Merge pull request #311 from nickeubank/add_dissolve

Add dissolve method to GeoDataFrame

geopandas/geodataframe.py

@@ -8,7 +8,7 @@ import os
 import sys
 
 import numpy as np
-from pandas import DataFrame, Series
+from pandas import DataFrame, Series, Index
 from shapely.geometry import mapping, shape
 from shapely.geometry.base import BaseGeometry
 from six import string_types
@@ -441,6 +441,53 @@ class GeoDataFrame(GeoPandasBase, DataFrame):
     plot.__doc__ = plot_dataframe.__doc__
 
 
+    def dissolve(self, by=None, aggfunc='first'):
+        """
+        Dissolve geometries within `groupby` into single observation. 
+
+        Parameters
+        ----------
+        by : string, default None
+            Column whose values define groups to be dissolved
+        aggfunc : function or string, default "first"
+            Aggregation function for manipulation of data associated
+            with each group. Passed to pandas `groupby.agg` method.
+
+        Returns
+        -------
+        GeoDataFrame
+        """
+
+        # Process non-spatial component
+        data = self.drop(labels=self.geometry.name, axis=1).copy()
+        aggregated_data = data.groupby(by=by).agg(aggfunc)
+
+
+        # Process spatial component
+        groupby_plus_geometry_cols = [self.geometry.name]
+        groupby_plus_geometry_cols.append(by)
+        geometry = self[groupby_plus_geometry_cols].copy()
+
+        def merge_geometries(block):
+
+            merged_geom = block.unary_union
+
+            new_index = block.drop(self.geometry.name, axis=1).iloc[0][by]
+            merged_w_index = GeoSeries(merged_geom, index=Index(Series(new_index),name=by), 
+                                          name=self.geometry.name)
+            return merged_w_index
+
+
+        g = geometry.groupby(by=by, group_keys=False).apply(merge_geometries)
+
+        aggregated_geometry = GeoDataFrame(g, 
+                                           index=g.index,
+                                           geometry=self.geometry.name)
+        # Recombine
+        aggregated = aggregated_geometry.join(aggregated_data)
+        aggregated = aggregated.set_geometry(self.geometry.name)
+        return aggregated
+
 def _dataframe_set_geometry(self, col, drop=False, inplace=False, crs=None):
     if inplace:
         raise ValueError("Can't do inplace setting when converting from"

tests/test_dissolve.py

@@ -0,0 +1,64 @@
+from __future__ import absolute_import
+import tempfile
+import shutil
+import numpy as np
+from shapely.geometry import Point
+from geopandas import GeoDataFrame, read_file
+from geopandas.tools import overlay
+from .util import unittest, download_nybb
+from pandas.util.testing import assert_frame_equal
+from pandas import Index
+
+class TestDataFrame(unittest.TestCase):
+
+    def setUp(self):
+
+        nybb_filename, nybb_zip_path = download_nybb()
+        self.polydf = read_file(nybb_zip_path, vfs='zip://' + nybb_filename)
+        self.polydf = self.polydf[['geometry', 'BoroName', 'BoroCode']]
+
+        self.polydf = self.polydf.rename(columns={'geometry':'myshapes'})
+        self.polydf = self.polydf.set_geometry('myshapes')
+
+        self.polydf['manhattan_bronx'] = 5
+        self.polydf.loc[3:4,'manhattan_bronx']=6
+
+        # Merged geometry
+        manhattan_bronx = self.polydf.loc[3:4,]
+        others = self.polydf.loc[0:2,]
+
+        collapsed = [others.geometry.unary_union, manhattan_bronx.geometry.unary_union]
+        merged_shapes = GeoDataFrame({'myshapes': collapsed}, geometry='myshapes', 
+                             index=Index([5,6], name='manhattan_bronx'))
+
+        # Different expected results
+        self.first = merged_shapes.copy()
+        self.first['BoroName'] = ['Staten Island', 'Manhattan']
+        self.first['BoroCode'] = [5, 1]
+
+        self.mean = merged_shapes.copy()
+        self.mean['BoroCode'] = [4,1.5]
+
+
+    def test_geom_dissolve(self):
+        test = self.polydf.dissolve('manhattan_bronx')
+        self.assertTrue(test.geometry.name == 'myshapes')
+        self.assertTrue(test.geom_almost_equals(self.first).all())
+
+    def test_first_dissolve(self):
+        test = self.polydf.dissolve('manhattan_bronx')
+        test = test.drop('myshapes', axis=1)
+        first = self.first.drop('myshapes', axis=1)
+        assert_frame_equal(first, test)
+
+    def test_mean_dissolve(self):
+        test = self.polydf.dissolve('manhattan_bronx', aggfunc='mean')
+        test = test.drop('myshapes', axis=1)
+        mean = self.mean.drop('myshapes', axis=1)
+        assert_frame_equal(mean, test)
+
+        test = self.polydf.dissolve('manhattan_bronx', aggfunc=np.mean)
+        test = test.drop('myshapes', axis=1)
+        assert_frame_equal(mean, test)
+
+