bigquery: add Client.list_rows, remove Table.fetch_data (#4119)

Author: jba

bigquery/google/cloud/bigquery/client.py

@@ -26,14 +26,16 @@
 from google.cloud.bigquery._http import Connection
 from google.cloud.bigquery.dataset import Dataset
 from google.cloud.bigquery.dataset import DatasetReference
-from google.cloud.bigquery.table import Table
+from google.cloud.bigquery.table import Table, _TABLE_HAS_NO_SCHEMA
 from google.cloud.bigquery.table import TableReference
 from google.cloud.bigquery.job import CopyJob
 from google.cloud.bigquery.job import ExtractJob
 from google.cloud.bigquery.job import LoadJob
 from google.cloud.bigquery.job import QueryJob
 from google.cloud.bigquery.job import QueryJobConfig
 from google.cloud.bigquery.query import QueryResults
+from google.cloud.bigquery._helpers import _item_to_row
+from google.cloud.bigquery._helpers import _rows_page_start
 
 
 class Project(object):
@@ -346,7 +348,6 @@ def delete_table(self, table):
         :type table: One of:
                      :class:`~google.cloud.bigquery.table.Table`
                      :class:`~google.cloud.bigquery.table.TableReference`
-
         :param table: the table to delete, or a reference to it.
         """
         if not isinstance(table, (Table, TableReference)):
@@ -667,6 +668,80 @@ def query_rows(self, query, job_config=None, job_id=None, timeout=None):
         job.begin()
         return job.result(timeout=timeout)
 
+    def list_rows(self, table, selected_fields=None, max_results=None,
+                  page_token=None, start_index=None):
+        """List the rows of the table.
+
+        See
+        https://cloud.google.com/bigquery/docs/reference/rest/v2/tabledata/list
+
+        .. note::
+
+           This method assumes that the provided schema is up-to-date with the
+           schema as defined on the back-end: if the two schemas are not
+           identical, the values returned may be incomplete. To ensure that the
+           local copy of the schema is up-to-date, call ``client.get_table``.
+
+        :type table: One of:
+                     :class:`~google.cloud.bigquery.table.Table`
+                     :class:`~google.cloud.bigquery.table.TableReference`
+        :param table: the table to list, or a reference to it.
+
+        :type selected_fields: list of :class:`SchemaField`
+        :param selected_fields:
+            The fields to return. Required if ``table`` is a
+            :class:`~google.cloud.bigquery.table.TableReference`.
+
+        :type max_results: int
+        :param max_results: maximum number of rows to return.
+
+        :type page_token: str
+        :param page_token: (Optional) Token representing a cursor into the
+                           table's rows.
+
+        :type start_index: int
+        :param page_token: (Optional) The zero-based index of the starting
+                           row to read.
+
+        :rtype: :class:`~google.api.core.page_iterator.Iterator`
+        :returns: Iterator of row data :class:`tuple`s. During each page, the
+                  iterator will have the ``total_rows`` attribute set,
+                  which counts the total number of rows **in the table**
+                  (this is distinct from the total number of rows in the
+                  current page: ``iterator.page.num_items``).
+
+        """
+        if selected_fields is not None:
+            schema = selected_fields
+        elif isinstance(table, TableReference):
+            raise ValueError('need selected_fields with TableReference')
+        elif isinstance(table, Table):
+            if len(table._schema) == 0:
+                raise ValueError(_TABLE_HAS_NO_SCHEMA)
+            schema = table.schema
+        else:
+            raise TypeError('table should be Table or TableReference')
+
+        params = {}
+        if selected_fields is not None:
+            params['selectedFields'] = [f.name for f in selected_fields]
+        if start_index is not None:
+            params['startIndex'] = start_index
+
+        iterator = page_iterator.HTTPIterator(
+            client=self,
+            api_request=self._connection.api_request,
+            path='%s/data' % (table.path,),
+            item_to_value=_item_to_row,
+            items_key='rows',
+            page_token=page_token,
+            next_token='pageToken',
+            max_results=max_results,
+            page_start=_rows_page_start,
+            extra_params=params)
+        iterator.schema = schema
+        return iterator
+
 
 # pylint: disable=unused-argument
 def _item_to_project(iterator, resource):

bigquery/google/cloud/bigquery/table.py

@@ -25,13 +25,10 @@
 from google.resumable_media.requests import MultipartUpload
 from google.resumable_media.requests import ResumableUpload
 
-from google.api.core import page_iterator
 from google.cloud import exceptions
 from google.cloud._helpers import _datetime_from_microseconds
 from google.cloud._helpers import _millis_from_datetime
 from google.cloud.bigquery.schema import SchemaField
-from google.cloud.bigquery._helpers import _item_to_row
-from google.cloud.bigquery._helpers import _rows_page_start
 from google.cloud.bigquery._helpers import _SCALAR_VALUE_TO_JSON_ROW
 
 
@@ -768,61 +765,6 @@ def update(self, client=None):
             method='PUT', path=self.path, data=self._build_resource())
         self._set_properties(api_response)
 
-    def fetch_data(self, max_results=None, page_token=None, client=None):
-        """API call:  fetch the table data via a GET request
-
-        See
-        https://cloud.google.com/bigquery/docs/reference/rest/v2/tabledata/list
-
-        .. note::
-
-           This method assumes that its instance's ``schema`` attribute is
-           up-to-date with the schema as defined on the back-end:  if the
-           two schemas are not identical, the values returned may be
-           incomplete.  To ensure that the local copy of the schema is
-           up-to-date, call ``client.get_table``.
-
-        :type max_results: int
-        :param max_results: (Optional) Maximum number of rows to return.
-
-        :type page_token: str
-        :param page_token: (Optional) Token representing a cursor into the
-                           table's rows.
-
-        :type client: :class:`~google.cloud.bigquery.client.Client`
-        :param client: (Optional) The client to use.  If not passed, falls
-                       back to the ``client`` stored on the current dataset.
-
-        :rtype: :class:`~google.api.core.page_iterator.Iterator`
-        :returns: Iterator of row data :class:`tuple`s. During each page, the
-                  iterator will have the ``total_rows`` attribute set,
-                  which counts the total number of rows **in the table**
-                  (this is distinct from the total number of rows in the
-                  current page: ``iterator.page.num_items``).
-        """
-        if len(self._schema) == 0:
-            raise ValueError(_TABLE_HAS_NO_SCHEMA)
-
-        params = {}
-
-        if max_results is not None:
-            params['maxResults'] = max_results
-
-        client = self._require_client(client)
-        path = '%s/data' % (self.path,)
-        iterator = page_iterator.HTTPIterator(
-            client=client,
-            api_request=client._connection.api_request,
-            path=path,
-            item_to_value=_item_to_row,
-            items_key='rows',
-            page_token=page_token,
-            page_start=_rows_page_start,
-            next_token='pageToken',
-            extra_params=params)
-        iterator.schema = self._schema
-        return iterator
-
     def row_from_mapping(self, mapping):
         """Convert a mapping to a row tuple using the schema.
 

bigquery/tests/system.py

@@ -295,7 +295,7 @@ def test_update_table(self):
 
     @staticmethod
     def _fetch_single_page(table):
-        iterator = table.fetch_data()
+        iterator = Config.CLIENT.list_rows(table)
         page = six.next(iterator.pages)
         return list(page)