Integration with geopandas #588

.vscode/settings.json

@@ -0,0 +1,3 @@
+{
+    "python.pythonPath": "/Users/tim/anaconda/anaconda/envs/altair_test/bin/python"
+}

.vscode/tasks.json

@@ -0,0 +1,15 @@
+{
+    // See https://go.microsoft.com/fwlink/?LinkId=733558
+    // for the documentation about the tasks.json format
+    "version": "2.0.0",
+    "tasks": [
+        {
+            "label": "pip instal",
+            "type": "shell",
+            "command": "pip install -e.",
+            "problemMatcher": [
+                "$go"
+            ]
+        }
+    ]
+}

altair/utils/core.py

@@ -227,9 +227,13 @@ def parse_shorthand(shorthand, data=None):
     {'aggregate': 'count', 'type': 'quantitative'}
     """
     attrs = _parse_shorthand(shorthand)
-    if isinstance(data, pd.DataFrame) and 'type' not in attrs:
+    if isinstance(data, pd.DataFrame):
         if 'field' in attrs and attrs['field'] in data.columns:
-            attrs['type'] = infer_vegalite_type(data[attrs['field']])
+                if 'type' not in attrs:
+                    attrs['type'] = infer_vegalite_type(data[attrs['field']])
+                if hasattr(data,'__geo_interface__'): #TODO: Add descripion in shorthand
+                    attrs['field']='properties.'+attrs['field']
+
     return attrs
 
 

altair/utils/data.py

@@ -33,7 +33,7 @@ class DataTransformerRegistry(PluginRegistry[DataTransformerType]):
 # form.
 #
 # A data model transformer has the following type signature:
-# DataModelType = Union[dict, pd.DataFrame]
+# DataModelType = Union[dict, pd.DataFrame, gpd.GeoDataFrame, geojson interface object]
 # DataModelTransformerType = Callable[[DataModelType, KwArgs], DataModelType]
 # ==============================================================================
 
@@ -57,6 +57,8 @@ def limit_rows(data, max_rows=5000):
             values = data['values']
         else:
             return data
+    else:
+        return data
     if max_rows is not None and len(values) > max_rows:
         raise MaxRowsError('The number of rows in your dataset is greater '
                            'than the maximum allowed ({0}). '
@@ -85,18 +87,29 @@ def to_json(data, prefix='altair-data'):
     check_data_type(data)
     ext = '.json'
     filename = _compute_filename(prefix=prefix, ext=ext)
+    data_format = {'type': 'json'}
     if isinstance(data, pd.DataFrame):
         data = sanitize_dataframe(data)
-        data.to_json(filename, orient='records')
+        if not hasattr(data,'__geo_interface__'):
+            data.to_json(filename, orient='records')
+        else: #GeoPandas
+            with open(filename,'w') as f:
+                json.dump(data.__geo_interface__, f)
+            data_format['property']='features'
+
+    elif hasattr(data,'__geo_interface__'): # geojson object
+        with open(filename,'w') as f:
+            json.dump(data.__geo_interface__, f)
+
     elif isinstance(data, dict):
         if 'values' not in data:
             raise KeyError('values expected in data dict, but not present.')
         values = data['values']
-        with open(filename) as f:
+        with open(filename,'w') as f:
             json.dump(values, f)
     return {
         'url': filename,
-        'format': {'type': 'json'}
+        'format': data_format
     }
 
 
@@ -107,12 +120,17 @@ def to_csv(data, prefix='altair-data'):
     ext = '.csv'
     filename = _compute_filename(prefix=prefix, ext=ext)
     if isinstance(data, pd.DataFrame):
+        if hasattr(data,'__geo_interface__'):#GeoPandas
+                raise NotImplementedError('use to_json or to_values with GeoDataFrame objects.')
+
         data = sanitize_dataframe(data)
         data.to_csv(filename)
         return {
             'url': filename,
             'format': {'type': 'csv'}
         }
+    elif hasattr(data,'__geo_interface__'):#GeoJSON
+            raise NotImplementedError('to_csv only works with Pandas DataFrame objects.')
     elif isinstance(data, dict):
         raise NotImplementedError('to_csv only works with Pandas DataFrame objects.')
 
@@ -123,7 +141,19 @@ def to_values(data):
     check_data_type(data)
     if isinstance(data, pd.DataFrame):
         data = sanitize_dataframe(data)
+        if hasattr(data,'__geo_interface__'):#GeoPandas
+            return {
+                    'values':data.__geo_interface__,
+                    'format':{'type':'json','property':'features'}
+                    }
+
         return {'values': data.to_dict(orient='records')}
+    elif hasattr(data,'__geo_interface__'):#GeoJSON
+            return {
+                    'values':data.__geo_interface__,
+                    'format':{'type':'json'}
+                    }
+
     elif isinstance(data, dict):
         if 'values' not in data:
             raise KeyError('values expected in data dict, but not present.')
@@ -132,8 +162,8 @@ def to_values(data):
 
 def check_data_type(data):
     """Raise if the data is not a dict or DataFrame."""
-    if not isinstance(data, (dict, pd.DataFrame)):
-        raise TypeError('Expected dict or DataFrame, got: {}'.format(type(data)))
+    if not (isinstance(data, (dict, pd.DataFrame)) or hasattr(data,'__geo_interface__')):
+        raise TypeError('Expected dict, DataFrame, GeoDataFrame or geojson inteface object , got: {}'.format(type(data)))
 
 
 # ==============================================================================

altair/utils/tests/test_data.py

@@ -59,7 +59,7 @@ def test_to_values():
 
 
 def test_type_error():
-    """Ensure that TypeError is raised for types other than dict/DataFrame."""
+    """Ensure that TypeError is raised for types other than dict/DataFrame/GeoDataFrame/__geo_interface__."""
     for f in (sample, limit_rows, to_values):
         with pytest.raises(TypeError):
             pipe(0, f)

altair/utils/tests/test_geojson.py

@@ -0,0 +1,153 @@
+import pytest
+import pandas as pd
+import altair.vegalite.v2 as alt
+
+from ..data import  pipe, to_values, to_csv
+from .. import parse_shorthand
+
+
+def _create_geojson():
+    return {
+                "type": "FeatureCollection",
+                "bbox": [
+                    -161.30174569731454,
+                    -60.39157788643298,
+                    172.67580002536624,
+                    42.438347020953984
+                ],
+                "features": [
+                    {
+                    "type": "Feature",
+                    "properties": {"prop": 1},
+                    "geometry": {
+                        "type": "LineString",
+                        "coordinates": [
+                        [-69.2980008004234, 23.18780298146116],
+                        [-161.30174569731454, -60.39157788643298],
+                        [172.67580002536624, 24.151450472748962]
+                        ]
+                    },
+                    "id": "0",
+                    "bbox": [
+                        -161.30174569731454,
+                        -60.39157788643298,
+                        172.67580002536624,
+                        24.151450472748962
+                    ]
+                    },
+                    {
+                    "type": "Feature",
+                    "properties": {"prop": 2},
+                    "geometry": {
+                        "type": "LineString",
+                        "coordinates": [
+                        [156.03047546751765, 42.438347020953984],
+                        [35.46296546950265, -18.185542212943375],
+                        [152.53211600051463, 23.471406463455793]
+                        ]
+                    },
+                    "id": "1",
+                    "bbox": [
+                        35.46296546950265,
+                        -18.185542212943375,
+                        156.03047546751765,
+                        42.438347020953984
+                    ]
+                    },
+                    {
+                    "type": "Feature",
+                    "properties": {"prop": 3},
+                    "geometry": {
+                        "type": "LineString",
+                        "coordinates": [
+                        [-133.98414913936503, 25.39468871174894],
+                        [145.04376601680605, 13.058626381790845],
+                        [170.30576801294046, 38.67128737163435]
+                        ]
+                    },
+                    "id": "2",
+                    "bbox": [
+                        -133.98414913936503,
+                        13.058626381790845,
+                        170.30576801294046,
+                        38.67128737163435
+                    ]
+                    }
+                ]
+            }
+
+def _create_fake_geo_interface():
+    class FakeGeoJSON:
+        __geo_interface__=_create_geojson()
+    return FakeGeoJSON()
+
+def _create_fake_geodatafarme():
+    class FakeGeoDataFrame(pd.DataFrame):
+        __geo_interface__=_create_geojson()
+        def copy(self, deep=True):
+            data = self._data
+            if deep:
+                data = data.copy()
+            return FakeGeoDataFrame(data).__finalize__(self)
+
+    return FakeGeoDataFrame({'prop':[1,2,3]})
+
+def test_to_values_geo():
+    """Test the to_values data transformer."""
+
+    data = _create_fake_geodatafarme()
+    result = pipe(data, to_values)
+    assert result['format'] == {'type':'json','property':'features'}
+    assert result['values']==data.__geo_interface__
+
+    data = _create_fake_geo_interface()
+    result = pipe(data, to_values)
+    assert result['format'] == {'type':'json'}
+    assert result['values']==data.__geo_interface__
+
+def test_chart_data_geotypes():
+    Chart = lambda data,**arg: alt.Chart(data).mark_geoshape().project().encode(**arg)
+
+    # Fake GeoPandas
+    data = _create_fake_geodatafarme()
+    dct = Chart(data,fill='prop').to_dict() 
+    assert dct['data']['format'] == {'type':'json','property':'features'}
+    assert dct['data']['values'] == data.__geo_interface__
+
+      # Fake GeoInteface
+    data = _create_fake_geo_interface()
+    dct = Chart(data).to_dict() 
+    assert dct['data']['format'] == {'type':'json'}
+    assert dct['data']['values'] == data.__geo_interface__
+
+def test_parse_shorthand_with_geodata():
+    def check(s, data, **kwargs):
+        assert parse_shorthand(s, data) == kwargs
+
+    data = _create_fake_geodatafarme()
+
+    check('prop', data, field='properties.prop', type='quantitative')
+    check('prop:N', data, field='properties.prop', type='nominal')
+    check('count(prop)', data, field='properties.prop', aggregate='count', type='quantitative')
+
+    data = _create_fake_geo_interface()
+
+    check('properties.prop:Q', data, field='properties.prop', type='quantitative')
+    check('prop', data, field='prop')
+
+def test_to_csv_geo():
+    """Test the to_csv raise error with geopandas."""
+
+    data = _create_fake_geodatafarme()
+    with pytest.raises(NotImplementedError):
+        pipe(data, to_csv)
+
+def test_geo_pandas(): 
+    gpd = pytest.importorskip('geopandas')
+
+    data = gpd.GeoDataFrame.from_features(_create_geojson())
+    dct = alt.Chart(data).mark_geoshape().project().encode(fill='prop').to_dict()
+
+    assert dct['data']['format'] == {'type':'json','property':'features'}
+    assert (gpd.GeoDataFrame.from_features(dct['data']['values']) == data).all().all()
+    assert dct['encoding'] == {'fill': {'field': 'properties.prop', 'type': 'quantitative'}}

altair/vegalite/v2/api.py

@@ -19,11 +19,11 @@ def _prepare_data(data):
     """Convert input data to data for use within schema"""
     if data is Undefined:
         return data
+    elif isinstance(data, pd.DataFrame) or hasattr(data, '__geo_interface__'):
+        return pipe(data, data_transformers.get())
     elif isinstance(data, (dict, core.Data, core.InlineData,
                          core.UrlData, core.NamedData)):
         return data
-    elif isinstance(data, pd.DataFrame):
-        return pipe(data, data_transformers.get())
     elif isinstance(data, six.string_types):
         return core.UrlData(data)
     else: