Add support for converting geojson to csv. Closes #183.

csvkit/convert/__init__.py

@@ -1,13 +1,14 @@
 #!/usr/bin/env python
 
 from csvitself import csv2csv
+from dbase import dbf2csv
 from fixed import fixed2csv
+from geojs import geojson2csv
 from js import json2csv
 from xls import xls2csv
 from xlsx import xlsx2csv
-from dbase import dbf2csv
 
-SUPPORTED_FORMATS = ['fixed', 'xls', 'xlsx', 'csv', 'json', 'dbf']
+SUPPORTED_FORMATS = ['fixed', 'xls', 'xlsx', 'csv', 'json', 'geojson', 'dbf']
 
 def convert(f, format, schema=None, key=None, **kwargs):
     """
@@ -30,6 +31,8 @@ def convert(f, format, schema=None, key=None, **kwargs):
         return xlsx2csv(f, **kwargs)
     elif format == 'json':
         return json2csv(f, key, **kwargs)
+    elif format == 'geojson':
+        return geojson2csv(f, **kwargs)
     elif format == 'csv':
         return csv2csv(f, **kwargs)
     elif format == 'dbf':

csvkit/convert/geojs.py

@@ -0,0 +1,66 @@
+#!/usr/bin/env python
+
+from cStringIO import StringIO
+import json
+
+from csvkit import CSVKitWriter
+
+def geojson2csv(f, key=None, **kwargs):
+    """
+    Convert a GeoJSON document into CSV format.
+    """
+    document = f.read()
+    js = json.loads(document)
+
+    if not isinstance(js, dict):
+        raise TypeError('JSON document is not valid GeoJSON: Root element is not an object.')
+
+    if 'type' not in js:
+        raise TypeError('JSON document is not valid GeoJSON: No top-level "type" key.')
+
+    if js['type'] != 'FeatureCollection':
+        raise TypeError('Only GeoJSON with root FeatureCollection type is supported. Not %s' % js['type']) 
+
+    if 'features' not in js:
+        raise TypeError('JSON document is not a valid FeatureCollection: No top-level "features" key.')
+
+    features = js['features']
+
+    features_parsed = []    # tuples in the format (id, properties, geometry)
+    property_set = set()
+
+    for feature in features:
+        geoid = feature.get('id', None)
+
+        properties = feature.get('properties') or {}
+        property_set.update(properties.keys())
+
+        geometry = json.dumps(feature['geometry'])
+
+        features_parsed.append((geoid, properties, geometry))
+
+    header = ['id']
+    fields = sorted(list(property_set))
+    header.extend(fields)
+    header.append('geojson')
+
+    o = StringIO()
+    writer = CSVKitWriter(o)
+
+    writer.writerow(header)
+
+    for geoid, properties, geometry in features_parsed:
+        row = [geoid]
+
+        for field in fields:
+            row.append(properties.get(field, None))
+
+        row.append(geometry)
+
+        writer.writerow(row)
+
+    output = o.getvalue()
+    o.close()
+
+    return output
+

examples/test_geojson.csv

@@ -0,0 +1,4 @@
+id,prop0,prop1,geojson
+,value0,,"{""type"": ""Point"", ""coordinates"": [102.0, 0.5]}"
+,value0,0.0,"{""type"": ""LineString"", ""coordinates"": [[102.0, 0.0], [103.0, 1.0], [104.0, 0.0], [105.0, 1.0]]}"
+,value0,{u'this': u'that'},"{""type"": ""Polygon"", ""coordinates"": [[[100.0, 0.0], [101.0, 0.0], [101.0, 1.0], [100.0, 1.0], [100.0, 0.0]]]}"

examples/test_geojson.json

@@ -0,0 +1,49 @@
+{
+	"type": "FeatureCollection",
+	"features": [{
+		"type": "Feature",
+		"geometry": {
+			"type": "Point",
+			"coordinates": [102.0, 0.5]
+		},
+		"properties": {
+			"prop0": "value0"
+		}
+	}, {
+		"type": "Feature",
+		"geometry": {
+			"type": "LineString",
+			"coordinates": [
+				[102.0, 0.0],
+				[103.0, 1.0],
+				[104.0, 0.0],
+				[105.0, 1.0]
+			]
+		},
+		"properties": {
+			"prop0": "value0",
+			"prop1": 0.0
+		}
+	}, {
+		"type": "Feature",
+		"geometry": {
+			"type": "Polygon",
+			"coordinates": [
+				[
+					[100.0, 0.0],
+					[101.0, 0.0],
+					[101.0, 1.0],
+					[100.0, 1.0],
+					[100.0, 0.0]
+				]
+			]
+		},
+		"properties": {
+			"prop0": "value0",
+			"prop1": {
+				"this": "that"
+			}
+		}
+	}]
+}
+

tests/test_convert/test_convert.py

@@ -26,6 +26,16 @@ def test_guess_fixed(self):
 
     def test_guess_xls(self):
         self.assertEqual('xls', convert.guess_format('testdata.xls'))
+
+    def test_guess_xlsx(self):
+        self.assertEqual('xlsx', convert.guess_format('testdata.xlsx'))
 
     def test_guess_csv(self):
         self.assertEqual('csv', convert.guess_format('testdata.csv'))
+
+    def test_guess_dbf(self):
+        self.assertEqual('dbf', convert.guess_format('testdata.dbf'))
+
+    def test_guess_json(self):
+        self.assertEqual('json', convert.guess_format('testdata.json'))
+