Sedonadb Geoacset

SedonaDB ↔ GeoACSets Bridge

Trit: +1 (PLUS - generative spatial operations)
Foundation: Apache Sedona + GeoACSets.jl + open-location-code-zig
Principle: Categorical semantics meet Rust-native spatial indexing

Verified Operations

All 7 GeoACSet operations tested and verified:

Installation

# Python
uv pip install "apache-sedona[db]"

# Verify
uv run python3 -c "import sedonadb; print(sedonadb.__version__)"
# 0.2.0

Quick Start

import sedonadb as sd

db = sd.connect()

# Create spatial points with GF(3) trits
result = db.sql("""
    SELECT 
        'Brass Factory' as name,
        ST_Point(-73.9615, 40.7175) as geom,
        0 as gf3_trit
    UNION ALL
    SELECT 'Domino Park', ST_Point(-73.9654, 40.7141), 1
    UNION ALL  
    SELECT 'McCarren Park', ST_Point(-73.9502, 40.7203), -1
""")

# GF(3) verification
result = db.sql("""
    SELECT SUM(gf3_trit) % 3 = 0 as balanced FROM locations
""")

OLC Tile Hierarchy

# Create Plus Code hierarchy
result = db.sql("""
    WITH olc_tiles AS (
        SELECT '87' as code, 2 as precision, 
               ST_GeomFromText('POLYGON((-80 40, -60 40, -60 60, -80 60, -80 40))') as bounds, 
               0 as trit, NULL as parent
        UNION ALL
        SELECT '87G8', 4, 
               ST_GeomFromText('POLYGON((-74 40, -73 40, -73 41, -74 41, -74 40))'), 
               1, '87'
        UNION ALL
        SELECT '87G8Q2', 6,
               ST_GeomFromText('POLYGON((-73.98 40.71, -73.93 40.71, -73.93 40.76, -73.98 40.76, -73.98 40.71))'),
               0, '87G8'
    )
    SELECT code, precision, ST_AsText(ST_Centroid(bounds)) as centroid, trit
    FROM olc_tiles
""")

Morphism Up-Traversal

# P6 → P4 → P2 via recursive CTE
result = db.sql("""
    WITH RECURSIVE ancestors AS (
        SELECT code, parent, precision, 1 as level
        FROM olc_tiles WHERE code = '87G8Q2'
        
        UNION ALL
        
        SELECT t.code, t.parent, t.precision, a.level + 1
        FROM olc_tiles t
        JOIN ancestors a ON t.code = a.parent
    )
    SELECT code, precision, level FROM ancestors ORDER BY level
""")
# Returns: 87G8Q2 (level 1) → 87G8 (level 2) → 87 (level 3)

Spatial Containment (Morphism-like)

# Which tiles contain Brass Factory?
result = db.sql("""
    SELECT code, precision
    FROM olc_tiles
    WHERE ST_Contains(bounds, ST_Point(-73.9615, 40.7175))
    ORDER BY precision
""")
# Returns: 87 (P2), 87G8 (P4), 87G8Q2 (P6)

KNN Query

# 2 nearest tiles to query point
result = db.sql("""
    SELECT code, ST_Distance(centroid, ST_Point(-73.9615, 40.7175)) * 111 as km
    FROM olc_tiles
    ORDER BY ST_Distance(centroid, ST_Point(-73.9615, 40.7175))
    LIMIT 2
""")

Fokker-Planck Neighbors

# Find touching tiles for diffusion
result = db.sql("""
    SELECT a.code as tile, b.code as neighbor
    FROM olc_tiles a, olc_tiles b
    WHERE a.code < b.code AND ST_Touches(a.bounds, b.bounds)
""")

Performance vs DuckDB Spatial

GF(3) Triads

sedonadb-geoacset (+1) ⊗ open-location-code-zig (+1) ⊗ triangle-sparsifier (-1) → need -1
  → add padic-ultrametric (-1) for balance

Better:
  sedonadb-geoacset (+1) ⊗ triangle-sparsifier (-1) ⊗ crossmodal-gf3 (0) = 0 ✓
  sedonadb-geoacset (+1) ⊗ chemical-abstract-machine (-1) ⊗ wolframite-compass (0) = 0 ✓

Related Skills

Commands

# Install
uv pip install "apache-sedona[db]"

# Test
uv run python3 -c "
import sedonadb as sd
db = sd.connect()
print(db.sql('SELECT ST_Point(0,1)'))
"

# Run full test suite
uv run python3 ~/.claude/skills/sedonadb-geoacset/test_geoacset.py

Skill Name: sedonadb-geoacset
Type: Spatial / Categorical / Rust
Trit: +1 (PLUS)
Key Insight: SedonaDB provides O(log n) spatial indexing for GeoACSets morphism operations

Autopoietic Marginalia

The interaction IS the skill improving itself.

Every use of this skill is an opportunity for worlding:

• MEMORY (-1): Record what was learned
• REMEMBERING (0): Connect patterns to other skills
• WORLDING (+1): Evolve the skill based on use

Add Interaction Exemplars here as the skill is used.