feat(vector store): improve chunking strategy type names (#1690)

openai · Sep 6, 2024 · f161000 · f161000
1 parent b2f58cb
commit f161000
Show file tree

Hide file tree

Showing 20 changed files with 201 additions and 321 deletions.
diff --git a/.stats.yml b/.stats.yml
@@ -1,2 +1,2 @@
 configured_endpoints: 68
-openapi_spec_url: https://storage.googleapis.com/stainless-sdk-openapi-specs/openai-1dbac0e95bdb5a89a0dd3d93265475a378214551b7d8c22862928e0d87ace94b.yml
+openapi_spec_url: https://storage.googleapis.com/stainless-sdk-openapi-specs/openai-85a85e0c08de456441431c0ae4e9c078cc8f9748c29430b9a9058340db6389ee.yml
diff --git a/api.md b/api.md
@@ -222,7 +222,17 @@ Methods:
 Types:
 
 ```python
-from openai.types.beta import VectorStore, VectorStoreDeleted
+from openai.types.beta import (
+    AutoFileChunkingStrategyParam,
+    FileChunkingStrategy,
+    FileChunkingStrategyParam,
+    OtherFileChunkingStrategyObject,
+    StaticFileChunkingStrategy,
+    StaticFileChunkingStrategyObject,
+    StaticFileChunkingStrategyParam,
+    VectorStore,
+    VectorStoreDeleted,
+)
 ```
 
 Methods:

diff --git a/src/openai/resources/beta/vector_stores/file_batches.py b/src/openai/resources/beta/vector_stores/file_batches.py
@@ -22,11 +22,10 @@
 from ...._resource import SyncAPIResource, AsyncAPIResource
 from ...._response import to_streamed_response_wrapper, async_to_streamed_response_wrapper
 from ....pagination import SyncCursorPage, AsyncCursorPage
-from ...._base_client import (
-    AsyncPaginator,
-    make_request_options,
-)
+from ....types.beta import FileChunkingStrategyParam
+from ...._base_client import AsyncPaginator, make_request_options
 from ....types.beta.vector_stores import file_batch_create_params, file_batch_list_files_params
+from ....types.beta.file_chunking_strategy_param import FileChunkingStrategyParam
 from ....types.beta.vector_stores.vector_store_file import VectorStoreFile
 from ....types.beta.vector_stores.vector_store_file_batch import VectorStoreFileBatch
 
@@ -47,7 +46,7 @@ def create(
         vector_store_id: str,
         *,
         file_ids: List[str],
-        chunking_strategy: file_batch_create_params.ChunkingStrategy | NotGiven = NOT_GIVEN,
+        chunking_strategy: FileChunkingStrategyParam | NotGiven = NOT_GIVEN,
         # 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,
@@ -64,7 +63,7 @@ def create(
               files.
 
           chunking_strategy: The chunking strategy used to chunk the file(s). If not set, will use the `auto`
-              strategy.
+              strategy. Only applicable if `file_ids` is non-empty.
 
           extra_headers: Send extra headers
 
@@ -174,7 +173,7 @@ def create_and_poll(
         *,
         file_ids: List[str],
         poll_interval_ms: int | NotGiven = NOT_GIVEN,
-        chunking_strategy: file_batch_create_params.ChunkingStrategy | NotGiven = NOT_GIVEN,
+        chunking_strategy: FileChunkingStrategyParam | NotGiven = NOT_GIVEN,
     ) -> VectorStoreFileBatch:
         """Create a vector store batch and poll until all files have been processed."""
         batch = self.create(
@@ -308,7 +307,7 @@ def upload_and_poll(
         max_concurrency: int = 5,
         file_ids: List[str] = [],
         poll_interval_ms: int | NotGiven = NOT_GIVEN,
-        chunking_strategy: file_batch_create_params.ChunkingStrategy | NotGiven = NOT_GIVEN,
+        chunking_strategy: FileChunkingStrategyParam | NotGiven = NOT_GIVEN,
     ) -> VectorStoreFileBatch:
         """Uploads the given files concurrently and then creates a vector store file batch.
 
@@ -365,7 +364,7 @@ async def create(
         vector_store_id: str,
         *,
         file_ids: List[str],
-        chunking_strategy: file_batch_create_params.ChunkingStrategy | NotGiven = NOT_GIVEN,
+        chunking_strategy: FileChunkingStrategyParam | NotGiven = NOT_GIVEN,
         # 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,
@@ -382,7 +381,7 @@ async def create(
               files.
 
           chunking_strategy: The chunking strategy used to chunk the file(s). If not set, will use the `auto`
-              strategy.
+              strategy. Only applicable if `file_ids` is non-empty.
 
           extra_headers: Send extra headers
 
@@ -492,7 +491,7 @@ async def create_and_poll(
         *,
         file_ids: List[str],
         poll_interval_ms: int | NotGiven = NOT_GIVEN,
-        chunking_strategy: file_batch_create_params.ChunkingStrategy | NotGiven = NOT_GIVEN,
+        chunking_strategy: FileChunkingStrategyParam | NotGiven = NOT_GIVEN,
     ) -> VectorStoreFileBatch:
         """Create a vector store batch and poll until all files have been processed."""
         batch = await self.create(
@@ -626,7 +625,7 @@ async def upload_and_poll(
         max_concurrency: int = 5,
         file_ids: List[str] = [],
         poll_interval_ms: int | NotGiven = NOT_GIVEN,
-        chunking_strategy: file_batch_create_params.ChunkingStrategy | NotGiven = NOT_GIVEN,
+        chunking_strategy: FileChunkingStrategyParam | NotGiven = NOT_GIVEN,
     ) -> VectorStoreFileBatch:
         """Uploads the given files concurrently and then creates a vector store file batch.
 

diff --git a/src/openai/resources/beta/vector_stores/files.py b/src/openai/resources/beta/vector_stores/files.py
@@ -18,11 +18,10 @@
 from ...._resource import SyncAPIResource, AsyncAPIResource
 from ...._response import to_streamed_response_wrapper, async_to_streamed_response_wrapper
 from ....pagination import SyncCursorPage, AsyncCursorPage
-from ...._base_client import (
-    AsyncPaginator,
-    make_request_options,
-)
+from ....types.beta import FileChunkingStrategyParam
+from ...._base_client import AsyncPaginator, make_request_options
 from ....types.beta.vector_stores import file_list_params, file_create_params
+from ....types.beta.file_chunking_strategy_param import FileChunkingStrategyParam
 from ....types.beta.vector_stores.vector_store_file import VectorStoreFile
 from ....types.beta.vector_stores.vector_store_file_deleted import VectorStoreFileDeleted
 
@@ -43,7 +42,7 @@ def create(
         vector_store_id: str,
         *,
         file_id: str,
-        chunking_strategy: file_create_params.ChunkingStrategy | NotGiven = NOT_GIVEN,
+        chunking_strategy: FileChunkingStrategyParam | NotGiven = NOT_GIVEN,
         # 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,
@@ -62,7 +61,7 @@ def create(
               files.
 
           chunking_strategy: The chunking strategy used to chunk the file(s). If not set, will use the `auto`
-              strategy.
+              strategy. Only applicable if `file_ids` is non-empty.
 
           extra_headers: Send extra headers
 
@@ -245,7 +244,7 @@ def create_and_poll(
         *,
         vector_store_id: str,
         poll_interval_ms: int | NotGiven = NOT_GIVEN,
-        chunking_strategy: file_create_params.ChunkingStrategy | NotGiven = NOT_GIVEN,
+        chunking_strategy: FileChunkingStrategyParam | NotGiven = NOT_GIVEN,
     ) -> VectorStoreFile:
         """Attach a file to the given vector store and wait for it to be processed."""
         self.create(vector_store_id=vector_store_id, file_id=file_id, chunking_strategy=chunking_strategy)
@@ -302,7 +301,7 @@ def upload(
         *,
         vector_store_id: str,
         file: FileTypes,
-        chunking_strategy: file_create_params.ChunkingStrategy | NotGiven = NOT_GIVEN,
+        chunking_strategy: FileChunkingStrategyParam | NotGiven = NOT_GIVEN,
     ) -> VectorStoreFile:
         """Upload a file to the `files` API and then attach it to the given vector store.
 
@@ -318,7 +317,7 @@ def upload_and_poll(
         vector_store_id: str,
         file: FileTypes,
         poll_interval_ms: int | NotGiven = NOT_GIVEN,
-        chunking_strategy: file_create_params.ChunkingStrategy | NotGiven = NOT_GIVEN,
+        chunking_strategy: FileChunkingStrategyParam | NotGiven = NOT_GIVEN,
     ) -> VectorStoreFile:
         """Add a file to a vector store and poll until processing is complete."""
         file_obj = self._client.files.create(file=file, purpose="assistants")
@@ -344,7 +343,7 @@ async def create(
         vector_store_id: str,
         *,
         file_id: str,
-        chunking_strategy: file_create_params.ChunkingStrategy | NotGiven = NOT_GIVEN,
+        chunking_strategy: FileChunkingStrategyParam | NotGiven = NOT_GIVEN,
         # 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,
@@ -363,7 +362,7 @@ async def create(
               files.
 
           chunking_strategy: The chunking strategy used to chunk the file(s). If not set, will use the `auto`
-              strategy.
+              strategy. Only applicable if `file_ids` is non-empty.
 
           extra_headers: Send extra headers
 
@@ -546,7 +545,7 @@ async def create_and_poll(
         *,
         vector_store_id: str,
         poll_interval_ms: int | NotGiven = NOT_GIVEN,
-        chunking_strategy: file_create_params.ChunkingStrategy | NotGiven = NOT_GIVEN,
+        chunking_strategy: FileChunkingStrategyParam | NotGiven = NOT_GIVEN,
     ) -> VectorStoreFile:
         """Attach a file to the given vector store and wait for it to be processed."""
         await self.create(vector_store_id=vector_store_id, file_id=file_id, chunking_strategy=chunking_strategy)
@@ -603,7 +602,7 @@ async def upload(
         *,
         vector_store_id: str,
         file: FileTypes,
-        chunking_strategy: file_create_params.ChunkingStrategy | NotGiven = NOT_GIVEN,
+        chunking_strategy: FileChunkingStrategyParam | NotGiven = NOT_GIVEN,
     ) -> VectorStoreFile:
         """Upload a file to the `files` API and then attach it to the given vector store.
 
@@ -621,7 +620,7 @@ async def upload_and_poll(
         vector_store_id: str,
         file: FileTypes,
         poll_interval_ms: int | NotGiven = NOT_GIVEN,
-        chunking_strategy: file_create_params.ChunkingStrategy | NotGiven = NOT_GIVEN,
+        chunking_strategy: FileChunkingStrategyParam | NotGiven = NOT_GIVEN,
     ) -> VectorStoreFile:
         """Add a file to a vector store and poll until processing is complete."""
         file_obj = await self._client.files.create(file=file, purpose="assistants")

diff --git a/src/openai/resources/beta/vector_stores/vector_stores.py b/src/openai/resources/beta/vector_stores/vector_stores.py
@@ -33,13 +33,16 @@
     AsyncFileBatchesWithStreamingResponse,
 )
 from ....pagination import SyncCursorPage, AsyncCursorPage
-from ....types.beta import vector_store_list_params, vector_store_create_params, vector_store_update_params
-from ...._base_client import (
-    AsyncPaginator,
-    make_request_options,
+from ....types.beta import (
+    FileChunkingStrategyParam,
+    vector_store_list_params,
+    vector_store_create_params,
+    vector_store_update_params,
 )
+from ...._base_client import AsyncPaginator, make_request_options
 from ....types.beta.vector_store import VectorStore
 from ....types.beta.vector_store_deleted import VectorStoreDeleted
+from ....types.beta.file_chunking_strategy_param import FileChunkingStrategyParam
 
 __all__ = ["VectorStores", "AsyncVectorStores"]
 
@@ -64,7 +67,7 @@ def with_streaming_response(self) -> VectorStoresWithStreamingResponse:
     def create(
         self,
         *,
-        chunking_strategy: vector_store_create_params.ChunkingStrategy | NotGiven = NOT_GIVEN,
+        chunking_strategy: FileChunkingStrategyParam | NotGiven = NOT_GIVEN,
         expires_after: vector_store_create_params.ExpiresAfter | NotGiven = NOT_GIVEN,
         file_ids: List[str] | NotGiven = NOT_GIVEN,
         metadata: Optional[object] | NotGiven = NOT_GIVEN,
@@ -331,7 +334,7 @@ def with_streaming_response(self) -> AsyncVectorStoresWithStreamingResponse:
     async def create(
         self,
         *,
-        chunking_strategy: vector_store_create_params.ChunkingStrategy | NotGiven = NOT_GIVEN,
+        chunking_strategy: FileChunkingStrategyParam | NotGiven = NOT_GIVEN,
         expires_after: vector_store_create_params.ExpiresAfter | NotGiven = NOT_GIVEN,
         file_ids: List[str] | NotGiven = NOT_GIVEN,
         metadata: Optional[object] | NotGiven = NOT_GIVEN,

diff --git a/src/openai/types/beta/__init__.py b/src/openai/types/beta/__init__.py
@@ -19,6 +19,7 @@
 from .assistant_tool_choice import AssistantToolChoice as AssistantToolChoice
 from .code_interpreter_tool import CodeInterpreterTool as CodeInterpreterTool
 from .assistant_stream_event import AssistantStreamEvent as AssistantStreamEvent
+from .file_chunking_strategy import FileChunkingStrategy as FileChunkingStrategy
 from .file_search_tool_param import FileSearchToolParam as FileSearchToolParam
 from .assistant_create_params import AssistantCreateParams as AssistantCreateParams
 from .assistant_update_params import AssistantUpdateParams as AssistantUpdateParams
@@ -28,11 +29,17 @@
 from .assistant_tool_choice_param import AssistantToolChoiceParam as AssistantToolChoiceParam
 from .code_interpreter_tool_param import CodeInterpreterToolParam as CodeInterpreterToolParam
 from .assistant_tool_choice_option import AssistantToolChoiceOption as AssistantToolChoiceOption
+from .file_chunking_strategy_param import FileChunkingStrategyParam as FileChunkingStrategyParam
 from .thread_create_and_run_params import ThreadCreateAndRunParams as ThreadCreateAndRunParams
+from .static_file_chunking_strategy import StaticFileChunkingStrategy as StaticFileChunkingStrategy
 from .assistant_tool_choice_function import AssistantToolChoiceFunction as AssistantToolChoiceFunction
 from .assistant_response_format_option import AssistantResponseFormatOption as AssistantResponseFormatOption
+from .auto_file_chunking_strategy_param import AutoFileChunkingStrategyParam as AutoFileChunkingStrategyParam
 from .assistant_tool_choice_option_param import AssistantToolChoiceOptionParam as AssistantToolChoiceOptionParam
+from .other_file_chunking_strategy_object import OtherFileChunkingStrategyObject as OtherFileChunkingStrategyObject
+from .static_file_chunking_strategy_param import StaticFileChunkingStrategyParam as StaticFileChunkingStrategyParam
 from .assistant_tool_choice_function_param import AssistantToolChoiceFunctionParam as AssistantToolChoiceFunctionParam
+from .static_file_chunking_strategy_object import StaticFileChunkingStrategyObject as StaticFileChunkingStrategyObject
 from .assistant_response_format_option_param import (
     AssistantResponseFormatOptionParam as AssistantResponseFormatOptionParam,
 )
diff --git a/src/openai/types/beta/assistant_create_params.py b/src/openai/types/beta/assistant_create_params.py
@@ -3,10 +3,11 @@
 from __future__ import annotations
 
 from typing import List, Union, Iterable, Optional
-from typing_extensions import Literal, Required, TypeAlias, TypedDict
+from typing_extensions import Required, TypedDict
 
 from ..chat_model import ChatModel
 from .assistant_tool_param import AssistantToolParam
+from .file_chunking_strategy_param import FileChunkingStrategyParam
 from .assistant_response_format_option_param import AssistantResponseFormatOptionParam
 
 __all__ = [
@@ -15,10 +16,6 @@
     "ToolResourcesCodeInterpreter",
     "ToolResourcesFileSearch",
     "ToolResourcesFileSearchVectorStore",
-    "ToolResourcesFileSearchVectorStoreChunkingStrategy",
-    "ToolResourcesFileSearchVectorStoreChunkingStrategyAuto",
-    "ToolResourcesFileSearchVectorStoreChunkingStrategyStatic",
-    "ToolResourcesFileSearchVectorStoreChunkingStrategyStaticStatic",
 ]
 
 
@@ -118,43 +115,12 @@ class ToolResourcesCodeInterpreter(TypedDict, total=False):
     """
 
 
-class ToolResourcesFileSearchVectorStoreChunkingStrategyAuto(TypedDict, total=False):
-    type: Required[Literal["auto"]]
-    """Always `auto`."""
-
-
-class ToolResourcesFileSearchVectorStoreChunkingStrategyStaticStatic(TypedDict, total=False):
-    chunk_overlap_tokens: Required[int]
-    """The number of tokens that overlap between chunks. The default value is `400`.
-
-    Note that the overlap must not exceed half of `max_chunk_size_tokens`.
-    """
-
-    max_chunk_size_tokens: Required[int]
-    """The maximum number of tokens in each chunk.
-
-    The default value is `800`. The minimum value is `100` and the maximum value is
-    `4096`.
-    """
-
-
-class ToolResourcesFileSearchVectorStoreChunkingStrategyStatic(TypedDict, total=False):
-    static: Required[ToolResourcesFileSearchVectorStoreChunkingStrategyStaticStatic]
-
-    type: Required[Literal["static"]]
-    """Always `static`."""
-
-
-ToolResourcesFileSearchVectorStoreChunkingStrategy: TypeAlias = Union[
-    ToolResourcesFileSearchVectorStoreChunkingStrategyAuto, ToolResourcesFileSearchVectorStoreChunkingStrategyStatic
-]
-
-
 class ToolResourcesFileSearchVectorStore(TypedDict, total=False):
-    chunking_strategy: ToolResourcesFileSearchVectorStoreChunkingStrategy
+    chunking_strategy: FileChunkingStrategyParam
     """The chunking strategy used to chunk the file(s).
 
-    If not set, will use the `auto` strategy.
+    If not set, will use the `auto` strategy. Only applicable if `file_ids` is
+    non-empty.
     """
 
     file_ids: List[str]

diff --git a/src/openai/types/beta/auto_file_chunking_strategy_param.py b/src/openai/types/beta/auto_file_chunking_strategy_param.py
@@ -0,0 +1,12 @@
+# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
+
+from __future__ import annotations
+
+from typing_extensions import Literal, Required, TypedDict
+
+__all__ = ["AutoFileChunkingStrategyParam"]
+
+
+class AutoFileChunkingStrategyParam(TypedDict, total=False):
+    type: Required[Literal["auto"]]
+    """Always `auto`."""
diff --git a/src/openai/types/beta/file_chunking_strategy.py b/src/openai/types/beta/file_chunking_strategy.py
@@ -0,0 +1,14 @@
+# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
+
+from typing import Union
+from typing_extensions import Annotated, TypeAlias
+
+from ..._utils import PropertyInfo
+from .other_file_chunking_strategy_object import OtherFileChunkingStrategyObject
+from .static_file_chunking_strategy_object import StaticFileChunkingStrategyObject
+
+__all__ = ["FileChunkingStrategy"]
+
+FileChunkingStrategy: TypeAlias = Annotated[
+    Union[StaticFileChunkingStrategyObject, OtherFileChunkingStrategyObject], PropertyInfo(discriminator="type")
+]
diff --git a/src/openai/types/beta/file_chunking_strategy_param.py b/src/openai/types/beta/file_chunking_strategy_param.py
@@ -0,0 +1,13 @@
+# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
+
+from __future__ import annotations
+
+from typing import Union
+from typing_extensions import TypeAlias
+
+from .auto_file_chunking_strategy_param import AutoFileChunkingStrategyParam
+from .static_file_chunking_strategy_param import StaticFileChunkingStrategyParam
+
+__all__ = ["FileChunkingStrategyParam"]
+
+FileChunkingStrategyParam: TypeAlias = Union[AutoFileChunkingStrategyParam, StaticFileChunkingStrategyParam]