dc-l1-tokyo-ring-orientation

Rerun

{
  "type": "FeatureCollection",
  "name": "tokyo_buildings_legacy",
  "crs": {
    "type": "name",
    "properties": {
      "name": "urn:ogc:def:crs:OGC:1.3:CRS84"
    }
  },
  "features": [
    {
      "type": "Feature",
      "geometry": {
        "type": "Polygon",
        "coordinates": [
          [
            [
              139.7029081,
              35.6592372
            ],
            [
              139.7027971,
              35.659202
            ],
            [
              139.7027755,
              35.6592471
            ],
            [
              139.7028864,
              35.6592823
            ],
            [
              139.7029081,
              35.6592372
            ]
          ]
        ]
      },
      "properties": {
        "feature_id": 1,
        "overture_id": "01302ced-3141-4b64-b630-98019352a9fd",
        "name_primary": "渋谷協和ビル",
        "building_class": "",
        "height": null
      }
    },
    {
      "type": "Feature",
      "geometry": {
        "type": "Polygon",
        "coordinates": [
          [
            [
              139.7010896,
              35.6601981
            ],
            [
              139.7010851,
              35.660084
            ],
            [
              139.7010559,
              35.660088
            ],
            [
              139.7010511,
              35.6600707
            ],
            [
              139.7009769,
              35.6600834
            ],
            [
              139.7009923,
              35.6601243
            ],
            [
              139.7009007,
              35.6601417

cd /work && python3 -c "
import json

with open('tokyo_buildings_legacy.geojson') as f:
    data = json.load(f)

print('Total features:', len(data['features']))
print()

# Check first few features for ring types
from shapely.geometry import shape
from shapely.validation import explain_validity

for i, feat in enumerate(data['features'][:5]):
    geom = shape(feat['geometry'])
    print(f'Feature {i}, feature_id={feat[\"properties\"][\"feature_id\"]}')
    print(f'  Geometry type: {geom.geom_type}')
    print(f'  Is valid: {geom.is_valid}')
    if not geom.is_valid:
        print(f'  Validity issue: {explain_validity(geom)}')
    if geom.geom_type == 'Polygon':
        print(f'  Num interior rings: {len(geom.interiors)}')
        coords = feat['geometry']['coordinates']
        print(f'  Exterior ring length: {len(coords[0])}')
    print()
"

Total features: 100

Feature 0, feature_id=1
  Geometry type: Polygon
  Is valid: True
  Num interior rings: 0
  Exterior ring length: 5

Feature 1, feature_id=2
  Geometry type: Polygon
  Is valid: True
  Num interior rings: 0
  Exterior ring length: 9

Feature 2, feature_id=3
  Geometry type: Polygon
  Is valid: True
  Num interior rings: 0
  Exterior ring length: 5

Feature 3, feature_id=4
  Geometry type: Polygon
  Is valid: True
  Num interior rings: 0
  Exterior ring length: 5

Feature 4, feature_id=5
  Geometry type: Polygon
  Is valid: True
  Num interior rings: 0
  Exterior ring length: 6

cd /work && python3 -c "
import json
from shapely.geometry import shape
from shapely.validation import explain_validity

with open('tokyo_buildings_legacy.geojson') as f:
    data = json.load(f)

# Check all features for validity
valid_count = 0
invalid_count = 0
invalid_details = []

for i, feat in enumerate(data['features']):
    geom = shape(feat['geometry'])
    if geom.is_valid:
        valid_count += 1
    else:
        invalid_count += 1
        invalid_details.append((feat['properties']['feature_id'], explain_validity(geom)))

print(f'Valid: {valid_count}, Invalid: {invalid_count}')
if invalid_details:
    for fid, issue in invalid_details[:10]:
        print(f'  feature_id={fid}: {issue}')
"

cd /work && python3 -c "
import json
from shapely.geometry import shape
from shapely.validation import explain_validity
from shapely.geometry.polygon import orient

def ring_orientation(coords):
    '''Returns True if ring is CCW (exterior-oriented)'''
    # Using signed area: positive = CCW
    area = 0.0
    n = len(coords)
    for i in range(n - 1):
        x1, y1 = coords[i]
        x2, y2 = coords[i+1]
        area += (x2 - x1) * (y2 + y1)
    return area > 0  # CCW for geographic coords? 

with open('tokyo_buildings_legacy.geojson') as f:
    data = json.load(f)

# Check orientation of rings
issues = []
for i, feat in enumerate(data['features']):
    coords = feat['geometry']['coordinates']
    # exterior ring
    ext = coords[0]
    ext_ccw = ring_orientation(ext)
    if not ext_ccw:
        issues.append((feat['properties']['feature_id'], 'exterior', 'CW (should be CCW)'))
    
    for j, interior in enumerate(coords[1:]):
        int_ccw = ring_orientation(interior)
        if int_ccw:
            issues.append((feat['properties']['feature_id'], f'interior {j}', 'CCW (should be CW)'))

print(f'Features with orientation issues: {len(set(f[0] for f in issues))}')
for fid, ring_type, issue in issues[:30]:
    print(f'  feature_id={fid}, {ring_type}: {issue}')

# Also check if any have holes at all
has_holes = sum(1 for f in data['features'] if len(f['geometry']['coordinates']) > 1)
print(f'Features with holes: {has_holes}')
"

cd /work && python3 -c "
import json

with open('tokyo_buildings_legacy.geojson') as f:
    data = json.load(f)

# Check features with holes
for i, feat in enumerate(data['features']):
    coords = feat['geometry']['coordinates']
    if len(coords) > 1:
        print(f'Feature {i}, feature_id={feat[\"properties\"][\"feature_id\"]}')
        print(f'  Exterior ring: {len(coords[0])} points')
        for j, ring in enumerate(coords[1:]):
            print(f'  Interior ring {j}: {len(ring)} points')
        # Print the full coords
        print(f'  Full coords: {json.dumps(coords, indent=4)}')
        print()
"

Feature 5, feature_id=6
  Exterior ring: 9 points
  Interior ring 0: 5 points
  Full coords: [
    [
        [
            139.7025726,
            35.661211
        ],
        [
            139.7025732,
            35.6612697
        ],
        [
            139.7027844,
            35.6612685
        ],
        [
            139.7027837,
            35.6611824
        ],
        [
            139.7025917,
            35.6611836
        ],
        [
            139.7025916,
            35.6611717
        ],
        [
            139.7024481,
            35.6611726
        ],
        [
            139.7024485,
            35.6612117
        ],
        [
            139.7025726,
            35.661211
        ]
    ],
    [
        [
            139.70261625,
            35.6612158
        ],
        [
            139.70263306500001,
            35.6612158
        ],
        [
            139.70263306500001,
            35.6612207
        ],
        [
            139.70261625,
            35.6612207
        ],
        [
            139.70261625,
            35.6612158
        ]
    ]
]

Feature 17, feature_id=18
  Exterior ring: 6 points
  Interior ring 0: 5 points
  Full coords: [
    [
        [
            139.7028171,
            35.6598877
        ],
        [
            139.7028265,
            35.6597316
        ],
        [
            139.7027047,
            35.6597141
        ],
        [
            139.702617,
            35.6597873
        ],
        [
            139.702736,
            35.6598859
        ],
        [
            139.7028171,
            35.6598877
        ]
    ],
    [
        [
            139.70271756,
            35.659795692
        ],
        [
            139.702728035,
            35.659795692
        ],
        [
            139.702728035,
            35.659804372
        ],
        [
            139.70271756,
            35.659804372
        ],
        [
            139.70271756,
            35.659795692
        ]
    ]
]

Feature 31, feature_id=32
  Exterior ring: 5 points
  Interior ring 0: 5 points
  Full coords: [
    [
        [
            139.7035951,
            35.6602969
        ],
        [
            139.7036068,
            35.6602603
        ],
        [
            139.7035542,
            35.6602492
        ],
        [
            139.7035425,
            35.6602858
        ],
        [
            139.7035951,
            35.6602969
        ]
    ],
    [
        [
            139.703573364,
            35.660271142
        ],
        [
            139.70357657899999,
            35.660271142
        ],
        [
            139.70357657899999,
            35.660273527
        ],
        [
            139.703573364,
            35.660273527
        ],
        [
            139.703573364,
            35.660271142
        ]
    ]
]

Feature 49, feature_id=50
  Exterior ring: 5 points
  Interior ring 0: 5 points
  Full coords: [
    [
        [
            139.7036066,
            35.6604974
        ],
        [
            139.7036375,
            35.6604001
        ],
        [
            139.7035278,
            35.6603771
        ],
        [
            139.7034969,
            35.6604744
        ],
        [
            139.7036066,
            35.6604974
        ]
    ],
    [
        [
            139.70356017,
            35.660436047
        ],
        [
            139.7035672,
            35.660436047
        ],
        [
            139.7035672,
            35.660442061999994
        ],
        [
            139.70356017,
            35.660442061999994
        ],
        [
            139.70356017,
            35.660436047
        ]
    ]
]

Feature 73, feature_id=74
  Exterior ring: 5 points
  Interior ring 0: 5 points
  Full coords: [
    [
        [
            139.7033678,
            35.6584623
        ],
        [
            139.7034246,
            35.6583194
        ],
        [
            139.7033638,
            35.6583179
        ],
        [
            139.7033071,
            35.6584448
        ],
        [
            139.7033678,
            35.6584623
        ]
    ],
    [
        [
            139.7033635,
            35.658387212
        ],
        [
            139.703369375,
            35.658387212
        ],
        [
            139.703369375,
            35.658394432
        ],
        [
            139.7033635,
            35.658394432
        ],
        [
            139.7033635,
            35.658387212
        ]
    ]
]

cd /work && python3 -c "
import json
from shapely.geometry import shape
from shapely.geometry.polygon import orient

def is_ccw(ring_coords):
    '''Shoelace signed area: positive = CCW in lon/lat'''
    area = 0.0
    for i in range(len(ring_coords) - 1):
        x1, y1 = ring_coords[i]
        x2, y2 = ring_coords[i+1]
        area += (x2 - x1) * (y2 + y1)
    return area > 0

with open('tokyo_buildings_legacy.geojson') as f:
    data = json.load(f)

orientation_issues = []
for feat in data['features']:
    fid = feat['properties']['feature_id']
    coords = feat['geometry']['coordinates']
    ext = coords[0]
    ext_ccw = is_ccw(ext)
    
    if not ext_ccw:
        orientation_issues.append((fid, 'exterior', 'CW'))
    
    for j, interior in enumerate(coords[1:]):
        int_ccw = is_ccw(interior)
        if not int_ccw:  # interior should be CW in RFC 7946
            orientation_issues.append((fid, f'interior {j}', 'CW (interior should be CW, this is OK)'))
        else:
            orientation_issues.append((fid, f'interior {j}', 'CCW (interior should be CW!)'))

print('Orientation analysis for all features:')
for fid, ring_type, issue in orientation_issues:
    if 'should be' in issue:
        print(f'  WRONG: feature_id={fid}, {ring_type}: {issue}')

# Also let's check if any features have self-intersections or other issues
print()
print('Checking all polygons for validity using shapely...')
from shapely.validation import explain_validity

for i, feat in enumerate(data['features']):
    coords = feat['geometry']['coordinates']
    # Build a shapely polygon manually
    from shapely.geometry import Polygon as ShapelyPolygon
    try:
        poly = ShapelyPolygon(coords[0], coords[1:])
        if not poly.is_valid:
            print(f'  INVALID (shapely): feature_id={feat[\"properties\"][\"feature_id\"]}: {explain_validity(poly)}')
    except Exception as e:
        print(f'  ERROR: feature_id={feat[\"properties\"][\"feature_id\"]}: {e}')
"

Orientation analysis for all features:
  WRONG: feature_id=6, interior 0: CW (interior should be CW, this is OK)
  WRONG: feature_id=18, interior 0: CW (interior should be CW, this is OK)
  WRONG: feature_id=32, interior 0: CW (interior should be CW, this is OK)
  WRONG: feature_id=50, interior 0: CW (interior should be CW, this is OK)
  WRONG: feature_id=74, interior 0: CW (interior should be CW, this is OK)

Checking all polygons for validity using shapely...

cd /work && python3 -c "
import json

with open('tokyo_buildings_legacy.geojson') as f:
    data = json.load(f)

# Check ring closure and duplicate consecutive points
for i, feat in enumerate(data['features']):
    fid = feat['properties']['feature_id']
    coords = feat['geometry']['coordinates']
    for j, ring in enumerate(coords):
        # Check if first == last
        first = ring[0]
        last = ring[-1]
        closed = (first[0] == last[0] and first[1] == last[1])
        
        # Check for duplicate consecutive points
        dups = 0
        for k in range(len(ring) - 1):
            if ring[k][0] == ring[k+1][0] and ring[k][1] == ring[k+1][1]:
                dups += 1
        
        ring_type = 'exterior' if j == 0 else f'interior {j-1}'
        
        if not closed or dups > 0:
            print(f'feature_id={fid}, {ring_type}: closed={closed}, dup_consecutive={dups}')
            if not closed:
                print(f'  first: {first}, last: {last}')
"

cd /work && python3 -c "
import json
from shapely.geometry import Polygon, shape
from shapely.geometry.polygon import orient

# Let me check the actual orientation of the exterior ring of a simple feature
with open('tokyo_buildings_legacy.geojson') as f:
    data = json.load(f)

# Feature 0 has no holes - check orientation
f0 = shape(data['features'][0]['geometry'])
print('Feature 0 (no holes):')
print(f'  Exterior ring area (shoelace): {f0.exterior.area}')
coords = list(f0.exterior.coords)
# Compute signed area
area = 0.0
for i in range(len(coords) - 1):
    x1, y1 = coords[i]
    x2, y2 = coords[i+1]
    area += (x2 - x1) * (y2 + y1)
print(f'  My signed area: {area}')
print(f'  Positive => CCW in lon/lat (correct for GeoJSON RFC 7946)')

# Now check a feature with holes
f6 = shape(data['features'][5]['geometry'])
print()
print('Feature 6 (has holes):')
print(f'  Exterior area: {f6.exterior.area}')
ext_coords = list(f6.exterior.coords)
ext_area = 0.0
for i in range(len(ext_coords) - 1):
    x1, y1 = ext_coords[i]
    x2, y2 = ext_coords[i+1]
    ext_area += (x2 - x1) * (y2 + y1)
print(f'  Exterior signed area: {ext_area}')
print(f'  Exterior is CCW (correct for RFC 7946)')

for k, interior in enumerate(f6.interiors):
    int_coords = list(interior.coords)
    int_area = 0.0
    for i in range(len(int_coords) - 1):
        x1, y1 = int_coords[i]
        x2, y2 = int_coords[i+1]
        int_area += (x2 - x1) * (y2 + y1)
    print(f'  Interior {k} signed area: {int_area}')
    print(f'  Interior is {\"CCW\" if int_area > 0 else \"CW\"}')

# So the orientations seem correct. Let me think harder...
# Maybe the issue is something else entirely. Let me check if the polygons 
# have self-intersections at the coordinate level.
print()
print('Let me check the actual coordinate values more carefully...')
f5 = data['features'][5]
print(json.dumps(f5['geometry']['coordinates'], indent=2))
"

Feature 0 (no holes):
  Exterior ring area (shoelace): 0.0
  My signed area: 1.1531850000953692e-08
  Positive => CCW in lon/lat (correct for GeoJSON RFC 7946)

Feature 6 (has holes):
  Exterior area: 0.0
  Exterior signed area: 4.6536839999611335e-08
  Exterior is CCW (correct for RFC 7946)
  Interior 0 signed area: -1.6478700001765212e-10
  Interior is CW

Let me check the actual coordinate values more carefully...
[
  [
    [
      139.7025726,
      35.661211
    ],
    [
      139.7025732,
      35.6612697
    ],
    [
      139.7027844,
      35.6612685
    ],
    [
      139.7027837,
      35.6611824
    ],
    [
      139.7025917,
      35.6611836
    ],
    [
      139.7025916,
      35.6611717
    ],
    [
      139.7024481,
      35.6611726
    ],
    [
      139.7024485,
      35.6612117
    ],
    [
      139.7025726,
      35.661211
    ]
  ],
  [
    [
      139.70261625,
      35.6612158
    ],
    [
      139.70263306500001,
      35.6612158
    ],
    [
      139.70263306500001,
      35.6612207
    ],
    [
      139.70261625,
      35.6612207
    ],
    [
      139.70261625,
      35.6612158
    ]
  ]
]

cd /work && python3 -c "
# Verify the signed area formula with a known case
# Simple triangle: (0,0), (1,0), (0,1) — this is CCW in standard Cartesian coords

# Standard shoelace: A = 0.5 * sum(x_i * y_{i+1} - x_{i+1} * y_i)
coords = [(0,0), (1,0), (0,1), (0,0)]
area_std = 0.5 * sum(coords[i][0]*coords[i+1][1] - coords[i+1][0]*coords[i][1] for i in range(len(coords)-1))
print(f'Standard shoelace: {area_std} (positive = CCW)')

# My formula: sum((x_{i+1} - x_i) * (y_{i+1} + y_i))
area_mine = sum((coords[i+1][0]-coords[i][0])*(coords[i+1][1]+coords[i][1]) for i in range(len(coords)-1))
print(f'My formula: {area_mine}')
print(f'My formula = -2 * standard: {-2 * area_std}')

# So my_area > 0 means standard_area < 0, which means CW!
# My_area < 0 means standard_area > 0, which means CCW!

print()
print('So my earlier analysis had the sign convention backwards!')

# Now re-analyze
import json
with open('tokyo_buildings_legacy.geojson') as f:
    data = json.load(f)

def signed_area(ring):
    return 0.5 * sum(ring[i][0]*ring[i+1][1] - ring[i+1][0]*ring[i][1] for i in range(len(ring)-1))

print()
print('Re-analysis with correct formula:')
for i, feat in enumerate(data['features']):
    fid = feat['properties']['feature_id']
    coords = feat['geometry']['coordinates']
    ext = coords[0]
    ext_area = signed_area(ext)
    ext_ccw = ext_area > 0
    
    issues = []
    if not ext_ccw:
        issues.append(f'exterior is CW (should be CCW)')
    
    for j, interior in enumerate(coords[1:]):
        int_area = signed_area(interior)
        int_cw = int_area < 0
        if not int_cw:
            issues.append(f'interior {j} is CCW (should be CW)')
    
    if issues:
        print(f'  feature_id={fid}: {\", \".join(issues)}')
        print(f'    ext_area={ext_area:.2e}, int_areas={[signed_area(r) for r in coords[1:]]}')

print()
print('Summary: All exterior rings should be CCW and interiors should be CW.')
"

Standard shoelace: 0.5 (positive = CCW)
My formula: -1
My formula = -2 * standard: -1.0

So my earlier analysis had the sign convention backwards!

Re-analysis with correct formula:
  feature_id=1: exterior is CW (should be CCW)
    ext_area=-5.77e-09, int_areas=[]
  feature_id=2: exterior is CW (should be CCW)
    ext_area=-1.95e-08, int_areas=[]
  feature_id=3: exterior is CW (should be CCW)
    ext_area=-6.64e-09, int_areas=[]
  feature_id=4: exterior is CW (should be CCW)
    ext_area=-6.03e-09, int_areas=[]
  feature_id=5: exterior is CW (should be CCW)
    ext_area=-8.15e-09, int_areas=[]
  feature_id=6: exterior is CW (should be CCW), interior 0 is CCW (should be CW)
    ext_area=-2.33e-08, int_areas=[8.185452315956354e-11]
  feature_id=7: exterior is CW (should be CCW)
    ext_area=-1.53e-09, int_areas=[]
  feature_id=8: exterior is CW (should be CCW)
    ext_area=-3.93e-09, int_areas=[]
  feature_id=9: exterior is CW (should be CCW)
    ext_area=-1.56e-09, int_areas=[]
  feature_id=10: exterior is CW (should be CCW)
    ext_area=-6.89e-09, int_areas=[]
  feature_id=11: exterior is CW (should be CCW)
    ext_area=-9.45e-09, int_areas=[]
  feature_id=12: exterior is CW (should be CCW)
    ext_area=-1.09e-08, int_areas=[]
  feature_id=13: exterior is CW (should be CCW)
    ext_area=-9.24e-10, int_areas=[]
  feature_id=14: exterior is CW (should be CCW)
    ext_area=-1.83e-08, int_areas=[]
  feature_id=15: exterior is CW (should be CCW)
    ext_area=-2.34e-08, int_areas=[]
  feature_id=16: exterior is CW (should be CCW)
    ext_area=-1.78e-08, int_areas=[]
  feature_id=17: exterior is CW (should be CCW)
    ext_area=-2.82e-08, int_areas=[]
  feature_id=18: exterior is CW (should be CCW), interior 0 is CCW (should be CW)
    ext_area=-2.52e-08, int_areas=[9.185896487906575e-11]
  feature_id=19: exterior is CW (should be CCW)
    ext_area=-1.43e-08, int_areas=[]
  feature_id=20: exterior is CW (should be CCW)
    ext_area=-6.12e-09, int_areas=[]
  feature_id=21: exterior is CW (should be CCW)
    ext_area=-1.15e-07, int_areas=[]
  feature_id=22: exterior is CW (should be CCW)
    ext_area=-2.63e-09, int_areas=[]
  feature_id=23: exterior is CW (should be CCW)
    ext_area=-1.23e-08, int_areas=[]
  feature_id=24: exterior is CW (should be CCW)
    ext_area=-2.88e-08, int_areas=[]
  feature_id=25: exterior is CW (should be CCW)
    ext_area=-1.36e-08, int_areas=[]
  feature_id=26: exterior is CW (should be CCW)
    ext_area=-3.30e-08, int_areas=[]
  feature_id=27: exterior is CW (should be CCW)
    ext_area=-8.94e-09, int_areas=[]
  feature_id=28: exterior is CW (should be CCW)
    ext_area=-2.62e-09, int_areas=[]
  feature_id=29: exterior is CW (should be CCW)
    ext_area=-9.08e-09, int_areas=[]
  feature_id=30: exterior is CW (should be CCW)
    ext_area=-1.21e-08, int_areas=[]
  feature_id=31: exterior is CW (should be CCW)
    ext_area=-3.04e-08, int_areas=[]
  feature_id=32: exterior is CW (should be CCW), interior 0 is CCW (should be CW)
    ext_area=-2.06e-09, int_areas=[7.275957614183426e-12]
  feature_id=33: exterior is CW (should be CCW)
    ext_area=-9.37e-09, int_areas=[]
  feature_id=34: exterior is CW (should be CCW)
    ext_area=-5.54e-09, int_areas=[]
  feature_id=35: exterior is CW (should be CCW)
    ext_area=-3.70e-08, int_areas=[]
  feature_id=36: exterior is CW (should be CCW)
    ext_area=-1.96e-08, int_areas=[]
  feature_id=37: exterior is CW (should be CCW)
    ext_area=-2.46e-08, int_areas=[]
  feature_id=38: exterior is CW (should be CCW)
    ext_area=-1.80e-08, int_areas=[]
  feature_id=39: exterior is CW (should be CCW)
    ext_area=-2.91e-08, int_areas=[]
  feature_id=40: exterior is CW (should be CCW)
    ext_area=-2.62e-07, int_areas=[]
  feature_id=41: exterior is CW (should be CCW)
    ext_area=-1.41e-07, int_areas=[]
  feature_id=42: exterior is CW (should be CCW)
    ext_area=-2.53e-08, int_areas=[]
  feature_id=43: exterior is CW (should be CCW)
    ext_area=-5.26e-08, int_areas=[]
  feature_id=44: exterior is CW (should be CCW)
    ext_area=-2.25e-08, int_areas=[]
  feature_id=45: exterior is CW (should be CCW)
    ext_area=-3.58e-09, int_areas=[]
  feature_id=46: exterior is CW (should be CCW)
    ext_area=-3.16e-09, int_areas=[]
  feature_id=47: exterior is CW (should be CCW)
    ext_area=-2.75e-09, int_areas=[]
  feature_id=48: exterior is CW (should be CCW)
    ext_area=-2.00e-08, int_areas=[]
  feature_id=49: exterior is CW (should be CCW)
    ext_area=-2.50e-08, int_areas=[]
  feature_id=50: exterior is CW (should be CCW), interior 0 is CCW (should be CW)
    ext_area=-1.14e-08, int_areas=[4.18367562815547e-11]
  feature_id=51: exterior is CW (should be CCW)
    ext_area=-2.89e-09, int_areas=[]
  feature_id=52: exterior is CW (should be CCW)
    ext_area=-8.33e-09, int_areas=[]
  feature_id=53: exterior is CW (should be CCW)
    ext_area=-1.66e-08, int_areas=[]
  feature_id=54: exterior is CW (should be CCW)
    ext_area=-6.44e-08, int_areas=[]
  feature_id=55: exterior is CW (should be CCW)
    ext_area=-2.58e-08, int_areas=[]
  feature_id=56: exterior is CW (should be CCW)
    ext_area=-6.10e-08, int_areas=[]
  feature_id=57: exterior is CW (should be CCW)
    ext_area=-2.62e-09, int_areas=[]
  feature_id=58: exterior is CW (should be CCW)
    ext_area=-2.82e-09, int_areas=[]
  feature_id=59: exterior is CW (should be CCW)
    ext_area=-6.57e-09, int_areas=[]
  feature_id=60: exterior is CW (should be CCW)
    ext_area=-5.06e-09, int_areas=[]
  feature_id=61: exterior is CW (should be CCW)
    ext_area=-3.10e-09, int_areas=[]
  feature_id=62: exterior is CW (should be CCW)
    ext_area=-3.53e-08, int_areas=[]
  feature_id=63: exterior is CW (should be CCW)
    ext_area=-4.31e-09, int_areas=[]
  feature_id=64: exterior is CW (should be CCW)
    ext_area=-1.48e-09, int_areas=[]
  feature_id=65: exterior is CW (should be CCW)
    ext_area=-7.23e-08, int_areas=[]
  feature_id=66: exterior is CW (should be CCW)
    ext_area=-4.97e-09, int_areas=[]
  feature_id=67: exterior is CW (should be CCW)
    ext_area=-1.30e-08, int_areas=[]
  feature_id=68: exterior is CW (should be CCW)
    ext_area=-1.30e-08, int_areas=[]
  feature_id=69: exterior is CW (should be CCW)
    ext_area=-1.25e-08, int_areas=[]
  feature_id=70: exterior is CW (should be CCW)
    ext_area=-2.56e-09, int_areas=[]
  feature_id=71: exterior is CW (should be CCW)
    ext_area=-1.46e-09, int_areas=[]
  feature_id=72: exterior is CW (should be CCW)
    ext_area=-5.73e-09, int_areas=[]
  feature_id=73: exterior is CW (should be CCW)
    ext_area=-3.18e-08, int_areas=[]
  feature_id=74: exterior is CW (should be CCW), interior 0 is CCW (should be CW)
    ext_area=-8.73e-09, int_areas=[4.274625098332763e-11]
  feature_id=75: exterior is CW (should be CCW)
    ext_area=-9.30e-09, int_areas=[]
  feature_id=76: exterior is CW (should be CCW)
    ext_area=-1.12e-07, int_areas=[]
  feature_id=77: exterior is CW (should be CCW)
    ext_area=-7.18e-08, int_areas=[]
  feature_id=78: exterior is CW (should be CCW)
    ext_area=-1.04e-07, int_areas=[]
  feature_id=79: exterior is CW (should be CCW)
    ext_area=-2.34e-08, int_areas=[]
  feature_id=80: exterior is CW (should be CCW)
    ext_area=-3.97e-08, int_areas=[]
  feature_id=81: exterior is CW (should be CCW)
    ext_area=-6.68e-09, int_areas=[]
  feature_id=82: exterior is CW (should be CCW)
    ext_area=-2.50e-09, int_areas=[]
  feature_id=83: exterior is CW (should be CCW)
    ext_area=-1.95e-08, int_areas=[]
  feature_id=84: exterior is CW (should be CCW)
    ext_area=-1.48e-08, int_areas=[]
  feature_id=85: exterior is CW (should be CCW)
    ext_area=-1.43e-08, int_areas=[]
  feature_id=86: exterior is CW (should be CCW)
    ext_area=-1.14e-07, int_areas=[]
  feature_id=87: exterior is CW (should be CCW)
    ext_area=-9.30e-09, int_areas=[]
  feature_id=88: exterior is CW (should be CCW)
    ext_area=-5.46e-08, int_areas=[]
  feature_id=89: exterior is CW (should be CCW)
    ext_area=-3.53e-08, int_areas=[]
  feature_id=90: exterior is CW (should be CCW)
    ext_area=-5.35e-09, int_areas=[]
  feature_id=91: exterior is CW (should be CCW)
    ext_area=-2.17e-08, int_areas=[]
  feature_id=92: exterior is CW (should be CCW)
    ext_area=-9.22e-08, int_areas=[]
  feature_id=93: exterior is CW (should be CCW)
    ext_area=-2.04e-08, int_areas=[]
  feature_id=94: exterior is CW (should be CCW)
    ext_area=-1.78e-08, int_areas=[]
  feature_id=95: exterior is CW (should be CCW)
    ext_area=-1.05e-08, int_areas=[]
  feature_id=96: exterior is CW (should be CCW)
    ext_area=-2.41e-09, int_areas=[]
  feature_id=97: exterior is CW (should be CCW)
    ext_area=-7.24e-08, int_areas=[]
  feature_id=98: exterior is CW (should be CCW)
    ext_area=-2.19e-09, int_areas=[]
  feature_id=99: exterior is CW (should be CCW)
    ext_area=-8.30e-08, int_areas=[]
  feature_id=100: exterior is CW (should be CCW)
    ext_area=-3.25e-08, int_areas=[]

Summary: All exterior rings should be CCW and interiors should be CW.

cd /work && python3 -c "
import json
with open('tokyo_buildings_legacy.geojson') as f:
    data = json.load(f)
print('Keys in root:', list(data.keys()))
print('CRS:', json.dumps(data.get('crs', 'N/A'), indent=2))
"

Keys in root: ['type', 'name', 'crs', 'features']
CRS: {
  "type": "name",
  "properties": {
    "name": "urn:ogc:def:crs:OGC:1.3:CRS84"
  }
}

cd /work && python3 -c "
import json

with open('tokyo_buildings_legacy.geojson') as f:
    data = json.load(f)

issues = []
for feat in data['features']:
    fid = feat['properties']['feature_id']
    coords = feat['geometry']['coordinates']
    
    for j, ring in enumerate(coords):
        ring_type = 'exterior' if j == 0 else f'interior {j-1}'
        
        # Check: need at least 4 coords to form a triangle + closure
        if len(ring) < 4:
            issues.append((fid, ring_type, f'too few points: {len(ring)}'))
        
        # Check: first == last (closed)
        if ring[0] != ring[-1]:
            issues.append((fid, ring_type, 'not closed'))
        
        # Check for duplicate consecutive points
        for k in range(len(ring) - 1):
            if ring[k] == ring[k+1]:
                issues.append((fid, ring_type, f'consecutive duplicate at index {k}'))
                break
        
        # Check for self-intersection (collinear overlapping edges too)
        # Just check if the ring has all unique points (excluding closure)
        pts = ring[:-1]
        if len(pts) != len(set(tuple(p) for p in pts)):
            issues.append((fid, ring_type, 'duplicate non-consecutive points'))

for fid, rt, issue in issues:
    print(f'  feature_id={fid}, {rt}: {issue}')

if not issues:
    print('No ring-level issues found beyond orientation')
"

cd /work && python3 -c "
import json

with open('tokyo_buildings_fixed.geojson') as f:
    data = json.load(f)

print('=== Top-level structure ===')
print('Keys:', list(data.keys()))
print('type:', data['type'])
print('Features:', len(data['features']))

# Check CRS was preserved (it's legacy non-RFC7946 but should be kept)
print('CRS:', json.dumps(data.get('crs'), indent=2))

# Verify each feature
from shapely.geometry import shape
from shapely.validation import explain_validity

def signed_area(ring):
    return 0.5 * sum(ring[i][0]*ring[i+1][1] - ring[i+1][0]*ring[i][1] for i in range(len(ring)-1))

errors = []
all_valid = True
for i, feat in enumerate(data['features']):
    fid = feat['properties']['feature_id']
    coords = feat['geometry']['coordinates']
    
    # 1. Geometry type
    if feat['geometry']['type'] != 'Polygon':
        errors.append(f'feature_id={fid}: wrong type {feat[\"geometry\"][\"type\"]}')
        continue
    
    # 2. Exterior ring orientation (should be CCW => positive signed area)
    ext_sa = signed_area(coords[0])
    if ext_sa <= 0:
        errors.append(f'feature_id={fid}: exterior is CW (area={ext_sa:.2e})')
        all_valid = False
    
    # 3. Interior ring orientation (should be CW => negative signed area)
    holes_before = 0
    for j, ring in enumerate(coords[1:]):
        int_sa = signed_area(ring)
        if int_sa >= 0:
            errors.append(f'feature_id={fid}: interior {j} is CCW (area={int_sa:.2e})')
            all_valid = False
        holes_before += 1
    
    # 4. Validate with shapely
    poly = shape(feat['geometry'])
    if not poly.is_valid:
        errors.append(f'feature_id={fid}: shapely invalid: {explain_validity(poly)}')
        all_valid = False
    
    # 5. Check first == last
    for j, ring in enumerate(coords):
        if ring[0] != ring[-1]:
            errors.append(f'feature_id={fid}, ring {j}: not closed')
            all_valid = False
    
    # 6. Check ring has >= 4 points
    for j, ring in enumerate(coords):
        if len(ring) < 4:
            errors.append(f'feature_id={fid}, ring {j}: too few points ({len(ring)})')
            all_valid = False

print()
print('=== Validation Results ===')
if errors:
    print(f'{len(errors)} errors:')
    for e in errors:
        print(f'  {e}')
else:
    print('All features valid!')

# Verify attributes untouched
print()
print('=== Attribute check (first 3 features) ===')
for i in range(3):
    print(f'Feature {i}: {json.dumps(data[\"features\"][i][\"properties\"], indent=2, ensure_ascii=False)}')
"

=== Top-level structure ===
Keys: ['type', 'name', 'crs', 'features']
type: FeatureCollection
Features: 100
CRS: {
  "type": "name",
  "properties": {
    "name": "urn:ogc:def:crs:OGC:1.3:CRS84"
  }
}

=== Validation Results ===
All features valid!

=== Attribute check (first 3 features) ===
Feature 0: {
  "feature_id": 1,
  "overture_id": "01302ced-3141-4b64-b630-98019352a9fd",
  "name_primary": "渋谷協和ビル",
  "building_class": "",
  "height": null
}
Feature 1: {
  "feature_id": 2,
  "overture_id": "046ab130-a9a2-4a0e-9142-b7ea36374fdb",
  "name_primary": "丸大ビル",
  "building_class": "",
  "height": null
}
Feature 2: {
  "feature_id": 3,
  "overture_id": "063348bb-5559-4839-bf7d-8c6fca154867",
  "name_primary": "",
  "building_class": "",
  "height": null
}

cd /work && python3 -c "
import json

# Verify attributes are identical between original and fixed
with open('tokyo_buildings_legacy.geojson') as f:
    orig = json.load(f)

with open('tokyo_buildings_fixed.geojson') as f:
    fixed = json.load(f)

# Check all properties are untouched
for o, f_ in zip(orig['features'], fixed['features']):
    if o['properties'] != f_['properties']:
        print(f'ATTRIBUTE MISMATCH: feature_id={o[\"properties\"][\"feature_id\"]}')
        break
    if o['geometry']['type'] != f_['geometry']['type']:
        print(f'TYPE MISMATCH: feature_id={o[\"properties\"][\"feature_id\"]}')
        break
    if len(o['geometry']['coordinates']) != len(f_['geometry']['coordinates']):
        print(f'RING COUNT MISMATCH: feature_id={o[\"properties\"][\"feature_id\"]}')
        break
    # Check same number of points per ring
    for orig_ring, fix_ring in zip(o['geometry']['coordinates'], f_['geometry']['coordinates']):
        if len(orig_ring) != len(fix_ring):
            print(f'POINT COUNT MISMATCH: feature_id={o[\"properties\"][\"feature_id\"]}')
            break
else:
    print('All attributes, geometry types, ring counts, and point counts preserved.')

# Show one hole feature before/after
print()
print('=== Hole feature sample (feature_id=6) before fix ===')
f6_orig = next(f for f in orig['features'] if f['properties']['feature_id'] == 6)
print(json.dumps(f6_orig['geometry']['coordinates'], indent=2))

print()
print('=== Hole feature sample (feature_id=6) after fix ===')
f6_fix = next(f for f in fixed['features'] if f['properties']['feature_id'] == 6)
print(json.dumps(f6_fix['geometry']['coordinates'], indent=2))
"

All attributes, geometry types, ring counts, and point counts preserved.

=== Hole feature sample (feature_id=6) before fix ===
[
  [
    [
      139.7025726,
      35.661211
    ],
    [
      139.7025732,
      35.6612697
    ],
    [
      139.7027844,
      35.6612685
    ],
    [
      139.7027837,
      35.6611824
    ],
    [
      139.7025917,
      35.6611836
    ],
    [
      139.7025916,
      35.6611717
    ],
    [
      139.7024481,
      35.6611726
    ],
    [
      139.7024485,
      35.6612117
    ],
    [
      139.7025726,
      35.661211
    ]
  ],
  [
    [
      139.70261625,
      35.6612158
    ],
    [
      139.70263306500001,
      35.6612158
    ],
    [
      139.70263306500001,
      35.6612207
    ],
    [
      139.70261625,
      35.6612207
    ],
    [
      139.70261625,
      35.6612158
    ]
  ]
]

=== Hole feature sample (feature_id=6) after fix ===
[
  [
    [
      139.7025726,
      35.661211
    ],
    [
      139.7024485,
      35.6612117
    ],
    [
      139.7024481,
      35.6611726
    ],
    [
      139.7025916,
      35.6611717
    ],
    [
      139.7025917,
      35.6611836
    ],
    [
      139.7027837,
      35.6611824
    ],
    [
      139.7027844,
      35.6612685
    ],
    [
      139.7025732,
      35.6612697
    ],
    [
      139.7025726,
      35.661211
    ]
  ],
  [
    [
      139.70261625,
      35.6612158
    ],
    [
      139.70261625,
      35.6612207
    ],
    [
      139.70263306500001,
      35.6612207
    ],
    [
      139.70263306500001,
      35.6612158
    ],
    [
      139.70261625,
      35.6612158
    ]
  ]
]