python(bug): surface get_data page_size parameter

python/lib/sift_client/_internal/grpc_transport/transport.py

@@ -165,6 +165,10 @@ def _compute_channel_options(opts: SiftChannelConfig) -> list[tuple[str, Any]]:
         # Primary cannot be overriden:
         #  https://github.com/grpc/grpc/blob/0498194240f55d7f4b12633ad01339fb690621bf/src/core/ext/filters/http/client/http_client_filter.cc#L97
         ("grpc.secondary_user_agent", _compute_user_agent()),
+        (
+            "grpc.max_receive_message_length",
+            50 * 1024 * 1024,
+        ),  # 50 MB — headroom for 1M-row GetData pages
     ]
 
     enable_keepalive = opts.get("enable_keepalive", True)

python/lib/sift_client/resources/channels.py

@@ -265,6 +265,7 @@ async def get_data(
         start_time: datetime | None = None,
         end_time: datetime | None = None,
         limit: int | None = None,
+        page_size: int | None = None,
         ignore_cache: bool = False,
         show_progress: bool | None = None,
     ) -> dict[str, pd.DataFrame]:
@@ -276,6 +277,9 @@ async def get_data(
             start_time: The start time to get data for.
             end_time: The end time to get data for.
             limit: The maximum number of data points to return. Will be in increments of page_size or default page size defined by the call if no page_size is provided.
+            page_size: Number of data points to fetch per request. Defaults to 10,000.
+                The server caps this at 1,000,000; values above that are coerced down.
+                Increase toward 1,000,000 to reduce round-trips on large datasets.
             ignore_cache: Whether to ignore cached data and fetch fresh data from the server.
             show_progress: If True, display a progress bar naming each channel as
                 its data is fetched. Defaults to True for sync, False for async.
@@ -296,6 +300,7 @@ async def get_data(
             start_time=start_time,
             end_time=end_time,
             max_results=limit,
+            page_size=page_size,
             ignore_cache=ignore_cache,
             show_progress=show_progress,
         )
@@ -308,6 +313,7 @@ async def get_data_as_arrow(
         start_time: datetime | None = None,
         end_time: datetime | None = None,
         limit: int | None = None,
+        page_size: int | None = None,
         ignore_cache: bool = False,
         show_progress: bool | None = None,
     ) -> dict[str, pa.Table]:
@@ -321,6 +327,7 @@ async def get_data_as_arrow(
             start_time=start_time,
             end_time=end_time,
             limit=limit,
+            page_size=page_size,
             ignore_cache=ignore_cache,
             show_progress=show_progress,
         )

python/lib/sift_client/resources/sync_stubs/__init__.pyi

@@ -494,6 +494,7 @@ class ChannelsAPI:
         start_time: datetime | None = None,
         end_time: datetime | None = None,
         limit: int | None = None,
+        page_size: int | None = None,
         ignore_cache: bool = False,
         show_progress: bool | None = None,
     ) -> dict[str, pd.DataFrame]:
@@ -505,6 +506,9 @@ class ChannelsAPI:
             start_time: The start time to get data for.
             end_time: The end time to get data for.
             limit: The maximum number of data points to return. Will be in increments of page_size or default page size defined by the call if no page_size is provided.
+            page_size: Number of data points to fetch per request. Defaults to 10,000.
+                The server caps this at 1,000,000; values above that are coerced down.
+                Increase toward 1,000,000 to reduce round-trips on large datasets.
             ignore_cache: Whether to ignore cached data and fetch fresh data from the server.
             show_progress: If True, display a progress bar naming each channel as
                 its data is fetched. Defaults to True for sync, False for async.
@@ -523,6 +527,7 @@ class ChannelsAPI:
         start_time: datetime | None = None,
         end_time: datetime | None = None,
         limit: int | None = None,
+        page_size: int | None = None,
         ignore_cache: bool = False,
         show_progress: bool | None = None,
     ) -> dict[str, pa.Table]: