Merge 0d65bbc9e4 into sapling-pr-archive-ehhuang

client-sdks/stainless/openapi.yml

@@ -2916,11 +2916,11 @@ paths:
       responses:
         '200':
           description: >-
-            A list of InterleavedContent representing the file contents.
+            Represents the parsed content of a vector store file.
           content:
             application/json:
               schema:
-                $ref: '#/components/schemas/VectorStoreFileContentsResponse'
+                $ref: '#/components/schemas/VectorStoreFileContentResponse'
         '400':
           $ref: '#/components/responses/BadRequest400'
         '429':
@@ -10463,41 +10463,36 @@ components:
       title: VectorStoreContent
       description: >-
         Content item from a vector store file or search result.
-    VectorStoreFileContentsResponse:
+    VectorStoreFileContentResponse:
       type: object
+      description: Represents the parsed content of a vector store file.
       properties:
-        file_id:
+        object:
           type: string
-          description: Unique identifier for the file
+          enum:
-        filename:
+            - vector_store.file_content.page
-          type: string
+          description: The object type, which is always `vector_store.file_content.page`
-          description: Name of the file
+          x-stainless-const: true
-        attributes:
+        data:
-          type: object
-          additionalProperties:
-            oneOf:
-              - type: 'null'
-              - type: boolean
-              - type: number
-              - type: string
-              - type: array
-              - type: object
-          description: >-
-            Key-value attributes associated with the file
-        content:
           type: array
+          description: Parsed content of the file.
           items:
             $ref: '#/components/schemas/VectorStoreContent'
-          description: List of content items from the file
+        has_more:
+          type: boolean
+          description: Indicates if there are more content pages to fetch.
+        next_page:
+          anyOf:
+            - type: string
+              description: The token for the next page, if any.
+            - type: 'null'
       additionalProperties: false
       required:
-        - file_id
+        - object
-        - filename
+        - data
-        - attributes
+        - has_more
-        - content
+        - next_page
-      title: VectorStoreFileContentsResponse
+      title: VectorStoreFileContentResponse
-      description: >-
-        Response from retrieving the contents of a vector store file.
     OpenaiSearchVectorStoreRequest:
       type: object
       properties:

src/llama_stack/apis/vector_io/vector_io.py

@@ -396,19 +396,19 @@ class VectorStoreListFilesResponse(BaseModel):
 
 
 @json_schema_type
-class VectorStoreFileContentsResponse(BaseModel):
+class VectorStoreFileContentResponse(BaseModel):
-    """Response from retrieving the contents of a vector store file.
+    """Represents the parsed content of a vector store file.
 
-    :param file_id: Unique identifier for the file
+    :param object: The object type, which is always `vector_store.file_content.page`
-    :param filename: Name of the file
+    :param data: Parsed content of the file
-    :param attributes: Key-value attributes associated with the file
+    :param has_more: Indicates if there are more content pages to fetch
-    :param content: List of content items from the file
+    :param next_page: The token for the next page, if any
     """
 
-    file_id: str
+    object: Literal["vector_store.file_content.page"] = "vector_store.file_content.page"
-    filename: str
+    data: list[VectorStoreContent]
-    attributes: dict[str, Any]
+    has_more: bool
-    content: list[VectorStoreContent]
+    next_page: str | None = None
 
 
 @json_schema_type
@@ -732,12 +732,12 @@ class VectorIO(Protocol):
         self,
         vector_store_id: str,
         file_id: str,
-    ) -> VectorStoreFileContentsResponse:
+    ) -> VectorStoreFileContentResponse:
         """Retrieves the contents of a vector store file.
 
         :param vector_store_id: The ID of the vector store containing the file to retrieve.
         :param file_id: The ID of the file to retrieve.
-        :returns: A list of InterleavedContent representing the file contents.
+        :returns: A VectorStoreFileContentResponse representing the file contents.
         """
         ...
 

src/llama_stack/core/routers/vector_io.py

@@ -24,7 +24,7 @@ from llama_stack.apis.vector_io import (
     VectorStoreChunkingStrategyStaticConfig,
     VectorStoreDeleteResponse,
     VectorStoreFileBatchObject,
-    VectorStoreFileContentsResponse,
+    VectorStoreFileContentResponse,
     VectorStoreFileDeleteResponse,
     VectorStoreFileObject,
     VectorStoreFilesListInBatchResponse,
@@ -338,7 +338,7 @@ class VectorIORouter(VectorIO):
         self,
         vector_store_id: str,
         file_id: str,
-    ) -> VectorStoreFileContentsResponse:
+    ) -> VectorStoreFileContentResponse:
         logger.debug(f"VectorIORouter.openai_retrieve_vector_store_file_contents: {vector_store_id}, {file_id}")
         provider = await self.routing_table.get_provider_impl(vector_store_id)
         return await provider.openai_retrieve_vector_store_file_contents(

src/llama_stack/core/routing_tables/vector_stores.py

@@ -15,7 +15,7 @@ from llama_stack.apis.vector_io.vector_io import (
     SearchRankingOptions,
     VectorStoreChunkingStrategy,
     VectorStoreDeleteResponse,
-    VectorStoreFileContentsResponse,
+    VectorStoreFileContentResponse,
     VectorStoreFileDeleteResponse,
     VectorStoreFileObject,
     VectorStoreFileStatus,
@@ -195,7 +195,7 @@ class VectorStoresRoutingTable(CommonRoutingTableImpl):
         self,
         vector_store_id: str,
         file_id: str,
-    ) -> VectorStoreFileContentsResponse:
+    ) -> VectorStoreFileContentResponse:
         await self.assert_action_allowed("read", "vector_store", vector_store_id)
         provider = await self.get_provider_impl(vector_store_id)
         return await provider.openai_retrieve_vector_store_file_contents(

src/llama_stack/providers/utils/memory/openai_vector_store_mixin.py

@@ -30,7 +30,7 @@ from llama_stack.apis.vector_io import (
     VectorStoreContent,
     VectorStoreDeleteResponse,
     VectorStoreFileBatchObject,
-    VectorStoreFileContentsResponse,
+    VectorStoreFileContentResponse,
     VectorStoreFileCounts,
     VectorStoreFileDeleteResponse,
     VectorStoreFileLastError,
@@ -921,22 +921,21 @@ class OpenAIVectorStoreMixin(ABC):
         self,
         vector_store_id: str,
         file_id: str,
-    ) -> VectorStoreFileContentsResponse:
+    ) -> VectorStoreFileContentResponse:
         """Retrieves the contents of a vector store file."""
         if vector_store_id not in self.openai_vector_stores:
             raise VectorStoreNotFoundError(vector_store_id)
 
-        file_info = await self._load_openai_vector_store_file(vector_store_id, file_id)
         dict_chunks = await self._load_openai_vector_store_file_contents(vector_store_id, file_id)
         chunks = [Chunk.model_validate(c) for c in dict_chunks]
         content = []
         for chunk in chunks:
             content.extend(self._chunk_to_vector_store_content(chunk))
-        return VectorStoreFileContentsResponse(
+        return VectorStoreFileContentResponse(
-            file_id=file_id,
+            object="vector_store.file_content.page",
-            filename=file_info.get("filename", ""),
+            data=content,
-            attributes=file_info.get("attributes", {}),
+            has_more=False,
-            content=content,
+            next_page=None,
         )
 
     async def openai_update_vector_store_file(

tests/integration/vector_io/test_openai_vector_stores.py

@@ -907,16 +907,16 @@ def test_openai_vector_store_retrieve_file_contents(
     )
 
     assert file_contents is not None
-    assert len(file_contents.content) == 1
+    assert file_contents.object == "vector_store.file_content.page"
-    content = file_contents.content[0]
+    assert len(file_contents.data) == 1
+    content = file_contents.data[0]
 
     # llama-stack-client returns a model, openai-python is a badboy and returns a dict
     if not isinstance(content, dict):
         content = content.model_dump()
     assert content["type"] == "text"
     assert content["text"] == test_content.decode("utf-8")
-    assert file_contents.filename == file_name
+    assert file_contents.has_more is False
-    assert file_contents.attributes == attributes
 
 
 @vector_provider_wrapper
@@ -1483,14 +1483,12 @@ def test_openai_vector_store_file_batch_retrieve_contents(
         )
 
         assert file_contents is not None
-        assert file_contents.filename == file_data[i][0]
+        assert file_contents.object == "vector_store.file_content.page"
-        assert len(file_contents.content) > 0
+        assert len(file_contents.data) > 0
 
         # Verify the content matches what we uploaded
         content_text = (
-            file_contents.content[0].text
+            file_contents.data[0].text if hasattr(file_contents.data[0], "text") else file_contents.data[0]["text"]
-            if hasattr(file_contents.content[0], "text")
-            else file_contents.content[0]["text"]
         )
         assert file_data[i][1].decode("utf-8") in content_text