feat: allow tag dict column

closes #130

Author: SlowMo24

CHANGELOG.md

@@ -8,6 +8,10 @@
  - accept shapely Polygon and MultiPolygon for `bpolys` input parameter
  - if a request fails a bash script containing the respective `curl` command is logged (if possible). This allows for easier debugging and sharing of failed requests.
 
+### Changed
+
+ - breaking: geodataframes now contain a `@other_tags` colum containing all OSM tags. This behaviour can be adapted using the `explode_tags` parameter that allows to specify tags that should be in a separate column or to disable the feature completely. The latter will result in a potentially wide but sparse data frame.
+
 ### Removed
 
  - support for python < 3.10

ohsome/response.py

@@ -4,6 +4,7 @@
 """Class for ohsome API response"""
 
 import json
+from typing import Optional
 
 import geopandas as gpd
 import pandas as pd
@@ -22,17 +23,22 @@ def __init__(self, response=None, url=None, params=None):
         self.parameters = params
         self.data = response.json()
 
-    def as_dataframe(self, multi_index=True):
+    def as_dataframe(
+        self, multi_index: Optional[bool] = True, explode_tags: Optional[tuple] = ()
+    ):
         """
         Converts the ohsome response to a pandas.DataFrame or a geopandas.GeoDataFrame if the
         response contains geometries
         :param multi_index: If true returns the dataframe with a multi index
+        :param explode_tags: By default, tags of extracted features are stored in a single dict-column. You can specify
+        a tuple of tags that should be popped from this column. To disable it completely, pass None. Yet, be aware that
+        you may get a large but sparse data frame.
         :return: pandas.DataFrame or geopandas.GeoDataFrame
         """
         if "features" not in self.data.keys():
             return self._as_dataframe(multi_index)
         else:
-            return self._as_geodataframe(multi_index)
+            return self._as_geodataframe(multi_index, explode_tags)
 
     def _as_dataframe(self, multi_index=True):
         """
@@ -67,7 +73,9 @@ def _as_dataframe(self, multi_index=True):
 
         return result_df.sort_index()
 
-    def _as_geodataframe(self, multi_index=True):
+    def _as_geodataframe(
+        self, multi_index: Optional[bool] = True, explode_tags: Optional[tuple] = ()
+    ):
         """
         Converts the ohsome response to a geopandas.GeoDataFrame
         :param multi_index: If true returns the dataframe with a multi index
@@ -78,7 +86,25 @@ def _as_geodataframe(self, multi_index=True):
             return gpd.GeoDataFrame(crs="epsg:4326", columns=["@osmId", "geometry"])
 
         try:
+            if explode_tags is not None:
+                for feature in self.data["features"]:
+                    properties = feature["properties"]
+                    tags = {}
+                    new_properties = {}
+                    for k in properties.keys():
+                        if (
+                            (k.startswith("@"))
+                            or (k == "timestamp")
+                            or (k in explode_tags)
+                        ):
+                            new_properties[k] = properties.get(k)
+                        else:
+                            tags[k] = properties.get(k)
+                    new_properties["@other_tags"] = tags
+                    feature["properties"] = new_properties
+
             features = gpd.GeoDataFrame().from_features(self.data, crs="epsg:4326")
+
         except TypeError:
             raise TypeError(
                 "This result type cannot be converted to a GeoPandas GeoDataFrame object."

ohsome/test/cassettes/test_response/test_extra_tags_argument.yaml

@@ -0,0 +1,98 @@
+interactions:
+- request:
+    body: bboxes=8.7137%2C49.4096%2C8.717%2C49.4119&time=2016-01-01&filter=name%3DKrautturm+and+type%3Away&properties=tags%2Cmetadata
+    headers:
+      Accept:
+      - '*/*'
+      Accept-Encoding:
+      - gzip, deflate
+      Connection:
+      - keep-alive
+      Content-Length:
+      - '123'
+      Content-Type:
+      - application/x-www-form-urlencoded
+      user-agent:
+      - ohsome-py/0.2.0
+    method: POST
+    uri: https://api.ohsome.org/v1/elements/geometry
+  response:
+    body:
+      string: "{\n  \"attribution\" : {\n    \"url\" : \"https://ohsome.org/copyrights\",\n
+        \   \"text\" : \"\xA9 OpenStreetMap contributors\"\n  },\n  \"apiVersion\"
+        : \"1.10.1\",\n  \"type\" : \"FeatureCollection\",\n  \"features\" : [{\n
+        \   \"type\" : \"Feature\",\n    \"geometry\" : {\n      \"type\" : \"Polygon\",\n
+        \     \"coordinates\" : [\n        [\n          [\n            8.7160632,\n
+        \           49.4102899\n          ],\n          [\n            8.7160749,\n
+        \           49.4103121\n          ],\n          [\n            8.7160827,\n
+        \           49.4103235\n          ],\n          [\n            8.7160963,\n
+        \           49.4103374\n          ],\n          [\n            8.716121,\n
+        \           49.4103549\n          ],\n          [\n            8.7161392,\n
+        \           49.4103691\n          ],\n          [\n            8.7161626,\n
+        \           49.4103819\n          ],\n          [\n            8.716186,\n
+        \           49.4103894\n          ],\n          [\n            8.716225,\n
+        \           49.4103952\n          ],\n          [\n            8.7162679,\n
+        \           49.4103926\n          ],\n          [\n            8.7162893,\n
+        \           49.4103882\n          ],\n          [\n            8.7163176,\n
+        \           49.410378\n          ],\n          [\n            8.7163507,\n
+        \           49.4103615\n          ],\n          [\n            8.7163829,\n
+        \           49.4103317\n          ],\n          [\n            8.7163975,\n
+        \           49.4103057\n          ],\n          [\n            8.7164024,\n
+        \           49.4102791\n          ],\n          [\n            8.7164004,\n
+        \           49.4102506\n          ],\n          [\n            8.7163868,\n
+        \           49.4102277\n          ],\n          [\n            8.7162815,\n
+        \           49.4102258\n          ],\n          [\n            8.7162659,\n
+        \           49.4102189\n          ],\n          [\n            8.7162503,\n
+        \           49.4102144\n          ],\n          [\n            8.7162289,\n
+        \           49.4102106\n          ],\n          [\n            8.7162113,\n
+        \           49.4102093\n          ],\n          [\n            8.7161987,\n
+        \           49.4101992\n          ],\n          [\n            8.7161665,\n
+        \           49.4101992\n          ],\n          [\n            8.7161519,\n
+        \           49.4101954\n          ],\n          [\n            8.7161343,\n
+        \           49.4101846\n          ],\n          [\n            8.7161119,\n
+        \           49.4101967\n          ],\n          [\n            8.7160934,\n
+        \           49.4102119\n          ],\n          [\n            8.7160807,\n
+        \           49.4102277\n          ],\n          [\n            8.7160719,\n
+        \           49.4102449\n          ],\n          [\n            8.7160671,\n
+        \           49.4102594\n          ],\n          [\n            8.7160641,\n
+        \           49.4102753\n          ],\n          [\n            8.7160632,\n
+        \           49.4102899\n          ]\n        ]\n      ]\n    },\n    \"properties\"
+        : {\n      \"@changesetId\" : 30687511,\n      \"@lastEdit\" : \"2015-05-01T10:33:22Z\",\n
+        \     \"@osmId\" : \"way/24885641\",\n      \"@osmType\" : \"way\",\n      \"@snapshotTimestamp\"
+        : \"2016-01-01T00:00:00Z\",\n      \"@version\" : 4,\n      \"building\" :
+        \"yes\",\n      \"name\" : \"Krautturm\"\n    }\n  }]\n}\n"
+    headers:
+      Access-Control-Allow-Credentials:
+      - 'true'
+      Access-Control-Allow-Headers:
+      - Origin,Accept,X-Requested-With,Content-Type,Access-Control-Request-Method,Access-Control-Request-Headers,Authorization
+      Access-Control-Allow-Methods:
+      - POST, GET
+      Access-Control-Allow-Origin:
+      - '*'
+      Access-Control-Max-Age:
+      - '3600'
+      Connection:
+      - Keep-Alive
+      Content-Encoding:
+      - gzip
+      Content-Type:
+      - application/geo+json;charset=utf-8
+      Content-disposition:
+      - attachment;filename=ohsome.geojson
+      Date:
+      - Fri, 17 Nov 2023 14:22:26 GMT
+      Keep-Alive:
+      - timeout=5, max=100
+      Server:
+      - Apache
+      Strict-Transport-Security:
+      - max-age=63072000; includeSubdomains;
+      Transfer-Encoding:
+      - chunked
+      vary:
+      - accept-encoding
+    status:
+      code: 200
+      message: ''
+version: 1

ohsome/test/test_response.py

@@ -265,6 +265,28 @@ def test_elements_geometry(base_client):
     assert len(result) == 1
 
 
+@pytest.mark.vcr
+def test_extra_tags_argument(base_client):
+    """
+    Tests whether the result of elements.geometry is converted to a geopandas.GeoDataFrame
+    :return:
+    """
+    bboxes = "8.7137,49.4096,8.717,49.4119"
+    time = "2016-01-01"
+    flter = "name=Krautturm and type:way"
+
+    response = base_client.elements.geometry.post(
+        bboxes=bboxes, time=time, filter=flter, properties="tags,metadata"
+    )
+    result = response.as_dataframe()
+
+    assert "@other_tags" in result.columns
+    assert "@version" in result.columns
+
+    assert result["@other_tags"].to_list() == [{"building": "yes", "name": "Krautturm"}]
+    assert result["@version"].to_list() == [4]
+
+
 @pytest.mark.vcr
 def test_elementsFullHistory_geometry(base_client):
     """