true, yes, on (case-insensitive)
CI Auto-Detection: Warnings are automatically suppressed on common CI platforms (GitHub Actions, Jenkins, Travis CI, CircleCI).
# Enable Windows CI performance optimizations
export FORTPLOT_USE_MEMORY_BACKEND=1
export FORTPLOT_MINIMIZE_IO=1
make test
# Run with performance monitoring enabled
export FORTPLOT_DEBUG=1
make test ARGS="--target test_pcolormesh_consolidated"
# Check for tests exceeding 30-second target
make test | grep "REGRESSION\|Performance target"
# Previously slow tests now optimized for CI
make test ARGS="--target test_pcolormesh_consolidated" # Was >2min, now <30sec
make test ARGS="--target test_histogram_consolidated" # Was >1.5min, now <30sec
make test ARGS="--target test_contour_filled_backend_rendering" # Was >3min, now <30sec
See Windows CI Performance Guide for full optimization details.
Run these guards after introducing or modifying plotting helpers so that OO and
stateful APIs stay aligned with pyplot-fortran positional expectations.
make test ARGS="--target test_pyplot_legacy_order"
make test ARGS="--target test_legacy_positional_order"
Passing criteria
- Legacy positional calls compile without keyword reordering.
- Global pyplot façade and figure_t helpers report matching plot counts.
make test
# Run tests without warning noise
FORTPLOT_SUPPRESS_WARNINGS=1 make test
# Figure and basic plotting
make test ARGS="--target test_figure_basics"
# Animation functionality
make test ARGS="--target test_animation_save"
# Unicode support
make test ARGS="--target test_unicode_detection"
# Warning suppression system
make test ARGS="--target test_warning_suppression_control"
make test ARGS="--target test_marker_example_guard"
Run this guard to validate marker rendering artifacts across PNG/PDF/ASCII
outputs. It is the canonical CI regression check for marker output and
supersedes the retired test_markers_rendering smoke test.