voxcity 0.6.26__py3-none-any.whl → 0.7.0__py3-none-any.whl

voxcity/geoprocessor/io.py

@@ -0,0 +1,101 @@
+"""
+I/O helpers for reading/writing vector data (GPKG, gzipped GeoJSON lines) and
+saving FeatureCollections.
+"""
+
+import copy
+import gzip
+import json
+from typing import List
+
+import geopandas as gpd
+
+from .conversion import filter_and_convert_gdf_to_geojson
+
+
+def get_geojson_from_gpkg(gpkg_path, rectangle_vertices):
+    """
+    Read a GeoPackage file and convert it to GeoJSON features within a bounding rectangle.
+    """
+    print(f"Opening GPKG file: {gpkg_path}")
+    gdf = gpd.read_file(gpkg_path)
+    geojson = filter_and_convert_gdf_to_geojson(gdf, rectangle_vertices)
+    return geojson
+
+
+def get_gdf_from_gpkg(gpkg_path, rectangle_vertices):
+    """
+    Read a GeoPackage file and convert it to a GeoDataFrame with consistent CRS.
+
+    Note: rectangle_vertices is currently unused but kept for signature compatibility.
+    """
+    print(f"Opening GPKG file: {gpkg_path}")
+    gdf = gpd.read_file(gpkg_path)
+
+    if gdf.crs is None:
+        gdf.set_crs(epsg=4326, inplace=True)
+    elif gdf.crs != "EPSG:4326":
+        gdf = gdf.to_crs(epsg=4326)
+
+    gdf['id'] = gdf.index
+    return gdf
+
+
+def load_gdf_from_multiple_gz(file_paths):
+    """
+    Load GeoJSON features from multiple gzipped files into a single GeoDataFrame.
+    Each line in each file must be a single GeoJSON Feature.
+    """
+    geojson_objects = []
+
+    for gz_file_path in file_paths:
+        with gzip.open(gz_file_path, 'rt', encoding='utf-8') as file:
+            for line in file:
+                try:
+                    data = json.loads(line)
+                    if 'properties' in data and 'height' in data['properties']:
+                        if data['properties']['height'] is None:
+                            data['properties']['height'] = 0
+                    else:
+                        if 'properties' not in data:
+                            data['properties'] = {}
+                        data['properties']['height'] = 0
+                    geojson_objects.append(data)
+                except json.JSONDecodeError as e:
+                    print(f"Skipping line in {gz_file_path} due to JSONDecodeError: {e}")
+
+    gdf = gpd.GeoDataFrame.from_features(geojson_objects)
+    gdf.set_crs(epsg=4326, inplace=True)
+    return gdf
+
+
+def swap_coordinates(features):
+    """
+    Swap coordinate ordering in GeoJSON features from (lat, lon) to (lon, lat).
+    Modifies the input features in-place.
+    """
+    for feature in features:
+        if feature['geometry']['type'] == 'Polygon':
+            new_coords = [[[lon, lat] for lat, lon in polygon] for polygon in feature['geometry']['coordinates']]
+            feature['geometry']['coordinates'] = new_coords
+        elif feature['geometry']['type'] == 'MultiPolygon':
+            new_coords = [[[[lon, lat] for lat, lon in polygon] for polygon in multipolygon] for multipolygon in feature['geometry']['coordinates']]
+            feature['geometry']['coordinates'] = new_coords
+
+
+def save_geojson(features, save_path):
+    """
+    Save GeoJSON features to a file with coordinate swapping and pretty printing.
+    """
+    geojson_features = copy.deepcopy(features)
+    swap_coordinates(geojson_features)
+
+    geojson = {
+        "type": "FeatureCollection",
+        "features": geojson_features
+    }
+
+    with open(save_path, 'w') as f:
+        json.dump(geojson, f, indent=2)
+
+

voxcity/geoprocessor/merge_utils.py

@@ -0,0 +1,91 @@
+"""
+Utilities to merge GeoDataFrames while resolving ID conflicts.
+"""
+
+import pandas as pd
+
+
+def _merge_gdfs_with_missing_columns(gdf_1, gdf_2):
+    """
+    Helper to merge two GeoDataFrames while handling missing columns by filling with None.
+    """
+    columns_1 = set(gdf_1.columns)
+    columns_2 = set(gdf_2.columns)
+
+    only_in_1 = columns_1 - columns_2
+    only_in_2 = columns_2 - columns_1
+
+    for col in only_in_2:
+        gdf_1[col] = None
+    for col in only_in_1:
+        gdf_2[col] = None
+
+    all_columns = sorted(list(columns_1.union(columns_2)))
+    gdf_1 = gdf_1[all_columns]
+    gdf_2 = gdf_2[all_columns]
+
+    merged_gdf = pd.concat([gdf_1, gdf_2], ignore_index=True)
+    return merged_gdf
+
+
+def merge_gdfs_with_id_conflict_resolution(gdf_1, gdf_2, id_columns=['id', 'building_id']):
+    """
+    Merge two GeoDataFrames while resolving ID conflicts by modifying IDs in the second GeoDataFrame.
+    """
+    gdf_primary = gdf_1.copy()
+    gdf_secondary = gdf_2.copy()
+
+    missing_columns = []
+    for col in id_columns:
+        if col not in gdf_primary.columns:
+            missing_columns.append(f"'{col}' missing from gdf_1")
+        if col not in gdf_secondary.columns:
+            missing_columns.append(f"'{col}' missing from gdf_2")
+
+    if missing_columns:
+        print(f"Warning: Missing ID columns: {', '.join(missing_columns)}")
+        id_columns = [col for col in id_columns if col in gdf_primary.columns and col in gdf_secondary.columns]
+
+    if not id_columns:
+        print("Warning: No valid ID columns found. Merging without ID conflict resolution.")
+        merged_gdf = _merge_gdfs_with_missing_columns(gdf_primary, gdf_secondary)
+        return merged_gdf
+
+    max_ids = {}
+    for col in id_columns:
+        if gdf_primary[col].dtype in ['int64', 'int32', 'float64', 'float32']:
+            max_ids[col] = gdf_primary[col].max()
+        else:
+            max_ids[col] = len(gdf_primary)
+
+    next_ids = {col: max_ids[col] + 1 for col in id_columns}
+    modified_buildings = 0
+
+    for idx, row in gdf_secondary.iterrows():
+        needs_new_ids = False
+        for col in id_columns:
+            current_id = row[col]
+            if current_id in gdf_primary[col].values:
+                needs_new_ids = True
+                break
+        if needs_new_ids:
+            modified_buildings += 1
+            for col in id_columns:
+                new_id = next_ids[col]
+                gdf_secondary.at[idx, col] = new_id
+                next_ids[col] += 1
+
+    merged_gdf = _merge_gdfs_with_missing_columns(gdf_primary, gdf_secondary)
+
+    total_buildings = len(merged_gdf)
+    primary_buildings = len(gdf_primary)
+    secondary_buildings = len(gdf_secondary)
+
+    print(f"Merged {primary_buildings} buildings from primary dataset with {secondary_buildings} buildings from secondary dataset.")
+    print(f"Total buildings in merged dataset: {total_buildings}")
+    if modified_buildings > 0:
+        print(f"Modified IDs for {modified_buildings} buildings in secondary dataset to resolve conflicts.")
+
+    return merged_gdf
+
+