next_index -> next_start_index

docs/_static/llama-stack-spec.html

@@ -8160,7 +8160,7 @@
                         },
                         "description": "The rows in the current page."
                     },
-                    "next_index": {
+                    "next_start_index": {
                         "type": "integer",
                         "description": "Index into dataset for the first row in the next page. None if there are no more rows."
                     }

docs/_static/llama-stack-spec.yaml

@@ -5557,7 +5557,7 @@ components:
                 - type: array
                 - type: object
           description: The rows in the current page.
-        next_index:
+        next_start_index:
           type: integer
           description: >-
             Index into dataset for the first row in the next page. None if there are

llama_stack/apis/datasetio/datasetio.py

@@ -18,11 +18,11 @@ class IterrowsResponse(BaseModel):
     A paginated list of rows from a dataset.
 
     :param data: The rows in the current page.
-    :param next_index: Index into dataset for the first row in the next page. None if there are no more rows.
+    :param next_start_index: Index into dataset for the first row in the next page. None if there are no more rows.
     """
 
     data: List[Dict[str, Any]]
-    next_index: Optional[int] = None
+    next_start_index: Optional[int] = None
 
 
 class DatasetStore(Protocol):
@@ -46,9 +46,11 @@ class DatasetIO(Protocol):
 
         :param dataset_id: The ID of the dataset to get the rows from.
         :param start_index: Index into dataset for the first row to get. Get all rows if None.
-        :param limit: The number of rows to get per page.
+        :param limit: The number of rows to get.
         """
         ...
 
     @webmethod(route="/datasetio/append-rows/{dataset_id:path}", method="POST")
-    async def append_rows(self, dataset_id: str, rows: List[Dict[str, Any]]) -> None: ...
+    async def append_rows(
+        self, dataset_id: str, rows: List[Dict[str, Any]]
+    ) -> None: ...

llama_stack/providers/inline/datasetio/localfs/datasetio.py

@@ -44,7 +44,9 @@ class PandasDataframeDataset:
         elif self.dataset_def.source.type == "rows":
             self.df = pandas.DataFrame(self.dataset_def.source.rows)
         else:
-            raise ValueError(f"Unsupported dataset source type: {self.dataset_def.source.type}")
+            raise ValueError(
+                f"Unsupported dataset source type: {self.dataset_def.source.type}"
+            )
 
         if self.df is None:
             raise ValueError(f"Failed to load dataset from {self.dataset_def.url}")
@@ -108,7 +110,7 @@ class LocalFSDatasetIOImpl(DatasetIO, DatasetsProtocolPrivate):
 
         return IterrowsResponse(
             data=rows,
-            next_index=end if end < len(dataset_impl) else None,
+            next_start_index=end if end < len(dataset_impl) else None,
         )
 
     async def append_rows(self, dataset_id: str, rows: List[Dict[str, Any]]) -> None:
@@ -117,4 +119,6 @@ class LocalFSDatasetIOImpl(DatasetIO, DatasetsProtocolPrivate):
         dataset_impl.load()
 
         new_rows_df = pandas.DataFrame(rows)
-        dataset_impl.df = pandas.concat([dataset_impl.df, new_rows_df], ignore_index=True)
+        dataset_impl.df = pandas.concat(
+            [dataset_impl.df, new_rows_df], ignore_index=True
+        )

llama_stack/providers/remote/datasetio/huggingface/huggingface.py

@@ -86,7 +86,7 @@ class HuggingfaceDatasetIOImpl(DatasetIO, DatasetsProtocolPrivate):
 
         return IterrowsResponse(
             data=rows,
-            next_index=end if end < len(loaded_dataset) else None,
+            next_start_index=end if end < len(loaded_dataset) else None,
         )
 
     async def append_rows(self, dataset_id: str, rows: List[Dict[str, Any]]) -> None:
@@ -98,9 +98,13 @@ class HuggingfaceDatasetIOImpl(DatasetIO, DatasetsProtocolPrivate):
         new_dataset = hf_datasets.Dataset.from_list(rows)
 
         # Concatenate the new rows with existing dataset
-        updated_dataset = hf_datasets.concatenate_datasets([loaded_dataset, new_dataset])
+        updated_dataset = hf_datasets.concatenate_datasets(
+            [loaded_dataset, new_dataset]
+        )
 
         if dataset_def.metadata.get("path", None):
             updated_dataset.push_to_hub(dataset_def.metadata["path"])
         else:
-            raise NotImplementedError("Uploading to URL-based datasets is not supported yet")
+            raise NotImplementedError(
+                "Uploading to URL-based datasets is not supported yet"
+            )