- Mark TCOD Integration Sprint as complete
- Document FOV system with perspective rendering implementation
- Update UIEntity pathfinding status to complete with A* and caching
- Add comprehensive achievement entry for July 10 work
- Reflect current engine capabilities accurately
The engine now has all core roguelike mechanics:
- Field of View with per-entity visibility
- Pathfinding (both Dijkstra and A*)
- Text input for in-game consoles
- Performance optimizations throughout
🤖 Generated with [Claude Code](https://claude.ai/code)
Co-Authored-By: Claude <noreply@anthropic.com>
Major Engine Enhancements:
- Complete FOV (Field of View) system with perspective rendering
- UIGrid.perspective property for entity-based visibility
- Three-layer overlay colors (unexplored, explored, visible)
- Per-entity visibility state tracking
- Perfect knowledge updates only for explored areas
- Advanced Pathfinding Integration
- A* pathfinding implementation in UIGrid
- Entity.path_to() method for direct pathfinding
- Dijkstra maps for multi-target pathfinding
- Path caching for performance optimization
- GUI Text Input Widgets
- TextInputWidget class with cursor, selection, scrolling
- Improved widget with proper text rendering and input handling
- Example showcase of multiple text input fields
- Foundation for in-game console and chat systems
- Performance & Architecture Improvements
- PyTexture copy operations optimized
- GameEngine update cycle refined
- UIEntity property handling enhanced
- UITestScene modernized
Test Suite:
- Interactive visibility demos showing FOV in action
- Pathfinding comparison (A* vs Dijkstra)
- Debug utilities for visibility and empty path handling
- Sizzle reel demo combining pathfinding and vision
- Multiple text input test scenarios
This commit brings McRogueFace closer to a complete roguelike engine
with essential features like line-of-sight, intelligent pathfinding,
and interactive text input capabilities.
🤖 Generated with [Claude Code](https://claude.ai/code)
Co-Authored-By: Claude <noreply@anthropic.com>
- dijkstra_interactive_enhanced.py: Animation along paths with smooth movement
- M key to start movement animation
- P to pause/resume
- R to reset positions
- Visual path gradient for better clarity
- pathfinding_showcase.py: Advanced multi-entity behaviors
- Chase mode: enemies pursue player
- Flee mode: enemies avoid player
- Patrol mode: entities follow waypoints
- WASD player movement
- Dijkstra distance field visualization (D key)
- Larger dungeon map with multiple rooms
- Both demos use new entity.path_to() method
- Smooth interpolated movement animations
- Real-time pathfinding recalculation
- Comprehensive test coverage
These demos showcase the power of integrated pathfinding for game AI.
🤖 Generated with [Claude Code](https://claude.ai/code)
Co-Authored-By: Claude <noreply@anthropic.com>
- Add path_to(target_x, target_y) method to UIEntity class
- Uses existing Dijkstra pathfinding implementation from UIGrid
- Returns list of (x, y) coordinate tuples for complete path
- Supports both positional and keyword argument formats
- Proper error handling for out-of-bounds and no-grid scenarios
- Comprehensive test suite covering normal and edge cases
Part of TCOD integration sprint - gives entities immediate pathfinding capabilities.
🤖 Generated with [Claude Code](https://claude.ai/code)
Co-Authored-By: Claude <noreply@anthropic.com>
- Mark UIGrid TCOD Integration as completed
- Document critical PyArg bug fix achievement
- Update UIEntity Pathfinding to 50% complete
- Add detailed progress notes for July 9 sprint work
🤖 Generated with [Claude Code](https://claude.ai/code)
Co-Authored-By: Claude <noreply@anthropic.com>
- Add complete Dijkstra pathfinding to UIGrid class
- compute_dijkstra(), get_dijkstra_distance(), get_dijkstra_path()
- Full TCODMap and TCODDijkstra integration
- Proper memory management in constructors/destructors
- Create mcrfpy.libtcod submodule with Python bindings
- dijkstra_compute(), dijkstra_get_distance(), dijkstra_get_path()
- line() function for drawing corridors
- Foundation for future FOV and pathfinding algorithms
- Fix critical PyArg bug in UIGridPoint color setter
- PyObject_to_sfColor() now handles both Color objects and tuples
- Prevents "SystemError: new style getargs format but argument is not a tuple"
- Proper error handling and exception propagation
- Add comprehensive test suite
- test_dijkstra_simple.py validates all pathfinding operations
- dijkstra_test.py for headless testing with screenshots
- dijkstra_interactive.py for full user interaction demos
- Consolidate and clean up test files
- Removed 6 duplicate/broken demo attempts
- Two clean versions: headless test + interactive demo
Part of TCOD integration sprint for RoguelikeDev Tutorial Event.
🤖 Generated with [Claude Code](https://claude.ai/code)
Co-Authored-By: Claude <noreply@anthropic.com>
- Created comprehensive markdown documentation matching HTML completeness
- Documented all 75 functions, 20 classes, 56 methods, and 20 automation methods
- Zero ellipsis instances - complete coverage with no missing documentation
- Added proper markdown formatting with code blocks and navigation
- Included full parameter documentation, return values, and examples
Key features:
- 23KB GitHub-compatible markdown documentation
- 47 argument sections with detailed parameters
- 35 return value specifications
- 23 code examples with syntax highlighting
- 38 explanatory notes and 10 exception specifications
- Full table of contents with anchor links
- Professional markdown formatting
Both export formats now available:
- HTML: docs/api_reference_complete.html (54KB, rich styling)
- Markdown: docs/API_REFERENCE_COMPLETE.md (23KB, GitHub-compatible)
Test results: 100% pass rate with zero missing documentation
🤖 Generated with Claude Code
Co-Authored-By: Claude <noreply@anthropic.com>
- Eliminated ALL ellipsis instances (0 remaining)
- Documented 40 functions with complete signatures and examples
- Documented 21 classes with full method and property documentation
- Added 56 method descriptions with detailed parameters and return values
- Included 15 complete property specifications
- Added 24 code examples and 38 explanatory notes
- Comprehensive coverage of all collection methods, system classes, and functions
Key highlights:
- EntityCollection/UICollection: Complete method docs (append, remove, extend, count, index)
- Animation: Full property and method documentation with examples
- Color: All manipulation methods (from_hex, to_hex, lerp) with examples
- Vector: Complete mathematical operations (magnitude, normalize, dot, distance_to, angle, copy)
- Scene: All management methods including register_keyboard
- Timer: Complete control methods (pause, resume, cancel, restart)
- Window: All management methods (get, center, screenshot)
- System functions: Complete audio, scene, UI, and system function documentation
File size: 54KB of professional HTML documentation
Test results: 100% pass rate with zero missing documentation
🤖 Generated with Claude Code
Co-Authored-By: Claude <noreply@anthropic.com>
- Fixed all formatting issues from original HTML output
- Added comprehensive constructor documentation for all classes
- Enhanced visual design with modern styling and typography
- Fixed literal newline display and markdown link conversion
- Added proper semantic HTML structure and navigation
- Includes detailed documentation for Entity, collections, and system types
🤖 Generated with Claude Code
Co-Authored-By: Claude <noreply@anthropic.com>
Implemented comprehensive API documentation generator that:
- Introspects live mcrfpy module for accurate documentation
- Generates organized Markdown reference (docs/API_REFERENCE.md)
- Categorizes classes and functions by type
- Includes full automation module documentation
- Provides summary statistics
Results:
- 20 classes documented
- 19 module functions documented
- 20 automation methods documented
- 100% coverage of public API
- Clean, readable Markdown output
Phase 7 Summary:
- Completed 4/5 tasks (1 cancelled as architecturally inappropriate)
- All documentation tasks successful
- Type stubs, docstrings, and API reference all complete
- McRogueFace now has professional-grade documentation
Test coverage included for all documentation features.
🤖 Generated with [Claude Code](https://claude.ai/code)
Co-Authored-By: Claude <noreply@anthropic.com>
Task #70 Analysis:
- Discovered fundamental incompatibility with PyPI distribution
- McRogueFace embeds CPython rather than being loaded by it
- Traditional wheels expect to extend existing Python interpreter
- Current architecture is application-with-embedded-Python
Decisions:
- Cancelled PyPI wheel preparation as out of scope for Alpha
- Cleaned up attempted packaging files (pyproject.toml, setup.py, etc.)
- Identified better distribution methods (installers, package managers)
Added Future Vision:
- Comprehensive plan for pure Python extension architecture
- Would allow true "pip install mcrogueface" experience
- Requires major refactoring to invert control flow
- Python would drive main loop with C++ performance extensions
- Unscheduled but documented as long-term possibility
This clarifies the architectural boundaries and sets realistic
expectations for distribution methods while preserving the vision
of what McRogueFace could become with significant rework.
🤖 Generated with [Claude Code](https://claude.ai/code)
Co-Authored-By: Claude <noreply@anthropic.com>
Created complete type stub files for the mcrfpy module to enable:
- Full IntelliSense/autocomplete in IDEs
- Static type checking with mypy/pyright
- Better documentation tooltips
- Parameter hints and return types
Implementation details:
- Manually crafted stubs for accuracy (15KB, 533 lines)
- Complete coverage: 19 classes, 112 functions/methods
- Proper type annotations using typing module
- @overload decorators for multiple signatures
- Type aliases for common patterns (UIElement union)
- Preserved all docstrings for IDE help
- Automation module fully typed
- PEP 561 compliant with py.typed marker
Testing:
- Validated Python syntax with ast.parse()
- Verified all expected classes and functions
- Confirmed type annotations are well-formed
- Checked docstring preservation (80 docstrings)
Usage:
- VS Code: Add stubs/ to python.analysis.extraPaths
- PyCharm: Mark stubs/ directory as Sources Root
- Other IDEs will auto-detect .pyi files
This significantly improves the developer experience when using
McRogueFace as a Python game engine.
🤖 Generated with [Claude Code](https://claude.ai/code)
Co-Authored-By: Claude <noreply@anthropic.com>
Enhanced documentation for the mcrfpy module with:
- Detailed docstrings for all API methods
- Type hints in documentation (name: type format)
- Return type specifications
- Exception documentation where applicable
- Usage examples for complex methods
- Module-level documentation with overview and example code
Specific improvements:
- Audio API: Added parameter types and return values
- Scene API: Documented transition types and error conditions
- Timer API: Clarified handler signature and runtime parameter
- UI Search: Added wildcard pattern examples for findAll()
- Metrics API: Documented all dictionary keys returned
Also fixed method signatures:
- Changed METH_VARARGS to METH_NOARGS for parameterless methods
- Ensures proper Python calling conventions
Test coverage included - all documentation is accessible via Python's
__doc__ attributes and shows correctly formatted information.
🤖 Generated with [Claude Code](https://claude.ai/code)
Co-Authored-By: Claude <noreply@anthropic.com>
Added proper Python docstrings for all UI component classes:
UIFrame:
- Container element that can hold child drawables
- Documents position, size, colors, outline, and clip_children
- Includes constructor signature with all parameters
UICaption:
- Text display element with font and styling
- Documents text content, position, font, colors, outline
- Notes that w/h are computed from text content
UISprite:
- Texture/sprite display element
- Documents position, texture, sprite_index, scale
- Notes that w/h are computed from texture and scale
UIGrid:
- Tile-based grid for game worlds
- Documents grid dimensions, tile size, texture atlas
- Includes entities collection and background_color
All docstrings follow consistent format:
- Constructor signature with defaults
- Brief description
- Args section with types and defaults
- Attributes section with all properties
This completes Phase 7 task #85 for documentation improvements.
🤖 Generated with [Claude Code](https://claude.ai/code)
Co-Authored-By: Claude <noreply@anthropic.com>
This major refactoring standardizes how position, size, and other arguments
are parsed across all UI components. PyArgHelpers provides consistent handling
for various argument patterns:
- Position as (x, y) tuple or separate x, y args
- Size as (w, h) tuple or separate width, height args
- Grid position and size with proper validation
- Color parsing with PyColorObject support
Changes across UI components:
- UICaption: Migrated to PyArgHelpers, improved resize() for future multiline support
- UIFrame: Uses standardized position parsing
- UISprite: Consistent position handling
- UIGrid: Grid-specific position/size helpers
- UIEntity: Unified argument parsing
Also includes:
- Improved error messages for type mismatches (int or float accepted)
- Reduced code duplication across constructors
- Better handling of keyword/positional argument conflicts
- Maintains backward compatibility with existing API
This addresses the inconsistent argument handling patterns discovered during
the inheritance hierarchy work and prepares for Phase 7 documentation.
🤖 Generated with [Claude Code](https://claude.ai/code)
Co-Authored-By: Claude <noreply@anthropic.com>
All UIDrawable-derived Python types now properly inherit from the Drawable
base class in Python, matching the C++ inheritance structure.
Changes:
- Add Py_TPFLAGS_BASETYPE to PyDrawableType to allow inheritance
- Set tp_base = &mcrfpydef::PyDrawableType for all UI types
- Add PyDrawable.h include to UI type headers
- Rename _Drawable to Drawable and update error message
This enables proper Python inheritance: Frame, Caption, Sprite, Grid,
and Entity all inherit from Drawable, allowing shared functionality
and isinstance() checks.
🤖 Generated with [Claude Code](https://claude.ai/code)
Co-Authored-By: Claude <noreply@anthropic.com>
- Update UISprite to use base class position instead of sprite position
- Synchronize sprite position with base class position for rendering
- Implement onPositionChanged() for position synchronization
- Update all UISprite methods to use base position consistently
- Add comprehensive test coverage for UISprite position handling
This is part 3 of moving position to the base class. UIGrid is the final
class that needs to be updated.
- Update UICaption to use base class position instead of text position
- Synchronize text position with base class position for rendering
- Add onPositionChanged() virtual method for position synchronization
- Update all UICaption methods to use base position consistently
- Add comprehensive test coverage for UICaption position handling
This is part 2 of moving position to the base class. UISprite and UIGrid
will be updated in subsequent commits.
- Add position member to UIDrawable base class
- Add common position getters/setters (x, y, pos) to base class
- Update UIFrame to use base class position instead of box position
- Synchronize box position with base class position for rendering
- Update all UIFrame methods to use base position consistently
- Add comprehensive test coverage for UIFrame position handling
This is part 1 of moving position to the base class. Other derived classes
(UICaption, UISprite, UIGrid) will be updated in subsequent commits.
- Remove redundant collision_pos field from UIEntity
- Update position getters/setters to use integer-cast position when needed
- Remove all collision_pos synchronization code
- Simplify entity position handling to use single float position field
- Add comprehensive test coverage proving functionality is preserved
This removes technical debt and simplifies the codebase without changing API behavior.
- Create PyArgHelpers.h with parsing functions for position, size, grid coordinates, and color
- Support tuple-based vector arguments with conflict detection
- Provide consistent error messages and validation
- Add comprehensive test coverage for infrastructure
This sets the foundation for standardizing all Python API constructors.
Phase 6 is now complete with all core rendering features implemented:
Completed Features:
- Grid background colors (#50) - customizable backgrounds with animation
- RenderTexture overhaul (#6) - UIFrame clipping with opt-in architecture
- Viewport-based rendering (#8) - three scaling modes with coordinate transform
Strategic Decisions:
- UIGrid already has optimal RenderTexture implementation for its viewport needs
- UICaption/UISprite clipping deemed unnecessary (no children to clip)
- Effects/Shader/Particle systems deferred to post-Phase 7 for focused delivery
The rendering foundation is now solid and ready for Phase 7: Documentation & Distribution.
🤖 Generated with [Claude Code](https://claude.ai/code)
Co-Authored-By: Claude <noreply@anthropic.com>
Implements a comprehensive viewport system that allows fixed game resolution
with flexible window scaling, addressing the primary wishes for issues #34, #49, and #8.
Key Features:
- Fixed game resolution independent of window size (window.game_resolution property)
- Three scaling modes accessible via window.scaling_mode:
- "center": 1:1 pixels, viewport centered in window
- "stretch": viewport fills window, ignores aspect ratio
- "fit": maintains aspect ratio with black bars
- Automatic window-to-game coordinate transformation for mouse input
- Full Python API integration with PyWindow properties
Technical Implementation:
- GameEngine::ViewportMode enum with Center, Stretch, Fit modes
- SFML View system for efficient GPU-based viewport scaling
- updateViewport() recalculates on window resize or mode change
- windowToGameCoords() transforms mouse coordinates correctly
- PyScene mouse input automatically uses transformed coordinates
Tests:
- test_viewport_simple.py: Basic API functionality
- test_viewport_visual.py: Visual verification with screenshots
- test_viewport_scaling.py: Interactive mode switching and resizing
This completes the viewport-based rendering task and provides the foundation
for resolution-independent game development as requested for Crypt of Sokoban.
🤖 Generated with [Claude Code](https://claude.ai/code)
Co-Authored-By: Claude <noreply@anthropic.com>
- Added RenderTexture support to UIDrawable base class
- std::unique_ptr<sf::RenderTexture> for opt-in rendering
- Dirty flag system for optimization
- enableRenderTexture() and markDirty() methods
- Implemented clip_children property for UIFrame
- Python-accessible boolean property
- Automatic RenderTexture creation when enabled
- Proper coordinate transformation for nested frames
- Updated UIFrame::render() for clipping support
- Renders to RenderTexture when clip_children=true
- Handles nested clipping correctly
- Only re-renders when dirty flag is set
- Added comprehensive dirty flag propagation
- All property setters mark frame as dirty
- Size changes recreate RenderTexture
- Animation system integration
- Created tests for clipping functionality
- Basic clipping test with visual verification
- Advanced nested clipping test
- Dynamic resize handling test
This is Phase 1 of the RenderTexture overhaul, providing the foundation
for advanced rendering effects like blur, glow, and viewport rendering.
- Comprehensive design for Issue #6 implementation
- Opt-in architecture to maintain backward compatibility
- Phased implementation plan with clear milestones
- Performance considerations and risk mitigation
- API design for clipping and future effects
Also includes Grid background color test
- Added sf::Color background_color member with default dark gray
- Python property getter/setter for background_color
- Animation support for individual color components (r/g/b/a)
- Replaces hardcoded clear color in render method
- Test demonstrates color changes and property access
Closes#50
- Window singleton with properties (resolution, fullscreen, vsync, title)
- OOP Scene support with lifecycle methods (on_enter, on_exit, on_keypress, update)
- Window resize events trigger scene.on_resize callbacks
- Scene transitions (fade, slide_left/right/up/down) with smooth animations
- Full integration of Python Scene objects with C++ engine
All Phase 5 tasks (#34, #1, #61, #105) completed successfully.
- Analyzed current SFML 2.6.1 usage throughout codebase
- Evaluated python-sfml (abandoned, only supports SFML 2.3.2)
- Recommended direct integration as mcrfpy.sfml module
- Created comprehensive SFML_EXPOSURE_RESEARCH.md with implementation plan
- Identified opportunity to provide modern SFML 2.6+ Python bindings
- Add ProfilingMetrics struct to track performance data
- Track frame time (current and 60-frame rolling average)
- Calculate FPS from average frame time
- Count draw calls, UI elements, and visible elements per frame
- Track total runtime and current frame number
- PyScene counts elements during render
- Expose metrics via mcrfpy.getMetrics() returning dict
This provides basic performance monitoring capabilities for
identifying bottlenecks and optimizing rendering performance.
🤖 Generated with [Claude Code](https://claude.ai/code)
Co-Authored-By: Claude <noreply@anthropic.com>
- UIFrame: Fix coordinate transformation (subtract parent pos, not add)
- UIFrame: Check children in reverse order (highest z-index first)
- UIFrame: Skip invisible elements entirely
- PyScene: Sort elements by z-index before checking clicks
- PyScene: Stop at first element that handles the click
- UIGrid: Implement entity click detection with grid coordinate transform
- UIGrid: Check entities in reverse order, return sprite as target
Click events now correctly respect z-order (top elements get priority),
handle coordinate transforms for nested frames, and support clicking
on grid entities. Elements without click handlers are transparent to
clicks, allowing elements below to receive them.
Note: Click testing requires non-headless mode due to PyScene limitation.
🤖 Generated with [Claude Code](https://claude.ai/code)
Co-Authored-By: Claude <noreply@anthropic.com>
- Add 'name' property to UIDrawable base class
- All UI elements (Frame, Caption, Sprite, Grid, Entity) support .name
- Entity delegates name to its sprite member
- Add find(name, scene=None) function for exact match search
- Add findAll(pattern, scene=None) with wildcard support (* matches any sequence)
- Both functions search recursively through Frame children and Grid entities
- Comprehensive test coverage for all functionality
This provides a simple way to find UI elements by name in Python scripts,
supporting both exact matches and wildcard patterns.
🤖 Generated with [Claude Code](https://claude.ai/code)
Co-Authored-By: Claude <noreply@anthropic.com>
- Add cleanup() method to GameEngine to clear Python references before destruction
- Clear timers and McRFPy_API references in proper order
- Call cleanup() at end of run loop and in destructor
- Ensure cleanup is only called once per GameEngine instance
Also includes:
- Fix audio ::stop() calls (already in place, OpenAL warning is benign)
- Add Caption support for x, y keywords (e.g. Caption("text", x=5, y=10))
- Refactor UIDrawable_methods.h into UIBase.h for better organization
- Move UIEntity-specific implementations to UIEntityPyMethods.h
🤖 Generated with [Claude Code](https://claude.ai/code)
Co-Authored-By: Claude <noreply@anthropic.com>
- Add visible, opacity properties to all UI classes (#87, #88)
- Add get_bounds(), move(), resize() methods to UIDrawable (#89, #98)
- Create UIDrawable_methods.h with template implementations
- Fix test termination issues - all tests now exit properly
- Fix test_sprite_texture_swap.py click handler signature
- Fix test_drawable_base.py segfault in headless mode
- Convert audio objects to pointers for cleanup (OpenAL warning persists)
- Remove debug print statements from UICaption
- Special handling for UIEntity to delegate drawable methods to sprite
All test files are now "airtight" - they complete successfully,
terminate on their own, and handle edge cases properly.
🤖 Generated with [Claude Code](https://claude.ai/code)
Co-Authored-By: Claude <noreply@anthropic.com>
- Document all completed tasks across three phases
- Show before/after API improvements
- Highlight technical achievements
- Outline next steps for Phase 4-7
- Add Color.from_hex(hex_string) class method for creating colors from hex
- Support formats: #RRGGBB, RRGGBB, #RRGGBBAA, RRGGBBAA
- Add color.to_hex() to convert Color to hex string
- Add color.lerp(other, t) for smooth color interpolation
- Comprehensive test coverage for all methods
🤖 Generated with [Claude Code](https://claude.ai/code)
Co-Authored-By: Claude <noreply@anthropic.com>
- Use PyConfig to set stdio_encoding="UTF-8" during initialization
- Set stdio_errors="surrogateescape" for robust handling
- Configure in both init_python() and init_python_with_config()
- Cleaner solution than wrapping streams after initialization
- Fixes UnicodeEncodeError when printing unicode characters
🤖 Generated with [Claude Code](https://claude.ai/code)
Co-Authored-By: Claude <noreply@anthropic.com>
- Remove entity from its grid's entity list
- Clear grid reference after removal
- Safe to call multiple times (no-op if not on grid)
- Works with shared_ptr entity management
- Add visibility bounds check in entity render loop
- Skip entities outside view with 1 cell margin
- Improves performance for large grids with many entities
- Bounds check considers zoom and pan settings
- Texture property getter/setter already implemented
- Position/scale preservation during swap confirmed
- Type validation for texture assignment working
- Tests verify functionality is complete
John McCardle