fix: fix total rows returned when both start_index and page_size are provided

Author: Linchin

google/cloud/bigquery/client.py

@@ -1975,6 +1975,7 @@ def _get_query_results(
         location: Optional[str] = None,
         timeout: TimeoutType = DEFAULT_TIMEOUT,
         page_size: int = 0,
+        start_index: Optional[int] = None,
     ) -> _QueryResults:
         """Get the query results object for a query job.
 
@@ -2002,7 +2003,7 @@ def _get_query_results(
                 A new ``_QueryResults`` instance.
         """
 
-        extra_params: Dict[str, Any] = {"maxResults": page_size}
+        extra_params: Dict[str, Any] = {"maxResults": page_size, "startIndex": start_index}
 
         if timeout is not None:
             if not isinstance(timeout, (int, float)):
@@ -2040,6 +2041,7 @@ def _get_query_results(
             query_params=extra_params,
             timeout=timeout,
         )
+        breakpoint()
         return _QueryResults.from_api_repr(resource)
 
     def job_from_resource(
@@ -4146,6 +4148,7 @@ def _list_rows_from_query_results(
             params["startIndex"] = start_index
 
         params["formatOptions.useInt64Timestamp"] = True
+        # breakpoint()
         row_iterator = RowIterator(
             client=self,
             api_request=functools.partial(self._call_api, retry, timeout=timeout),

google/cloud/bigquery/job/query.py

@@ -1409,6 +1409,7 @@ def _reload_query_results(
         retry: "retries.Retry" = DEFAULT_RETRY,
         timeout: Optional[float] = None,
         page_size: int = 0,
+        start_index: Optional[int] = None,
     ):
         """Refresh the cached query results unless already cached and complete.
 
@@ -1468,6 +1469,7 @@ def _reload_query_results(
             location=self.location,
             timeout=transport_timeout,
             page_size=page_size,
+            start_index=start_index,
         )
 
     def result(  # type: ignore  # (incompatible with supertype)
@@ -1570,6 +1572,9 @@ def result(  # type: ignore  # (incompatible with supertype)
         if page_size is not None:
             reload_query_results_kwargs["page_size"] = page_size
 
+        if start_index is not None:
+            reload_query_results_kwargs["start_index"] = start_index
+
         try:
             retry_do_query = getattr(self, "_retry_do_query", None)
             if retry_do_query is not None:

google/cloud/bigquery/table.py

@@ -1991,12 +1991,14 @@ def _get_next_page_response(self):
             return response
 
         params = self._get_query_params()
+        import copy
+        params_copy = copy.deepcopy(params)
         if self._page_size is not None:
             if self.page_number and "startIndex" in params:
-                del params["startIndex"]
+                del params_copy["startIndex"]
 
         return self.api_request(
-            method=self._HTTP_METHOD, path=self.path, query_params=params
+            method=self._HTTP_METHOD, path=self.path, query_params=params_copy
         )
 
     @property