feat: Adding option to return embeddings and metadata from /vector_stores/*/files/*/content and UI updates

Author: stainless-app[bot]

.stats.yml

@@ -1,4 +1,4 @@
 configured_endpoints: 89
-openapi_spec_url: https://storage.googleapis.com/stainless-sdk-openapi-specs/llamastack%2Fllama-stack-client-0ace6fde1fda4dcf15bb1177074f61b15ffa1e574127f1d99c570f2a5fae04e9.yml
-openapi_spec_hash: 620ae49556af9e59880cfcf033058def
+openapi_spec_url: https://storage.googleapis.com/stainless-sdk-openapi-specs/llamastack%2Fllama-stack-client-af20fa1866f461e9fef4f7fd226d757b0dddee907e2a083fa582ac0580735e20.yml
+openapi_spec_hash: 68caf264f8ade02c34456c526d7300b1
 config_hash: e8a35d9d37cb4774b4b0fe1b167dc156

api.md

@@ -255,7 +255,7 @@ Methods:
 - <code title="post /v1/vector_stores/{vector_store_id}/files/{file_id}">client.vector_stores.files.<a href="./src/llama_stack_client/resources/vector_stores/files.py">update</a>(file_id, \*, vector_store_id, \*\*<a href="src/llama_stack_client/types/vector_stores/file_update_params.py">params</a>) -> <a href="./src/llama_stack_client/types/vector_stores/vector_store_file.py">VectorStoreFile</a></code>
 - <code title="get /v1/vector_stores/{vector_store_id}/files">client.vector_stores.files.<a href="./src/llama_stack_client/resources/vector_stores/files.py">list</a>(vector_store_id, \*\*<a href="src/llama_stack_client/types/vector_stores/file_list_params.py">params</a>) -> <a href="./src/llama_stack_client/types/vector_stores/vector_store_file.py">SyncOpenAICursorPage[VectorStoreFile]</a></code>
 - <code title="delete /v1/vector_stores/{vector_store_id}/files/{file_id}">client.vector_stores.files.<a href="./src/llama_stack_client/resources/vector_stores/files.py">delete</a>(file_id, \*, vector_store_id) -> <a href="./src/llama_stack_client/types/vector_stores/file_delete_response.py">FileDeleteResponse</a></code>
-- <code title="get /v1/vector_stores/{vector_store_id}/files/{file_id}/content">client.vector_stores.files.<a href="./src/llama_stack_client/resources/vector_stores/files.py">content</a>(file_id, \*, vector_store_id) -> <a href="./src/llama_stack_client/types/vector_stores/file_content_response.py">FileContentResponse</a></code>
+- <code title="get /v1/vector_stores/{vector_store_id}/files/{file_id}/content">client.vector_stores.files.<a href="./src/llama_stack_client/resources/vector_stores/files.py">content</a>(file_id, \*, vector_store_id, \*\*<a href="src/llama_stack_client/types/vector_stores/file_content_params.py">params</a>) -> <a href="./src/llama_stack_client/types/vector_stores/file_content_response.py">FileContentResponse</a></code>
 
 ## FileBatches
 

src/llama_stack_client/resources/vector_stores/files.py

@@ -25,7 +25,7 @@
 )
 from ...pagination import SyncOpenAICursorPage, AsyncOpenAICursorPage
 from ..._base_client import AsyncPaginator, make_request_options
-from ...types.vector_stores import file_list_params, file_create_params, file_update_params
+from ...types.vector_stores import file_list_params, file_create_params, file_update_params, file_content_params
 from ...types.vector_stores.vector_store_file import VectorStoreFile
 from ...types.vector_stores.file_delete_response import FileDeleteResponse
 from ...types.vector_stores.file_content_response import FileContentResponse
@@ -286,6 +286,8 @@ def content(
         file_id: str,
         *,
         vector_store_id: str,
+        include_embeddings: bool | Omit = omit,
+        include_metadata: bool | Omit = omit,
         # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
         # The extra values given here take precedence over values defined on the client or passed to this method.
         extra_headers: Headers | None = None,
@@ -297,6 +299,10 @@ def content(
         Retrieves the contents of a vector store file.
 
         Args:
+          include_embeddings: Whether to include embedding vectors in the response.
+
+          include_metadata: Whether to include chunk metadata in the response.
+
           extra_headers: Send extra headers
 
           extra_query: Add additional query parameters to the request
@@ -312,7 +318,17 @@ def content(
         return self._get(
             f"/v1/vector_stores/{vector_store_id}/files/{file_id}/content",
             options=make_request_options(
-                extra_headers=extra_headers, extra_query=extra_query, extra_body=extra_body, timeout=timeout
+                extra_headers=extra_headers,
+                extra_query=extra_query,
+                extra_body=extra_body,
+                timeout=timeout,
+                query=maybe_transform(
+                    {
+                        "include_embeddings": include_embeddings,
+                        "include_metadata": include_metadata,
+                    },
+                    file_content_params.FileContentParams,
+                ),
             ),
             cast_to=FileContentResponse,
         )
@@ -571,6 +587,8 @@ async def content(
         file_id: str,
         *,
         vector_store_id: str,
+        include_embeddings: bool | Omit = omit,
+        include_metadata: bool | Omit = omit,
         # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
         # The extra values given here take precedence over values defined on the client or passed to this method.
         extra_headers: Headers | None = None,
@@ -582,6 +600,10 @@ async def content(
         Retrieves the contents of a vector store file.
 
         Args:
+          include_embeddings: Whether to include embedding vectors in the response.
+
+          include_metadata: Whether to include chunk metadata in the response.
+
           extra_headers: Send extra headers
 
           extra_query: Add additional query parameters to the request
@@ -597,7 +619,17 @@ async def content(
         return await self._get(
             f"/v1/vector_stores/{vector_store_id}/files/{file_id}/content",
             options=make_request_options(
-                extra_headers=extra_headers, extra_query=extra_query, extra_body=extra_body, timeout=timeout
+                extra_headers=extra_headers,
+                extra_query=extra_query,
+                extra_body=extra_body,
+                timeout=timeout,
+                query=await async_maybe_transform(
+                    {
+                        "include_embeddings": include_embeddings,
+                        "include_metadata": include_metadata,
+                    },
+                    file_content_params.FileContentParams,
+                ),
             ),
             cast_to=FileContentResponse,
         )

src/llama_stack_client/types/vector_store_search_response.py

@@ -11,7 +11,45 @@
 
 from .._models import BaseModel
 
-__all__ = ["VectorStoreSearchResponse", "Data", "DataContent"]
+__all__ = ["VectorStoreSearchResponse", "Data", "DataContent", "DataContentChunkMetadata"]
+
+
+class DataContentChunkMetadata(BaseModel):
+    chunk_embedding_dimension: Optional[int] = None
+    """The dimension of the embedding vector for the chunk."""
+
+    chunk_embedding_model: Optional[str] = None
+    """The embedding model used to create the chunk's embedding."""
+
+    chunk_id: Optional[str] = None
+    """The ID of the chunk.
+
+    If not set, it will be generated based on the document ID and content.
+    """
+
+    chunk_tokenizer: Optional[str] = None
+    """The tokenizer used to create the chunk. Default is Tiktoken."""
+
+    chunk_window: Optional[str] = None
+    """The window of the chunk, which can be used to group related chunks together."""
+
+    content_token_count: Optional[int] = None
+    """The number of tokens in the content of the chunk."""
+
+    created_timestamp: Optional[int] = None
+    """An optional timestamp indicating when the chunk was created."""
+
+    document_id: Optional[str] = None
+    """The ID of the document this chunk belongs to."""
+
+    metadata_token_count: Optional[int] = None
+    """The number of tokens in the metadata of the chunk."""
+
+    source: Optional[str] = None
+    """The source of the content, such as a URL, file path, or other identifier."""
+
+    updated_timestamp: Optional[int] = None
+    """An optional timestamp indicating when the chunk was last updated."""
 
 
 class DataContent(BaseModel):
@@ -21,6 +59,15 @@ class DataContent(BaseModel):
     type: Literal["text"]
     """Content type, currently only "text" is supported"""
 
+    chunk_metadata: Optional[DataContentChunkMetadata] = None
+    """Optional chunk metadata"""
+
+    embedding: Optional[List[float]] = None
+    """Optional embedding vector for this content chunk"""
+
+    metadata: Optional[Dict[str, Union[bool, float, str, List[object], object, None]]] = None
+    """Optional user-defined metadata"""
+
 
 class Data(BaseModel):
     content: List[DataContent]

src/llama_stack_client/types/vector_stores/__init__.py

@@ -12,6 +12,7 @@
 from .vector_store_file import VectorStoreFile as VectorStoreFile
 from .file_create_params import FileCreateParams as FileCreateParams
 from .file_update_params import FileUpdateParams as FileUpdateParams
+from .file_content_params import FileContentParams as FileContentParams
 from .file_delete_response import FileDeleteResponse as FileDeleteResponse
 from .file_content_response import FileContentResponse as FileContentResponse
 from .file_batch_create_params import FileBatchCreateParams as FileBatchCreateParams

src/llama_stack_client/types/vector_stores/file_content_params.py

@@ -0,0 +1,17 @@
+# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
+
+from __future__ import annotations
+
+from typing_extensions import Required, TypedDict
+
+__all__ = ["FileContentParams"]
+
+
+class FileContentParams(TypedDict, total=False):
+    vector_store_id: Required[str]
+
+    include_embeddings: bool
+    """Whether to include embedding vectors in the response."""
+
+    include_metadata: bool
+    """Whether to include chunk metadata in the response."""

src/llama_stack_client/types/vector_stores/file_content_response.py

@@ -6,12 +6,50 @@
 
 # File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
 
-from typing import List, Optional
+from typing import Dict, List, Union, Optional
 from typing_extensions import Literal
 
 from ..._models import BaseModel
 
-__all__ = ["FileContentResponse", "Data"]
+__all__ = ["FileContentResponse", "Data", "DataChunkMetadata"]
+
+
+class DataChunkMetadata(BaseModel):
+    chunk_embedding_dimension: Optional[int] = None
+    """The dimension of the embedding vector for the chunk."""
+
+    chunk_embedding_model: Optional[str] = None
+    """The embedding model used to create the chunk's embedding."""
+
+    chunk_id: Optional[str] = None
+    """The ID of the chunk.
+
+    If not set, it will be generated based on the document ID and content.
+    """
+
+    chunk_tokenizer: Optional[str] = None
+    """The tokenizer used to create the chunk. Default is Tiktoken."""
+
+    chunk_window: Optional[str] = None
+    """The window of the chunk, which can be used to group related chunks together."""
+
+    content_token_count: Optional[int] = None
+    """The number of tokens in the content of the chunk."""
+
+    created_timestamp: Optional[int] = None
+    """An optional timestamp indicating when the chunk was created."""
+
+    document_id: Optional[str] = None
+    """The ID of the document this chunk belongs to."""
+
+    metadata_token_count: Optional[int] = None
+    """The number of tokens in the metadata of the chunk."""
+
+    source: Optional[str] = None
+    """The source of the content, such as a URL, file path, or other identifier."""
+
+    updated_timestamp: Optional[int] = None
+    """An optional timestamp indicating when the chunk was last updated."""
 
 
 class Data(BaseModel):
@@ -21,6 +59,15 @@ class Data(BaseModel):
     type: Literal["text"]
     """Content type, currently only "text" is supported"""
 
+    chunk_metadata: Optional[DataChunkMetadata] = None
+    """Optional chunk metadata"""
+
+    embedding: Optional[List[float]] = None
+    """Optional embedding vector for this content chunk"""
+
+    metadata: Optional[Dict[str, Union[bool, float, str, List[object], object, None]]] = None
+    """Optional user-defined metadata"""
+
 
 class FileContentResponse(BaseModel):
     data: List[Data]

tests/api_resources/vector_stores/test_files.py

@@ -287,6 +287,16 @@ def test_method_content(self, client: LlamaStackClient) -> None:
         )
         assert_matches_type(FileContentResponse, file, path=["response"])
 
+    @parametrize
+    def test_method_content_with_all_params(self, client: LlamaStackClient) -> None:
+        file = client.vector_stores.files.content(
+            file_id="file_id",
+            vector_store_id="vector_store_id",
+            include_embeddings=True,
+            include_metadata=True,
+        )
+        assert_matches_type(FileContentResponse, file, path=["response"])
+
     @parametrize
     def test_raw_response_content(self, client: LlamaStackClient) -> None:
         response = client.vector_stores.files.with_raw_response.content(
@@ -592,6 +602,16 @@ async def test_method_content(self, async_client: AsyncLlamaStackClient) -> None
         )
         assert_matches_type(FileContentResponse, file, path=["response"])
 
+    @parametrize
+    async def test_method_content_with_all_params(self, async_client: AsyncLlamaStackClient) -> None:
+        file = await async_client.vector_stores.files.content(
+            file_id="file_id",
+            vector_store_id="vector_store_id",
+            include_embeddings=True,
+            include_metadata=True,
+        )
+        assert_matches_type(FileContentResponse, file, path=["response"])
+
     @parametrize
     async def test_raw_response_content(self, async_client: AsyncLlamaStackClient) -> None:
         response = await async_client.vector_stores.files.with_raw_response.content(