Making updates per jreback's requested changes

Author: mr-mcox

docs/source/changelog.rst

@@ -6,6 +6,8 @@ Changelog
 
 - Resolve issue where the optional ``--noauth_local_webserver`` command line argument would not be propagated during the authentication process. (:issue:`35`)
 - Drop support for Python 3.4 (:issue:`40`)
+- When using ```to_gbq``` if ```if_exists``` is set to ```append```, dataframe needs to contain only a subset of the fields in the BigQuery schema. (:issue: `24`)
+
 
 0.1.6 / 2017-05-03
 ------------------
@@ -14,10 +16,6 @@ Changelog
 
 0.1.5 / 2017-04-20
 ------------------
-- When using ```to_gbq``` if ```if_exists``` is set to ```append```, dataframe needs to contain only a subset of the fields in the BigQuery schema. GH#24
-
-0.1.4 / 2017-03-17
-------------------
 
 - ``InvalidIndexColumn`` will be raised instead of ``InvalidColumnOrder`` in ``read_gbq`` when the index column specified does not exist in the BigQuery schema. (:issue:`6`)
 

pandas_gbq/gbq.py

@@ -563,11 +563,17 @@ def schema(self, dataset_id, table_id):
         Obtain from BigQuery the field names and field types
         for the table defined by the parameters
 
-        :param str dataset_id: Name of the BigQuery dataset for the table
-        :param str table_id: Name of the BigQuery table
+        Parameters
+        ----------
+        dataset_id : str
+            Name of the BigQuery dataset for the table
+        table_id : str
+            Name of the BigQuery table
 
-        :return: Fields representing the schema
-        :rtype: list of dicts
+        Returns
+        -------
+        list of dicts
+            Fields representing the schema
         """
 
         try:
@@ -596,13 +602,20 @@ def verify_schema(self, dataset_id, table_id, schema):
         the schema passed in and indicate whether all fields in the former
         are present in the latter. Order is not considered.
 
-        :param str dataset_id: Name of the BigQuery dataset for the table
-        :param str table_id: Name of the BigQuery table
-        :param list(dict) schema: Schema for comparison. Each item should have
+        Parameters
+        ----------
+        dataset_id :str
+            Name of the BigQuery dataset for the table
+        table_id : str
+            Name of the BigQuery table
+        schema : list(dict)
+            Schema for comparison. Each item should have
             a 'name' and a 'type'
 
-        :return: Whether the schemas match
-        :rtype: bool
+        Returns
+        -------
+        bool
+            Whether the schemas match
         """
 
         fields_remote = sorted(self.schema(dataset_id, table_id),
@@ -618,13 +631,20 @@ def schema_is_subset(self, dataset_id, table_id, schema):
         the schema passed in and indicate whether a subset of the fields in
         the former are present in the latter. Order is not considered.
 
-        :param str dataset_id: Name of the BigQuery dataset for the table
-        :param str table_id: Name of the BigQuery table
-        :param list(dict) schema: Schema for comparison. Each item should have
+        Parameters
+        ----------
+        dataset_id : str
+            Name of the BigQuery dataset for the table
+        table_id : str
+            Name of the BigQuery table
+        schema : list(dict)  
+            Schema for comparison. Each item should have
             a 'name' and a 'type'
 
-        :return: Whether the passed schema is a subset
-        :rtype: bool
+        Returns
+        -------
+        bool
+            Whether the passed schema is a subset
         """
 
         fields_remote = self.schema(dataset_id, table_id)

pandas_gbq/tests/test_gbq.py

@@ -1072,7 +1072,8 @@ def test_upload_data_if_table_exists_append(self):
                        private_key=_get_private_key_path())
 
     def test_upload_subset_columns_if_table_exists_append(self):
-        # For pull request #24
+        # Issue 24: Upload is succesful if dataframe has columns
+        # which are a subset of the current schema
         test_id = "16"
         test_size = 10
         df = make_mixed_dataframe_v2(test_size)
@@ -1283,7 +1284,7 @@ def test_verify_schema_ignores_field_mode(self):
             self.dataset_prefix + "1", TABLE_ID + test_id, test_schema_2)
 
     def test_retrieve_schema(self):
-        # For pull request #24
+        # Issue #24 schema function returns the schema in biquery
         test_id = "15"
         test_schema = {'fields': [{'name': 'A', 'type': 'FLOAT'},
                                   {'name': 'B', 'type': 'FLOAT'},
@@ -1296,7 +1297,8 @@ def test_retrieve_schema(self):
         assert expected == actual, 'Expected schema used to create table'
 
     def test_schema_is_subset_passes_if_subset(self):
-        # For pull request #24
+        # Issue #24 schema_is_subset indicates whether the schema of the
+        # dataframe is a subset of the schema of the bigquery table
         test_id = '16'
 
         table_name = TABLE_ID + test_id