fio-l1-vienna-shapefile-recovery

Rerun

import geopandas as gpd
import pandas as pd
import json

def migrate_parcels(shp_path, map_csv_path, output_geojson):
    # 1. Load the column map
    # Based on the Read output, it looks like it's tab-separated or space-separated?
    # Let's try reading it with sep=None to auto-detect, but it looks like whitespace or tab.
    # Actually, looking at the output: "     1\ttruncated,original"
    # It seems to be tab separated for the index, then comma for the columns? 
    # Or maybe it's a CSV with some weird formatting.
    # Let's re-examine the Read output.
    # "     1	truncated,original"
    # This looks like: [index][tab][truncated],[original]
    
    mapping_df = pd.read_csv(map_csv_path, sep='\t', header=None, names=['id', 'cols'])
    # Now mapping_df['cols'] contains 'truncated,original'
    # We need to split 'cols' by ',' and create a dictionary.
    
    rename_map = {}
    for _, row in mapping_df.iterrows():
        parts = row['cols'].split(',')
        if len(parts) == 2:
            old_name = parts[0].strip()
            new_name = parts[1].strip()
            rename_map[old_name] = new_name
            
    print(f"Mapping found: {rename_map}")

    # 2. Load the shapefile
    gdf = gpd.read_file(shp_path)
    print(f"Original columns: {gdf.columns.tolist()}")

    # 3. Rename columns
    # We only want the columns that are in our map. 
    # Actually, the instructions say "with the correct full column names".
    # It's safer to rename what we can and keep the rest, or only keep mapped ones?
    # Usually, migration implies keeping all data but fixing names.
    # However, if "truncated" refers to the short names in the SHP, we rename.
    
    gdf = gdf.rename(columns=rename_map)
    
    # Filter to only include the renamed columns and the geometry? 
    # Or just rename everything possible. Let's keep all for now.
    # But let's check if the renamed columns are actually there.
    
    print(f"New columns: {gdf.columns.tolist()}")

    # 4. Ensure correct geometry type (Polygon/MultiPolygon)
    # Geopandas usually handles this, but let's ensure we don't have points.
    # The user said "The features should come through as Polygon or MultiPolygon".
    # We should probably filter out anything that isn't a polygon just in case.
    gdf = gdf[gdf.geometry.type.isin(['Polygon', 'MultiPolygon'])]

    # 5. Export to GeoJSON
    # Ensure proper encoding (UTF-8 is default for GeoJSON)
    gdf.to_file(output_geojson, driver='GeoJSON')
    print(f"Successfully saved to {output_geojson}")

if __name__ == "__main__":
    migrate_parcels('parcels.shp', 'column_map.csv', 'parcels.geojson')

import geopandas as gpd
import pandas as pd

def migrate_parcels(shp_path, map_csv_path, output_geojson):
    # The column_map.csv file has a very specific structure.
    # Let's read it line by line to avoid pandas' automatic parsing issues with the header-like line.
    
    rename_map = {}
    with open(map_csv_path, 'r', encoding='utf-8') as f:
        lines = f.readlines()
        for line in lines:
            # Skip the first line "1 truncated,original"
            if 'truncated,original' in line:
                continue
            
            # The lines look like "     2	KATASTRALG,KATASTRALGEMEINDE_NAME"
            # They are separated by a tab (or whitespace) and then a comma.
            parts = line.strip().split('\t')
            if len(parts) >= 2:
                cols_part = parts[1]
                col_mapping = cols_part.split(',')
                if len(col_mapping) == 2:
                    old_name = col_mapping[0].strip()
                    new_name = col_mapping[1].strip()
                    rename_map[old_name] = new_name
    
    print(f"Mapping found: {rename_map}")

    # 2. Load the shapefile
    gdf = gpd.read_file(shp_path)
    print(f"Original columns: {gdf.columns.tolist()}")

    # 3. Rename columns
    # We only want to keep the columns that are in our map and the geometry.
    # Or we can rename everything in the map and then potentially filter.
    # The request says "with the correct full column names".
    # This implies the output should have these names.
    
    gdf = gdf.rename(columns=rename_map)
    
    # Let's verify if columns were actually renamed.
    # If they weren't, it might be because of casing or whitespace.
    # Let's check the actual columns in gdf.
    
    # 4. Ensure correct geometry type
    gdf = gdf[gdf.geometry.type.isin(['Polygon', 'MultiPolygon'])]

    # 5. Export to GeoJSON
    gdf.to_file(output_geojson, driver='GeoJSON')
    print(f"Successfully saved to {output_geojson}")

if __name__ == "__main__":
    migrate_parcels('parcels.shp', 'column_map.csv', 'parcels.geojson')

import geopandas as gpd
import pandas as pd

def migrate_parcels(shp_path, map_csv_path, output_geojson):
    rename_map = {}
    with open(map_csv_path, 'r', encoding='utf-8') as f:
        lines = f.readlines()
        for line in lines:
            line = line.strip()
            if not line or 'truncated,original' in line:
                continue
            
            # The format is [ID][TAB][TRUNCATED],[ORIGINAL]
            # Let's split by tab first
            parts = line.split('\t')
            if len(parts) >= 2:
                col_mapping = parts[1].split(',')
                if len(col_mapping) == 2:
                    old_name = col_mapping[0].strip()
                    new_name = col_mapping[1].strip()
                    rename_map[old_name] = new_name
            else:
                # If tab split fails, try whitespace split (though tab is expected)
                # parts[0] is ID, parts[1] is col_mapping
                parts_ws = line.split(None, 1)
                if len(parts_ws) == 2:
                    col_mapping = parts_ws[1].split(',')
                    if len(col_mapping) == 2:
                        old_name = col_mapping[0].strip()
                        new_name = col_mapping[1].strip()
                        rename_map[old_name] = new_name
    
    print(f"Mapping found: {rename_map}")

    gdf = gpd.read_file(shp_path)
    print(f"Original columns: {gdf.columns.tolist()}")

    # Rename columns
    gdf = gdf.rename(columns=rename_map)
    print(f"New columns: {gdf.columns.tolist()}")

    # Ensure correct geometry type
    gdf = gdf[gdf.geometry.type.isin(['Polygon', 'MultiPolygon'])]

    # Export to GeoJSON
    gdf.to_file(output_geojson, driver='GeoJSON')
    print(f"Successfully saved to {output_geojson}")

if __name__ == "__main__":
    migrate_parcels('parcels.shp', 'column_map.csv', 'parcels.geojson')

import geopandas as gpd
import pandas as pd

def migrate_parcels(shp_path, map_csv_path, output_geojson):
    rename_map = {}
    with open(map_csv_path, 'r', encoding='utf-8') as f:
        lines = f.readlines()
        for line in lines:
            line = line.strip()
            if not line or 'truncated,original' in line:
                continue
            
            print(f"Processing line: '{line}'")
            # The line is: "2\tKATASTRALG,KATASTRALGEMEINDE_NAME"
            # Actually, looking at the raw output of Read:
            # "     2\tKATASTRALG,KATASTRALGEMEINDE_NAME"
            # It might have multiple spaces or tabs.
            
            # Let's try a more robust approach. 
            # Split by the first comma that is preceded by text.
            # Or just split by tab/whitespace and then handle the second part.
            
            parts = line.split()
            # parts[0] = '2', parts[1] = 'KATASTRALG,KATASTRALGEMEINDE_NAME'
            if len(parts) >= 2:
                col_mapping = parts[1].split(',')
                if len(col_mapping) == 2:
                    old_name = col_mapping[0].strip()
                    new_name = col_mapping[1].strip()
                    rename_map[old_name] = new_name
                else:
                    print(f"  Failed to split col_mapping in parts: {parts[1]}")
            else:
                print(f"  Failed to split line into parts: {line}")
    
    print(f"Final Mapping: {rename_map}")

    gdf = gpd.read_file(shp_path)
    print(f"Original columns: {gdf.columns.tolist()}")

    # Rename columns
    gdf = gdf.rename(columns=rename_map)
    print(f"New columns: {gdf.columns.tolist()}")

    # Ensure correct geometry type
    gdf = gdf[gdf.geometry.type.isin(['Polygon', 'MultiPolygon'])]

    # Export to GeoJSON
    gdf.to_file(output_geojson, driver='GeoJSON')
    print(f"Successfully saved to {output_geojson}")

if __name__ == "__main__":
    migrate_parcels('parcels.shp', 'column_map.csv', 'parcels.geojson')

import geopandas as gpd
import pandas as pd

def migrate_parcels(shp_path, map_csv_path, output_geojson):
    rename_map = {}
    with open(map_csv_path, 'r', encoding='utf-8') as f:
        for line in f:
            line = line.strip()
            if not line or 'truncated,original' in line:
                continue
            
            # Line is like "2	KATASTRALG,KATASTRALGEMEINDE_NAME"
            # It seems there is a tab between the ID and the mapping.
            # But if it's not working, let's try splitting by everything.
            
            # If there's a comma, the last part of the line (after the first whitespace) 
            # should contain the mapping.
            
            # Let's find the first comma and look backwards for the mapping part.
            # Or just split by comma and then clean up the first part.
            
            parts = line.split(',')
            if len(parts) == 2:
                # parts[0] is something like "2\tKATASTRALG"
                # parts[1] is "KATASTRALGEMEINDE_NAME"
                
                # Let's split parts[0] by whitespace to get the truncated name.
                prefix_parts = parts[0].split()
                if prefix_parts:
                    old_name = prefix_parts[-1]
                    new_name = parts[1].strip()
                    rename_map[old_name] = new_name

    print(f"Final Mapping: {rename_map}")

    gdf = gpd.read_file(shp_path)
    print(f"Original columns: {gdf.columns.tolist()}")

    # Rename columns
    gdf = gdf.rename(columns=rename_map)
    print(f"New columns: {gdf.columns.tolist()}")

    # Ensure correct geometry type
    gdf = gdf[gdf.geometry.type.isin(['Polygon', 'MultiPolygon'])]

    # Export to GeoJSON
    gdf.to_file(output_geojson, driver='GeoJSON')
    print(f"Successfully saved to {output_geojson}")

if __name__ == "__main__":
    migrate_parcels('parcels.shp', 'column_map.csv', 'parcels.geojson')