Python GeoPandas – Loading Polygons from INSPIRE XML to GeoPandas Dataframe

geopandasinspirepythonxml

The German state of Schleswig-Holstein has a geoportal, where they provide a link to their WFS endpoint:

https://service.gdi-sh.de/SH_INSPIREDOWNLOAD_AI_CP_ALKIS?service=WFS&version=2.0.0&request=GetCapabilities

making it possible to access the state's cadastral data. I load some parcels to QGIS canvas using this endpoint, and observe the calls made via the Network Logger:

One of the calls:

6701 GET https://service.gdi-sh.de/SH_INSPIREDOWNLOAD_AI_CP_ALKIS?SERVICE=WFS&REQUEST=GetFeature&VERSION=2.0.0&TYPENAMES=cp:CadastralParcel&COUNT=10000&SRSNAME=urn:ogc:def:crs:EPSG::25832&BBOX=599711.71880523720756173,5915094.58108168374747038,600677.40076130512170494,5916482.59453260432928801,urn:ogc:def:crs:EPSG::25832&NAMESPACES=xmlns(cp,http://inspire.ec.europa.eu/schemas/cp/4.0)&NAMESPACE=xmlns(cp,http://inspire.ec.europa.eu/schemas/cp/4.0)

I am attempting to recreate what QGIS does using GeoPandas: make this call, convert response to Polygons. I try:

import fiona
fiona.supported_drivers["NAS"] = "raw"

import geopandas as gpd
url = "https://service.gdi-sh.de/SH_INSPIREDOWNLOAD_AI_CP_ALKIS?SERVICE=WFS&REQUEST=GetFeature&VERSION=2.0.0&TYPENAMES=cp:CadastralParcel&COUNT=10000&SRSNAME=urn:ogc:def:crs:EPSG::25832&BBOX=599711.71880523720756173,5915094.58108168374747038,600677.40076130512170494,5916482.59453260432928801,urn:ogc:def:crs:EPSG::25832&NAMESPACES=xmlns(cp,http://inspire.ec.europa.eu/schemas/cp/4.0)&NAMESPACE=xmlns(cp,http://inspire.ec.europa.eu/schemas/cp/4.0)"
df = gpd.read_file(url)

The resulting dataframe df does not however contain POLYGONs, only POINTs:

I can download the response manually:

URL="https://service.gdi-sh.de/SH_INSPIREDOWNLOAD_AI_CP_ALKIS?SERVICE=WFS&REQUEST=GetFeature&VERSION=2.0.0&TYPENAMES=cp:CadastralParcel&COUNT=10000&SRSNAME=urn:ogc:def:crs:EPSG::25832&BBOX=599711.71880523720756173,5915094.58108168374747038,600677.40076130512170494,5916482.59453260432928801,urn:ogc:def:crs:EPSG::25832&NAMESPACES=xmlns(cp,http://inspire.ec.europa.eu/schemas/cp/4.0)&NAMESPACE=xmlns(cp,http://inspire.ec.europa.eu/schemas/cp/4.0)"
curl $URL -o inspire.xml

How can I make GeoPandas load the parcel polygons, instead of the reference points, into the dataframe?

This is a similar question, where the accepted answer relies on ogrinfo. I try ogrinfo inspire.xml, get:

ERROR 1: No schema information loaded
INFO: Open of `inspire.xml'
      using driver `NAS' successful.
1: CadastralParcel (Point)
2: ALKIS_beziehungen (None)

i.e. even ogrinfo does not seem to see the Polygons. They are clearly there (see screenshot of XML in browser above), but something is wrong with the schema, I believe.

Apparently this does not happen for everyone, as pointed out in this answer. In my environment, fiona.__version__ returns 1.8.21, gpd.__version__ returns 0.11.1.

Tried the same code in Google Colab (after a !pip3 install geopandas), it works. There, fiona.__version__ returns 1.8.22, gpd.__version__ returns 0.12.2 .

Most probably my problem is due to old GeoPandas or Fiona packages.

Best Answer

I don't understand your problem, when I run your script, I get polygons:

import fiona
fiona.supported_drivers["NAS"] = "raw"
import geopandas as gpd
url = "https://service.gdi-sh.de/SH_INSPIREDOWNLOAD_AI_CP_ALKIS?SERVICE=WFS&REQUEST=GetFeature&VERSION=2.0.0&TYPENAMES=cp:CadastralParcel&COUNT=10000&SRSNAME=urn:ogc:def:crs:EPSG::25832&BBOX=599711.71880523720756173,5915094.58108168374747038,600677.40076130512170494,5916482.59453260432928801,urn:ogc:def:crs:EPSG::25832&NAMESPACES=xmlns(cp,http://inspire.ec.europa.eu/schemas/cp/4.0)&NAMESPACE=xmlns(cp,http://inspire.ec.europa.eu/schemas/cp/4.0)"
df = gpd.read_file(url)
print(df.geometry.head())
0    POLYGON ((599690.313 5915268.949, 599721.202 5...
1    POLYGON ((597116.805 5916472.623, 597248.259 5...
2    POLYGON ((600516.724 5915145.596, 600554.597 5...
3    MULTIPOLYGON (((600598.558 5915149.776, 600599...
4    POLYGON ((600510.934 5915180.078, 600512.871 5...

Related Solutions

[GIS] Python geopandas dataframe of polygons — determine nearest neighbor polygon

You can use the Python rtree library to build up a spatial index, which then has a nearest method you can use to get the nearest geometry in the index to any given query. I think Shapely also comes with an rtree implementation which behaves similarly, but I could be wrong - I always use rtree. This is probably the fastest way, as it only requires one calculation for each record.

Otherwise, you'll need to compare every geometry to every other one using the shapely distance method, and choose the smallest one. I guess in Pandas that would be a full outer join of your dataset to itself, and add a column of distance, and then query out the records where ID's are not equal (implying you calculated the distance to the same polygon) and distance is the shortest.

GeoPandas – How to Read and Import Inspire XML Data with GeoPandas

This is a GML file with layers. To read a layer, pass the layer name:

In [6]: db = gpd.read_file("DB-Netz_INSPIRE_20200217.xml",layer="RailwayLine")
/home/rowlings/.local/lib/python3.8/site-packages/geopandas/geodataframe.py:600: RuntimeWarning: Sequential read of iterator was interrupted. Resetting iterator. This can negatively impact the performance.
  for feature in features_lst:

In [7]: db
Out[7]: 
           gml_id                     identifier beginLifespanVersion      localId  ... grammaticalNumber validFrom railwayLineCode geometry
0     Line-953080  urn:x-dbnetze:oid:Line-953080                 None  Line-953080  ...              None      None            6321     None
1     Line-953081  urn:x-dbnetze:oid:Line-953081                 None  Line-953081  ...              None      None            6406     None
2     Line-953082  urn:x-dbnetze:oid:Line-953082                 None  Line-953082  ...              None      None            2112     None
3     Line-953083  urn:x-dbnetze:oid:Line-953083                 None  Line-953083  ...              None      None            5864     None
4     Line-953084  urn:x-dbnetze:oid:Line-953084                 None  Line-953084  ...              None      None            6270     None
...           ...                            ...                  ...          ...  ...               ...       ...             ...      ...
1504  Line-954584  urn:x-dbnetze:oid:Line-954584                 None  Line-954584  ...              None      None            2666     None
1505  Line-954585  urn:x-dbnetze:oid:Line-954585                 None  Line-954585  ...              None      None            3525     None
1506  Line-954586  urn:x-dbnetze:oid:Line-954586                 None  Line-954586  ...              None      None            2901     None
1507  Line-954587  urn:x-dbnetze:oid:Line-954587                 None  Line-954587  ...              None      None            1923     None
1508  Line-954588  urn:x-dbnetze:oid:Line-954588                 None  Line-954588  ...              None      None            6070     None

I'm not sure how to get all the layer names in python, but ogrinfo at the command line will do this for me:

INFO: Open of `DB-Netz_INSPIRE_20200217.xml'
      using driver `GML' successful.
1: Network (None)
2: ConditionOfFacility (None)
3: MarkerPost (Point)
4: TrafficFlowDirection (None)
5: VerticalPosition (None)
6: DesignSpeed (None)
7: NominalTrackGauge (None)
8: NumberOfTracks (None)
9: RailwayElectrification (None)
10: RailwayLine (None)
11: RailwayLink (Line String)
12: RailwayLinkSequence (None)
13: RailwayNode (Point)
14: RailwayStationNode (Point)
15: RailwayStationCode (None)
16: RailwayType (None)
17: RailwayUse (None)

What you are seeing without a layer option is the first layer, which has no geometry. How the various layers relate will be in the documentation.

Best Answer

Related Solutions

[GIS] Python geopandas dataframe of polygons — determine nearest neighbor polygon

GeoPandas – How to Read and Import Inspire XML Data with GeoPandas

Related Question