BSP: add Binary Space Partitioning for procedural dungeon generation

Implements #202, #203, #204, #205; partially implements #206:
- BSP class: core tree structure with bounds, split_once, split_recursive, clear
- BSPNode class: lightweight node reference with bounds, level, is_leaf,
  split_horizontal, split_position; navigation via left/right/parent/sibling;
  contains() and center() methods
- Traversal enum: PRE_ORDER, IN_ORDER, POST_ORDER, LEVEL_ORDER, INVERTED_LEVEL_ORDER
- BSP iteration: leaves() for leaf nodes only, traverse(order) for all nodes
- BSP query: find(pos) returns deepest node containing position
- BSP.to_heightmap(): converts BSP to HeightMap with select, shrink, value options

Note: #206's BSPMap subclass deferred - to_heightmap returns plain HeightMap.
The HeightMap already has all necessary operations (inverse, threshold, etc.)
for procedural generation workflows.

Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
This commit is contained in:
John McCardle 2026-01-12 07:02:54 -05:00
commit 8699bba9e6
5 changed files with 1703 additions and 0 deletions

View file

@ -0,0 +1,28 @@
"""Simple BSP test to identify crash."""
import sys
import mcrfpy
print("Step 1: Import complete")
print("Step 2: Creating BSP...")
bsp = mcrfpy.BSP(bounds=((0, 0), (100, 80)))
print("Step 2: BSP created:", bsp)
print("Step 3: Getting bounds...")
bounds = bsp.bounds
print("Step 3: Bounds:", bounds)
print("Step 4: Getting root...")
root = bsp.root
print("Step 4: Root:", root)
print("Step 5: Split once...")
bsp.split_once(horizontal=True, position=40)
print("Step 5: Split complete")
print("Step 6: Get leaves...")
leaves = list(bsp.leaves())
print("Step 6: Leaves count:", len(leaves))
print("PASS")
sys.exit(0)