Convert NaN values to None in numpy arrays.

geodimm · geodimm · commit 8bf57cb84f85 · 2015-06-25T23:12:00.000+01:00
This allows the default JSON encoder to do serialize panda dataframes
diff --git a/influxdb/influxdb08/dataframe_client.py b/influxdb/influxdb08/dataframe_client.py
@@ -15,18 +15,17 @@ class DataFrameClient(InfluxDBClient):
     The client reads and writes from pandas DataFrames.
     """
 
-    def __init__(self, *args, **kwargs):
+    def __init__(self, ignore_nan=True, *args, **kwargs):
         super(DataFrameClient, self).__init__(*args, **kwargs)
+
         try:
             global pd
             import pandas as pd
         except ImportError as ex:
-            raise ImportError(
-                'DataFrameClient requires Pandas, "{ex}" problem importing'
-                .format(ex=str(ex))
-            )
-
+            raise ImportError('DataFrameClient requires Pandas, '
+                              '"{ex}" problem importing'.format(ex=str(ex)))
         self.EPOCH = pd.Timestamp('1970-01-01 00:00:00.000+00:00')
+        self.ignore_nan = ignore_nan
 
     def write_points(self, data, *args, **kwargs):
         """
@@ -135,9 +134,24 @@ def _convert_dataframe_to_json(self, dataframe, name, time_precision='s'):
                              for dt in dataframe.index]
         data = {'name': name,
                 'columns': [str(column) for column in dataframe.columns],
-                'points': list([list(x) for x in dataframe.values])}
+                'points': [self._convert_array(x) for x in dataframe.values]}
         return data
 
+    def _convert_array(self, array):
+        try:
+            global np
+            import numpy as np
+        except ImportError as ex:
+            raise ImportError('DataFrameClient requires Numpy, '
+                              '"{ex}" problem importing'.format(ex=str(ex)))
+        if self.ignore_nan:
+            number_types = (int, float, np.number)
+            condition = (all(isinstance(el, number_types) for el in array) and
+                         np.isnan(array))
+            return list(np.where(condition, None, array))
+        else:
+            return list(array)
+
     def _datetime_to_epoch(self, datetime, time_precision='s'):
         seconds = (datetime - self.EPOCH).total_seconds()
         if time_precision == 's':
diff --git a/tests/influxdb/influxdb08/dataframe_client_test.py b/tests/influxdb/influxdb08/dataframe_client_test.py
@@ -52,6 +52,32 @@ def test_write_points_from_dataframe(self):
 
             self.assertListEqual(json.loads(m.last_request.body), points)
 
+    def test_write_points_from_dataframe_with_float_nan(self):
+        now = pd.Timestamp('1970-01-01 00:00+00:00')
+        dataframe = pd.DataFrame(data=[[1, float("NaN"), 1.0], [2, 2, 2.0]],
+                                 index=[now, now + timedelta(hours=1)],
+                                 columns=["column_one", "column_two",
+                                          "column_three"])
+        points = [
+            {
+                "points": [
+                    [1, None, 1.0, 0],
+                    [2, 2, 2.0, 3600]
+                ],
+                "name": "foo",
+                "columns": ["column_one", "column_two", "column_three", "time"]
+            }
+        ]
+
+        with requests_mock.Mocker() as m:
+            m.register_uri(requests_mock.POST,
+                           "http://localhost:8086/db/db/series")
+
+            cli = DataFrameClient(database='db')
+            cli.write_points({"foo": dataframe})
+
+            self.assertListEqual(json.loads(m.last_request.body), points)
+
     def test_write_points_from_dataframe_in_batches(self):
         now = pd.Timestamp('1970-01-01 00:00+00:00')
         dataframe = pd.DataFrame(data=[["1", 1, 1.0], ["2", 2, 2.0]],
