Read xml with CurvePolygon with Geopandas / Fiona

Question

I am trying to use the Dutch BGT Api via https://api.pdok.nl/lv/bgt/download/v1_0/ui/ with the following code:

# Native
import requests
from requests.structures import CaseInsensitiveDict
import json
import time
from zipfile import ZipFile
from io import BytesIO

# Third Party
import pandas as pd
import geopandas as gpd

# %% Function to request the download link through the BGT API
def request_bgt_download_url(base_url, featurelist, fileformat, geofilter):
    api_url = F"{base_url}/lv/bgt/download/v1_0/delta/custom"
    
    #  Headers
    headers = CaseInsensitiveDict()
    headers["accept"] = "application/json"
    headers["Content-Type"] = "application/json"
    
    features = json.dumps(featurelist)
    data = f'{{"featuretypes": {features}, "format": "{fileformat}", "geofilter": "{geofilter}"}}'

    # Request
    response = requests.post(api_url, headers=headers, data=data)
    
    return response.json()['_links']['status']['href']

# %% Get the download zip URL from the download url requested with: request_bgt_download_url
def request_bgt_zip_url(base_url, download_url):
    api_url = F"{base_url}{download_url}"
    
    response = requests.get(api_url)

    print(response.json())
    while response.json()['status'] != 'COMPLETED':
        response = requests.get(api_url)
        print(response.json()['progress'])
        time.sleep(1)
    
    print(response.json())
    return response.json()['_links']['download']['href']

# %% Main
base_url = "https://api.pdok.nl"
featurelist = ["bak","begroeidterreindeel","bord","buurt","functioneelgebied",
               "gebouwinstallatie","installatie","kast","kunstwerkdeel",
               "mast","onbegroeidterreindeel","ondersteunendwaterdeel",
               "ondersteunendwegdeel","ongeclassificeerdobject","openbareruimte",
               "openbareruimtelabel","overbruggingsdeel","overigbouwwerk","overigescheiding",
               "paal","pand","plaatsbepalingspunt","put","scheiding","sensor","spoor",
               "stadsdeel","straatmeubilair","tunneldeel","vegetatieobject","waterdeel",
               "waterinrichtingselement","waterschap","wegdeel","weginrichtingselement",
               "wijk"]
fileformat = "citygml"
geofilter = 'POLYGON((211417.92 475752.4800000001,212390.64000000004 475896.12,212916.48000000004 475818.84,212879.52000000005 475360.2,212950.08000000002 475203.12,212839.2 475065.36,212819.04 474981.36,212819.04 474877.2,212772 474857.04,212792.16 474769.68,212832.48 474705.84,212889.6 474695.76,213010.56000000003 474685.68,213044.16 474611.76,213030.72 474450.48,212637.6 474423.6,212708.16 473956.56,211122.24000000002 473849.04,210453.6 473896.08,210315.84000000003 473970,211417.92 475752.4800000001))'

# Get the ZIP URL
download_url = request_bgt_download_url(base_url, featurelist, fileformat, geofilter)
zip_url = request_bgt_zip_url(base_url, download_url)

# Request the ZIP URL and unpack it in python
zip_response = requests.get(f"{base_url}{zip_url}")
zipfile = ZipFile(BytesIO(zip_response.content))
files = [zipfile.open(file_name) for file_name in zipfile.namelist()]

# Read the xml files with Geopandas
gdf = gpd.read_file(files[4])

The problem is that I get an error with certain XML files, because of the CurvePolygons present in the XML:

Traceback (most recent call last):

  File "<ipython-input-71-beceeaf42900>", line 1, in <module>
    gdf = gpd.read_file(files[4])

  File "C:\ProgramData\Anaconda3\envs\gis\lib\site-packages\geopandas\io\file.py", line 197, in _read_file
    columns = list(features.schema["properties"])

  File "C:\ProgramData\Anaconda3\envs\gis\lib\site-packages\fiona\collection.py", line 200, in schema
    self._schema = self.session.get_schema()

  File "fiona/ogrext.pyx", line 661, in fiona.ogrext.Session.get_schema

  File "fiona/_geometry.pyx", line 80, in fiona._geometry.normalize_geometry_type_code

UnsupportedGeometryTypeError: 10

TypeError 10 refers to (https://github.com/Toblerity/Fiona/blob/master/fiona/_geometry.pyx):

# Unsupported types.
    #8: 'CircularString',
    #9: 'CompoundCurve',
    #10: 'CurvePolygon',
    #11: 'MultiCurve',
    #12: 'MultiSurface',
    #13: 'Curve',
    #14: 'Surface',
    #15: 'PolyhedralSurface',
    #16: 'TIN',
    #17: 'Triangle'

Is there a way to read the CurvePolygons fom the XML files and make them into normal Polygons so they will fit into a Geopandas GeoDataFrame?

Answer 1

Thanks Gabriel De Luca for your comment it pushed me into the right direction.

from osgeo import ogr

# I saved the zipfile first to a local directory and uzipped it
zipfile.extractall(r"C:\MapWithData")

# OGR was used to read the xml files from the directory
ds = driver_cloud.Open(r"C:\MapWithData\bgt_begroeidterreindeel.xml")
dataLayer = ds.GetLayer(0)

# You can then loop over the features from the dataLayer and alter the Geometry
# For now I selected the last feature
num_features = dataLayer.GetFeatureCount()
last_feature = dataLayer.GetFeature(num_features - 1)

# From https://gis.stackexchange.com/questions/256123/how-to-convert-curved-features-into-geojson
polygon_geometry = ogr.CreateGeometryFromWkt(last_feature.GetGeometryRef().ExportToWkt()).GetLinearGeometry()
last_feature.SetGeometryDirectly(polygon_geometry)

# write the feature to geojson
geojson_out = last_feature.ExportToJson()

print(geojson_out)

'{"type": "Feature", "geometry": {"type": "Polygon", "coordinates": [[[212368.186, 475094.89], [212368.152, 475095.028], [212363.707, 475094.019], [212363.6498651207, 475094.0053874058], [212363.5937045143, 475093.9881919697], [212363.53874261674, 475093.96748241776], [212363.48519905654, 475093.94334150845], [212363.43328781423, 475093.91586570436], [212363.3832163438, 475093.88516482705], [212363.33518473068, 475093.8513615355], [212363.28938493883, 475093.814590956], [212363.246, 475093.775], [212363.20546198025, 475093.7330512827], [212363.16766718827, 475093.68861503154], [212363.1327679306, 475093.6418703496], [212363.10090482977, 475093.5930055379], [212363.0722062811, 475093.5422175675], [212363.0467879325, 475093.48971104616], [212363.02475221452, 475093.4356976002], [212363.00618793067, 475093.38039486104], [212362.99116987042, 475093.3240256831], [212362.979758562, 475093.26681722456], [212362.972, 475093.209], [212363.327, 475091.812], [212367.455, 475091.408], [212368.966, 475091.713], [212368.778, 475092.479], [212368.186, 475094.89]]]}, "properties": {"gml_id": "b5451a320-90d2-2d5d-2aae-71af9df6d5de", "creationDate": "2015-02-26", "LV-publicatiedatum": "2016-12-09T07:48:11", "relatieveHoogteligging": 0, "inOnderzoek": false, "eindRegistratie": null, "tijdstipRegistratie": "2016-12-09T07:22:29.000", "namespace": "NL.IMGeo", "lokaalID": "G0150.d1e9c2997cec4060b406057e58461b8d", "bronhouder": "G0150", "bgt-status": "bestaand", "plus-status": "geenWaarde", "class": "groenvoorziening", "begroeidTerreindeelOpTalud": false, "kruinlijnBegroeidTerreindeel": null, "plus-fysiekVoorkomen": "gras- en kruidachtigen", "terminationDate": null}, "id": 23645}'