diff options
author | S. Solomon Darnell | 2025-03-28 21:52:21 -0500 |
---|---|---|
committer | S. Solomon Darnell | 2025-03-28 21:52:21 -0500 |
commit | 4a52a71956a8d46fcb7294ac71734504bb09bcc2 (patch) | |
tree | ee3dc5af3b6313e921cd920906356f5d4febc4ed /.venv/lib/python3.12/site-packages/openai/resources/vector_stores | |
parent | cc961e04ba734dd72309fb548a2f97d67d578813 (diff) | |
download | gn-ai-master.tar.gz |
Diffstat (limited to '.venv/lib/python3.12/site-packages/openai/resources/vector_stores')
4 files changed, 2649 insertions, 0 deletions
diff --git a/.venv/lib/python3.12/site-packages/openai/resources/vector_stores/__init__.py b/.venv/lib/python3.12/site-packages/openai/resources/vector_stores/__init__.py new file mode 100644 index 00000000..96ae16c3 --- /dev/null +++ b/.venv/lib/python3.12/site-packages/openai/resources/vector_stores/__init__.py @@ -0,0 +1,47 @@ +# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details. + +from .files import ( + Files, + AsyncFiles, + FilesWithRawResponse, + AsyncFilesWithRawResponse, + FilesWithStreamingResponse, + AsyncFilesWithStreamingResponse, +) +from .file_batches import ( + FileBatches, + AsyncFileBatches, + FileBatchesWithRawResponse, + AsyncFileBatchesWithRawResponse, + FileBatchesWithStreamingResponse, + AsyncFileBatchesWithStreamingResponse, +) +from .vector_stores import ( + VectorStores, + AsyncVectorStores, + VectorStoresWithRawResponse, + AsyncVectorStoresWithRawResponse, + VectorStoresWithStreamingResponse, + AsyncVectorStoresWithStreamingResponse, +) + +__all__ = [ + "Files", + "AsyncFiles", + "FilesWithRawResponse", + "AsyncFilesWithRawResponse", + "FilesWithStreamingResponse", + "AsyncFilesWithStreamingResponse", + "FileBatches", + "AsyncFileBatches", + "FileBatchesWithRawResponse", + "AsyncFileBatchesWithRawResponse", + "FileBatchesWithStreamingResponse", + "AsyncFileBatchesWithStreamingResponse", + "VectorStores", + "AsyncVectorStores", + "VectorStoresWithRawResponse", + "AsyncVectorStoresWithRawResponse", + "VectorStoresWithStreamingResponse", + "AsyncVectorStoresWithStreamingResponse", +] diff --git a/.venv/lib/python3.12/site-packages/openai/resources/vector_stores/file_batches.py b/.venv/lib/python3.12/site-packages/openai/resources/vector_stores/file_batches.py new file mode 100644 index 00000000..9b4b64d3 --- /dev/null +++ b/.venv/lib/python3.12/site-packages/openai/resources/vector_stores/file_batches.py @@ -0,0 +1,801 @@ +# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details. + +from __future__ import annotations + +import asyncio +from typing import Dict, List, Iterable, Optional +from typing_extensions import Union, Literal +from concurrent.futures import Future, ThreadPoolExecutor, as_completed + +import httpx +import sniffio + +from ... import _legacy_response +from ...types import FileChunkingStrategyParam +from ..._types import NOT_GIVEN, Body, Query, Headers, NotGiven, FileTypes +from ..._utils import ( + is_given, + maybe_transform, + async_maybe_transform, +) +from ..._compat import cached_property +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.file_object import FileObject +from ...types.vector_stores import file_batch_create_params, file_batch_list_files_params +from ...types.file_chunking_strategy_param import FileChunkingStrategyParam +from ...types.vector_stores.vector_store_file import VectorStoreFile +from ...types.vector_stores.vector_store_file_batch import VectorStoreFileBatch + +__all__ = ["FileBatches", "AsyncFileBatches"] + + +class FileBatches(SyncAPIResource): + @cached_property + def with_raw_response(self) -> FileBatchesWithRawResponse: + """ + This property can be used as a prefix for any HTTP method call to return + the raw response object instead of the parsed content. + + For more information, see https://www.github.com/openai/openai-python#accessing-raw-response-data-eg-headers + """ + return FileBatchesWithRawResponse(self) + + @cached_property + def with_streaming_response(self) -> FileBatchesWithStreamingResponse: + """ + An alternative to `.with_raw_response` that doesn't eagerly read the response body. + + For more information, see https://www.github.com/openai/openai-python#with_streaming_response + """ + return FileBatchesWithStreamingResponse(self) + + def create( + self, + vector_store_id: str, + *, + file_ids: List[str], + attributes: Optional[Dict[str, Union[str, float, bool]]] | 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, + extra_query: Query | None = None, + extra_body: Body | None = None, + timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN, + ) -> VectorStoreFileBatch: + """ + Create a vector store file batch. + + Args: + file_ids: A list of [File](https://platform.openai.com/docs/api-reference/files) IDs that + the vector store should use. Useful for tools like `file_search` that can access + files. + + attributes: Set of 16 key-value pairs that can be attached to an object. This can be useful + for storing additional information about the object in a structured format, and + querying for objects via API or the dashboard. Keys are strings with a maximum + length of 64 characters. Values are strings with a maximum length of 512 + characters, booleans, or numbers. + + chunking_strategy: The chunking strategy used to chunk the file(s). If not set, will use the `auto` + strategy. Only applicable if `file_ids` is non-empty. + + extra_headers: Send extra headers + + extra_query: Add additional query parameters to the request + + extra_body: Add additional JSON properties to the request + + timeout: Override the client-level default timeout for this request, in seconds + """ + if not vector_store_id: + raise ValueError(f"Expected a non-empty value for `vector_store_id` but received {vector_store_id!r}") + extra_headers = {"OpenAI-Beta": "assistants=v2", **(extra_headers or {})} + return self._post( + f"/vector_stores/{vector_store_id}/file_batches", + body=maybe_transform( + { + "file_ids": file_ids, + "attributes": attributes, + "chunking_strategy": chunking_strategy, + }, + file_batch_create_params.FileBatchCreateParams, + ), + options=make_request_options( + extra_headers=extra_headers, extra_query=extra_query, extra_body=extra_body, timeout=timeout + ), + cast_to=VectorStoreFileBatch, + ) + + def retrieve( + self, + batch_id: str, + *, + vector_store_id: str, + # 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, + extra_query: Query | None = None, + extra_body: Body | None = None, + timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN, + ) -> VectorStoreFileBatch: + """ + Retrieves a vector store file batch. + + Args: + extra_headers: Send extra headers + + extra_query: Add additional query parameters to the request + + extra_body: Add additional JSON properties to the request + + timeout: Override the client-level default timeout for this request, in seconds + """ + if not vector_store_id: + raise ValueError(f"Expected a non-empty value for `vector_store_id` but received {vector_store_id!r}") + if not batch_id: + raise ValueError(f"Expected a non-empty value for `batch_id` but received {batch_id!r}") + extra_headers = {"OpenAI-Beta": "assistants=v2", **(extra_headers or {})} + return self._get( + f"/vector_stores/{vector_store_id}/file_batches/{batch_id}", + options=make_request_options( + extra_headers=extra_headers, extra_query=extra_query, extra_body=extra_body, timeout=timeout + ), + cast_to=VectorStoreFileBatch, + ) + + def cancel( + self, + batch_id: str, + *, + vector_store_id: str, + # 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, + extra_query: Query | None = None, + extra_body: Body | None = None, + timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN, + ) -> VectorStoreFileBatch: + """Cancel a vector store file batch. + + This attempts to cancel the processing of + files in this batch as soon as possible. + + Args: + extra_headers: Send extra headers + + extra_query: Add additional query parameters to the request + + extra_body: Add additional JSON properties to the request + + timeout: Override the client-level default timeout for this request, in seconds + """ + if not vector_store_id: + raise ValueError(f"Expected a non-empty value for `vector_store_id` but received {vector_store_id!r}") + if not batch_id: + raise ValueError(f"Expected a non-empty value for `batch_id` but received {batch_id!r}") + extra_headers = {"OpenAI-Beta": "assistants=v2", **(extra_headers or {})} + return self._post( + f"/vector_stores/{vector_store_id}/file_batches/{batch_id}/cancel", + options=make_request_options( + extra_headers=extra_headers, extra_query=extra_query, extra_body=extra_body, timeout=timeout + ), + cast_to=VectorStoreFileBatch, + ) + + def create_and_poll( + self, + vector_store_id: str, + *, + file_ids: List[str], + poll_interval_ms: int | 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( + vector_store_id=vector_store_id, + file_ids=file_ids, + chunking_strategy=chunking_strategy, + ) + # TODO: don't poll unless necessary?? + return self.poll( + batch.id, + vector_store_id=vector_store_id, + poll_interval_ms=poll_interval_ms, + ) + + def list_files( + self, + batch_id: str, + *, + vector_store_id: str, + after: str | NotGiven = NOT_GIVEN, + before: str | NotGiven = NOT_GIVEN, + filter: Literal["in_progress", "completed", "failed", "cancelled"] | NotGiven = NOT_GIVEN, + limit: int | NotGiven = NOT_GIVEN, + order: Literal["asc", "desc"] | 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, + extra_query: Query | None = None, + extra_body: Body | None = None, + timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN, + ) -> SyncCursorPage[VectorStoreFile]: + """ + Returns a list of vector store files in a batch. + + Args: + after: A cursor for use in pagination. `after` is an object ID that defines your place + in the list. For instance, if you make a list request and receive 100 objects, + ending with obj_foo, your subsequent call can include after=obj_foo in order to + fetch the next page of the list. + + before: A cursor for use in pagination. `before` is an object ID that defines your place + in the list. For instance, if you make a list request and receive 100 objects, + starting with obj_foo, your subsequent call can include before=obj_foo in order + to fetch the previous page of the list. + + filter: Filter by file status. One of `in_progress`, `completed`, `failed`, `cancelled`. + + limit: A limit on the number of objects to be returned. Limit can range between 1 and + 100, and the default is 20. + + order: Sort order by the `created_at` timestamp of the objects. `asc` for ascending + order and `desc` for descending order. + + extra_headers: Send extra headers + + extra_query: Add additional query parameters to the request + + extra_body: Add additional JSON properties to the request + + timeout: Override the client-level default timeout for this request, in seconds + """ + if not vector_store_id: + raise ValueError(f"Expected a non-empty value for `vector_store_id` but received {vector_store_id!r}") + if not batch_id: + raise ValueError(f"Expected a non-empty value for `batch_id` but received {batch_id!r}") + extra_headers = {"OpenAI-Beta": "assistants=v2", **(extra_headers or {})} + return self._get_api_list( + f"/vector_stores/{vector_store_id}/file_batches/{batch_id}/files", + page=SyncCursorPage[VectorStoreFile], + options=make_request_options( + extra_headers=extra_headers, + extra_query=extra_query, + extra_body=extra_body, + timeout=timeout, + query=maybe_transform( + { + "after": after, + "before": before, + "filter": filter, + "limit": limit, + "order": order, + }, + file_batch_list_files_params.FileBatchListFilesParams, + ), + ), + model=VectorStoreFile, + ) + + def poll( + self, + batch_id: str, + *, + vector_store_id: str, + poll_interval_ms: int | NotGiven = NOT_GIVEN, + ) -> VectorStoreFileBatch: + """Wait for the given file batch to be processed. + + Note: this will return even if one of the files failed to process, you need to + check batch.file_counts.failed_count to handle this case. + """ + headers: dict[str, str] = {"X-Stainless-Poll-Helper": "true"} + if is_given(poll_interval_ms): + headers["X-Stainless-Custom-Poll-Interval"] = str(poll_interval_ms) + + while True: + response = self.with_raw_response.retrieve( + batch_id, + vector_store_id=vector_store_id, + extra_headers=headers, + ) + + batch = response.parse() + if batch.file_counts.in_progress > 0: + if not is_given(poll_interval_ms): + from_header = response.headers.get("openai-poll-after-ms") + if from_header is not None: + poll_interval_ms = int(from_header) + else: + poll_interval_ms = 1000 + + self._sleep(poll_interval_ms / 1000) + continue + + return batch + + def upload_and_poll( + self, + vector_store_id: str, + *, + files: Iterable[FileTypes], + max_concurrency: int = 5, + file_ids: List[str] = [], + poll_interval_ms: int | NotGiven = NOT_GIVEN, + chunking_strategy: FileChunkingStrategyParam | NotGiven = NOT_GIVEN, + ) -> VectorStoreFileBatch: + """Uploads the given files concurrently and then creates a vector store file batch. + + If you've already uploaded certain files that you want to include in this batch + then you can pass their IDs through the `file_ids` argument. + + By default, if any file upload fails then an exception will be eagerly raised. + + The number of concurrency uploads is configurable using the `max_concurrency` + parameter. + + Note: this method only supports `asyncio` or `trio` as the backing async + runtime. + """ + results: list[FileObject] = [] + + with ThreadPoolExecutor(max_workers=max_concurrency) as executor: + futures: list[Future[FileObject]] = [ + executor.submit( + self._client.files.create, + file=file, + purpose="assistants", + ) + for file in files + ] + + for future in as_completed(futures): + exc = future.exception() + if exc: + raise exc + + results.append(future.result()) + + batch = self.create_and_poll( + vector_store_id=vector_store_id, + file_ids=[*file_ids, *(f.id for f in results)], + poll_interval_ms=poll_interval_ms, + chunking_strategy=chunking_strategy, + ) + return batch + + +class AsyncFileBatches(AsyncAPIResource): + @cached_property + def with_raw_response(self) -> AsyncFileBatchesWithRawResponse: + """ + This property can be used as a prefix for any HTTP method call to return + the raw response object instead of the parsed content. + + For more information, see https://www.github.com/openai/openai-python#accessing-raw-response-data-eg-headers + """ + return AsyncFileBatchesWithRawResponse(self) + + @cached_property + def with_streaming_response(self) -> AsyncFileBatchesWithStreamingResponse: + """ + An alternative to `.with_raw_response` that doesn't eagerly read the response body. + + For more information, see https://www.github.com/openai/openai-python#with_streaming_response + """ + return AsyncFileBatchesWithStreamingResponse(self) + + async def create( + self, + vector_store_id: str, + *, + file_ids: List[str], + attributes: Optional[Dict[str, Union[str, float, bool]]] | 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, + extra_query: Query | None = None, + extra_body: Body | None = None, + timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN, + ) -> VectorStoreFileBatch: + """ + Create a vector store file batch. + + Args: + file_ids: A list of [File](https://platform.openai.com/docs/api-reference/files) IDs that + the vector store should use. Useful for tools like `file_search` that can access + files. + + attributes: Set of 16 key-value pairs that can be attached to an object. This can be useful + for storing additional information about the object in a structured format, and + querying for objects via API or the dashboard. Keys are strings with a maximum + length of 64 characters. Values are strings with a maximum length of 512 + characters, booleans, or numbers. + + chunking_strategy: The chunking strategy used to chunk the file(s). If not set, will use the `auto` + strategy. Only applicable if `file_ids` is non-empty. + + extra_headers: Send extra headers + + extra_query: Add additional query parameters to the request + + extra_body: Add additional JSON properties to the request + + timeout: Override the client-level default timeout for this request, in seconds + """ + if not vector_store_id: + raise ValueError(f"Expected a non-empty value for `vector_store_id` but received {vector_store_id!r}") + extra_headers = {"OpenAI-Beta": "assistants=v2", **(extra_headers or {})} + return await self._post( + f"/vector_stores/{vector_store_id}/file_batches", + body=await async_maybe_transform( + { + "file_ids": file_ids, + "attributes": attributes, + "chunking_strategy": chunking_strategy, + }, + file_batch_create_params.FileBatchCreateParams, + ), + options=make_request_options( + extra_headers=extra_headers, extra_query=extra_query, extra_body=extra_body, timeout=timeout + ), + cast_to=VectorStoreFileBatch, + ) + + async def retrieve( + self, + batch_id: str, + *, + vector_store_id: str, + # 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, + extra_query: Query | None = None, + extra_body: Body | None = None, + timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN, + ) -> VectorStoreFileBatch: + """ + Retrieves a vector store file batch. + + Args: + extra_headers: Send extra headers + + extra_query: Add additional query parameters to the request + + extra_body: Add additional JSON properties to the request + + timeout: Override the client-level default timeout for this request, in seconds + """ + if not vector_store_id: + raise ValueError(f"Expected a non-empty value for `vector_store_id` but received {vector_store_id!r}") + if not batch_id: + raise ValueError(f"Expected a non-empty value for `batch_id` but received {batch_id!r}") + extra_headers = {"OpenAI-Beta": "assistants=v2", **(extra_headers or {})} + return await self._get( + f"/vector_stores/{vector_store_id}/file_batches/{batch_id}", + options=make_request_options( + extra_headers=extra_headers, extra_query=extra_query, extra_body=extra_body, timeout=timeout + ), + cast_to=VectorStoreFileBatch, + ) + + async def cancel( + self, + batch_id: str, + *, + vector_store_id: str, + # 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, + extra_query: Query | None = None, + extra_body: Body | None = None, + timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN, + ) -> VectorStoreFileBatch: + """Cancel a vector store file batch. + + This attempts to cancel the processing of + files in this batch as soon as possible. + + Args: + extra_headers: Send extra headers + + extra_query: Add additional query parameters to the request + + extra_body: Add additional JSON properties to the request + + timeout: Override the client-level default timeout for this request, in seconds + """ + if not vector_store_id: + raise ValueError(f"Expected a non-empty value for `vector_store_id` but received {vector_store_id!r}") + if not batch_id: + raise ValueError(f"Expected a non-empty value for `batch_id` but received {batch_id!r}") + extra_headers = {"OpenAI-Beta": "assistants=v2", **(extra_headers or {})} + return await self._post( + f"/vector_stores/{vector_store_id}/file_batches/{batch_id}/cancel", + options=make_request_options( + extra_headers=extra_headers, extra_query=extra_query, extra_body=extra_body, timeout=timeout + ), + cast_to=VectorStoreFileBatch, + ) + + async def create_and_poll( + self, + vector_store_id: str, + *, + file_ids: List[str], + poll_interval_ms: int | 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( + vector_store_id=vector_store_id, + file_ids=file_ids, + chunking_strategy=chunking_strategy, + ) + # TODO: don't poll unless necessary?? + return await self.poll( + batch.id, + vector_store_id=vector_store_id, + poll_interval_ms=poll_interval_ms, + ) + + def list_files( + self, + batch_id: str, + *, + vector_store_id: str, + after: str | NotGiven = NOT_GIVEN, + before: str | NotGiven = NOT_GIVEN, + filter: Literal["in_progress", "completed", "failed", "cancelled"] | NotGiven = NOT_GIVEN, + limit: int | NotGiven = NOT_GIVEN, + order: Literal["asc", "desc"] | 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, + extra_query: Query | None = None, + extra_body: Body | None = None, + timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN, + ) -> AsyncPaginator[VectorStoreFile, AsyncCursorPage[VectorStoreFile]]: + """ + Returns a list of vector store files in a batch. + + Args: + after: A cursor for use in pagination. `after` is an object ID that defines your place + in the list. For instance, if you make a list request and receive 100 objects, + ending with obj_foo, your subsequent call can include after=obj_foo in order to + fetch the next page of the list. + + before: A cursor for use in pagination. `before` is an object ID that defines your place + in the list. For instance, if you make a list request and receive 100 objects, + starting with obj_foo, your subsequent call can include before=obj_foo in order + to fetch the previous page of the list. + + filter: Filter by file status. One of `in_progress`, `completed`, `failed`, `cancelled`. + + limit: A limit on the number of objects to be returned. Limit can range between 1 and + 100, and the default is 20. + + order: Sort order by the `created_at` timestamp of the objects. `asc` for ascending + order and `desc` for descending order. + + extra_headers: Send extra headers + + extra_query: Add additional query parameters to the request + + extra_body: Add additional JSON properties to the request + + timeout: Override the client-level default timeout for this request, in seconds + """ + if not vector_store_id: + raise ValueError(f"Expected a non-empty value for `vector_store_id` but received {vector_store_id!r}") + if not batch_id: + raise ValueError(f"Expected a non-empty value for `batch_id` but received {batch_id!r}") + extra_headers = {"OpenAI-Beta": "assistants=v2", **(extra_headers or {})} + return self._get_api_list( + f"/vector_stores/{vector_store_id}/file_batches/{batch_id}/files", + page=AsyncCursorPage[VectorStoreFile], + options=make_request_options( + extra_headers=extra_headers, + extra_query=extra_query, + extra_body=extra_body, + timeout=timeout, + query=maybe_transform( + { + "after": after, + "before": before, + "filter": filter, + "limit": limit, + "order": order, + }, + file_batch_list_files_params.FileBatchListFilesParams, + ), + ), + model=VectorStoreFile, + ) + + async def poll( + self, + batch_id: str, + *, + vector_store_id: str, + poll_interval_ms: int | NotGiven = NOT_GIVEN, + ) -> VectorStoreFileBatch: + """Wait for the given file batch to be processed. + + Note: this will return even if one of the files failed to process, you need to + check batch.file_counts.failed_count to handle this case. + """ + headers: dict[str, str] = {"X-Stainless-Poll-Helper": "true"} + if is_given(poll_interval_ms): + headers["X-Stainless-Custom-Poll-Interval"] = str(poll_interval_ms) + + while True: + response = await self.with_raw_response.retrieve( + batch_id, + vector_store_id=vector_store_id, + extra_headers=headers, + ) + + batch = response.parse() + if batch.file_counts.in_progress > 0: + if not is_given(poll_interval_ms): + from_header = response.headers.get("openai-poll-after-ms") + if from_header is not None: + poll_interval_ms = int(from_header) + else: + poll_interval_ms = 1000 + + await self._sleep(poll_interval_ms / 1000) + continue + + return batch + + async def upload_and_poll( + self, + vector_store_id: str, + *, + files: Iterable[FileTypes], + max_concurrency: int = 5, + file_ids: List[str] = [], + poll_interval_ms: int | NotGiven = NOT_GIVEN, + chunking_strategy: FileChunkingStrategyParam | NotGiven = NOT_GIVEN, + ) -> VectorStoreFileBatch: + """Uploads the given files concurrently and then creates a vector store file batch. + + If you've already uploaded certain files that you want to include in this batch + then you can pass their IDs through the `file_ids` argument. + + By default, if any file upload fails then an exception will be eagerly raised. + + The number of concurrency uploads is configurable using the `max_concurrency` + parameter. + + Note: this method only supports `asyncio` or `trio` as the backing async + runtime. + """ + uploaded_files: list[FileObject] = [] + + async_library = sniffio.current_async_library() + + if async_library == "asyncio": + + async def asyncio_upload_file(semaphore: asyncio.Semaphore, file: FileTypes) -> None: + async with semaphore: + file_obj = await self._client.files.create( + file=file, + purpose="assistants", + ) + uploaded_files.append(file_obj) + + semaphore = asyncio.Semaphore(max_concurrency) + + tasks = [asyncio_upload_file(semaphore, file) for file in files] + + await asyncio.gather(*tasks) + elif async_library == "trio": + # We only import if the library is being used. + # We support Python 3.7 so are using an older version of trio that does not have type information + import trio # type: ignore # pyright: ignore[reportMissingTypeStubs] + + async def trio_upload_file(limiter: trio.CapacityLimiter, file: FileTypes) -> None: + async with limiter: + file_obj = await self._client.files.create( + file=file, + purpose="assistants", + ) + uploaded_files.append(file_obj) + + limiter = trio.CapacityLimiter(max_concurrency) + + async with trio.open_nursery() as nursery: + for file in files: + nursery.start_soon(trio_upload_file, limiter, file) # pyright: ignore [reportUnknownMemberType] + else: + raise RuntimeError( + f"Async runtime {async_library} is not supported yet. Only asyncio or trio is supported", + ) + + batch = await self.create_and_poll( + vector_store_id=vector_store_id, + file_ids=[*file_ids, *(f.id for f in uploaded_files)], + poll_interval_ms=poll_interval_ms, + chunking_strategy=chunking_strategy, + ) + return batch + + +class FileBatchesWithRawResponse: + def __init__(self, file_batches: FileBatches) -> None: + self._file_batches = file_batches + + self.create = _legacy_response.to_raw_response_wrapper( + file_batches.create, + ) + self.retrieve = _legacy_response.to_raw_response_wrapper( + file_batches.retrieve, + ) + self.cancel = _legacy_response.to_raw_response_wrapper( + file_batches.cancel, + ) + self.list_files = _legacy_response.to_raw_response_wrapper( + file_batches.list_files, + ) + + +class AsyncFileBatchesWithRawResponse: + def __init__(self, file_batches: AsyncFileBatches) -> None: + self._file_batches = file_batches + + self.create = _legacy_response.async_to_raw_response_wrapper( + file_batches.create, + ) + self.retrieve = _legacy_response.async_to_raw_response_wrapper( + file_batches.retrieve, + ) + self.cancel = _legacy_response.async_to_raw_response_wrapper( + file_batches.cancel, + ) + self.list_files = _legacy_response.async_to_raw_response_wrapper( + file_batches.list_files, + ) + + +class FileBatchesWithStreamingResponse: + def __init__(self, file_batches: FileBatches) -> None: + self._file_batches = file_batches + + self.create = to_streamed_response_wrapper( + file_batches.create, + ) + self.retrieve = to_streamed_response_wrapper( + file_batches.retrieve, + ) + self.cancel = to_streamed_response_wrapper( + file_batches.cancel, + ) + self.list_files = to_streamed_response_wrapper( + file_batches.list_files, + ) + + +class AsyncFileBatchesWithStreamingResponse: + def __init__(self, file_batches: AsyncFileBatches) -> None: + self._file_batches = file_batches + + self.create = async_to_streamed_response_wrapper( + file_batches.create, + ) + self.retrieve = async_to_streamed_response_wrapper( + file_batches.retrieve, + ) + self.cancel = async_to_streamed_response_wrapper( + file_batches.cancel, + ) + self.list_files = async_to_streamed_response_wrapper( + file_batches.list_files, + ) diff --git a/.venv/lib/python3.12/site-packages/openai/resources/vector_stores/files.py b/.venv/lib/python3.12/site-packages/openai/resources/vector_stores/files.py new file mode 100644 index 00000000..7d93798a --- /dev/null +++ b/.venv/lib/python3.12/site-packages/openai/resources/vector_stores/files.py @@ -0,0 +1,933 @@ +# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details. + +from __future__ import annotations + +from typing import TYPE_CHECKING, Dict, Union, Optional +from typing_extensions import Literal, assert_never + +import httpx + +from ... import _legacy_response +from ...types import FileChunkingStrategyParam +from ..._types import NOT_GIVEN, Body, Query, Headers, NotGiven, FileTypes +from ..._utils import ( + is_given, + maybe_transform, + async_maybe_transform, +) +from ..._compat import cached_property +from ..._resource import SyncAPIResource, AsyncAPIResource +from ..._response import to_streamed_response_wrapper, async_to_streamed_response_wrapper +from ...pagination import SyncPage, AsyncPage, SyncCursorPage, AsyncCursorPage +from ..._base_client import AsyncPaginator, make_request_options +from ...types.vector_stores import file_list_params, file_create_params, file_update_params +from ...types.file_chunking_strategy_param import FileChunkingStrategyParam +from ...types.vector_stores.vector_store_file import VectorStoreFile +from ...types.vector_stores.file_content_response import FileContentResponse +from ...types.vector_stores.vector_store_file_deleted import VectorStoreFileDeleted + +__all__ = ["Files", "AsyncFiles"] + + +class Files(SyncAPIResource): + @cached_property + def with_raw_response(self) -> FilesWithRawResponse: + """ + This property can be used as a prefix for any HTTP method call to return + the raw response object instead of the parsed content. + + For more information, see https://www.github.com/openai/openai-python#accessing-raw-response-data-eg-headers + """ + return FilesWithRawResponse(self) + + @cached_property + def with_streaming_response(self) -> FilesWithStreamingResponse: + """ + An alternative to `.with_raw_response` that doesn't eagerly read the response body. + + For more information, see https://www.github.com/openai/openai-python#with_streaming_response + """ + return FilesWithStreamingResponse(self) + + def create( + self, + vector_store_id: str, + *, + file_id: str, + attributes: Optional[Dict[str, Union[str, float, bool]]] | 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, + extra_query: Query | None = None, + extra_body: Body | None = None, + timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN, + ) -> VectorStoreFile: + """ + Create a vector store file by attaching a + [File](https://platform.openai.com/docs/api-reference/files) to a + [vector store](https://platform.openai.com/docs/api-reference/vector-stores/object). + + Args: + file_id: A [File](https://platform.openai.com/docs/api-reference/files) ID that the + vector store should use. Useful for tools like `file_search` that can access + files. + + attributes: Set of 16 key-value pairs that can be attached to an object. This can be useful + for storing additional information about the object in a structured format, and + querying for objects via API or the dashboard. Keys are strings with a maximum + length of 64 characters. Values are strings with a maximum length of 512 + characters, booleans, or numbers. + + chunking_strategy: The chunking strategy used to chunk the file(s). If not set, will use the `auto` + strategy. Only applicable if `file_ids` is non-empty. + + extra_headers: Send extra headers + + extra_query: Add additional query parameters to the request + + extra_body: Add additional JSON properties to the request + + timeout: Override the client-level default timeout for this request, in seconds + """ + if not vector_store_id: + raise ValueError(f"Expected a non-empty value for `vector_store_id` but received {vector_store_id!r}") + extra_headers = {"OpenAI-Beta": "assistants=v2", **(extra_headers or {})} + return self._post( + f"/vector_stores/{vector_store_id}/files", + body=maybe_transform( + { + "file_id": file_id, + "attributes": attributes, + "chunking_strategy": chunking_strategy, + }, + file_create_params.FileCreateParams, + ), + options=make_request_options( + extra_headers=extra_headers, extra_query=extra_query, extra_body=extra_body, timeout=timeout + ), + cast_to=VectorStoreFile, + ) + + def retrieve( + self, + file_id: str, + *, + vector_store_id: str, + # 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, + extra_query: Query | None = None, + extra_body: Body | None = None, + timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN, + ) -> VectorStoreFile: + """ + Retrieves a vector store file. + + Args: + extra_headers: Send extra headers + + extra_query: Add additional query parameters to the request + + extra_body: Add additional JSON properties to the request + + timeout: Override the client-level default timeout for this request, in seconds + """ + if not vector_store_id: + raise ValueError(f"Expected a non-empty value for `vector_store_id` but received {vector_store_id!r}") + if not file_id: + raise ValueError(f"Expected a non-empty value for `file_id` but received {file_id!r}") + extra_headers = {"OpenAI-Beta": "assistants=v2", **(extra_headers or {})} + return self._get( + f"/vector_stores/{vector_store_id}/files/{file_id}", + options=make_request_options( + extra_headers=extra_headers, extra_query=extra_query, extra_body=extra_body, timeout=timeout + ), + cast_to=VectorStoreFile, + ) + + def update( + self, + file_id: str, + *, + vector_store_id: str, + attributes: Optional[Dict[str, Union[str, float, bool]]], + # 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, + extra_query: Query | None = None, + extra_body: Body | None = None, + timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN, + ) -> VectorStoreFile: + """ + Update attributes on a vector store file. + + Args: + attributes: Set of 16 key-value pairs that can be attached to an object. This can be useful + for storing additional information about the object in a structured format, and + querying for objects via API or the dashboard. Keys are strings with a maximum + length of 64 characters. Values are strings with a maximum length of 512 + characters, booleans, or numbers. + + extra_headers: Send extra headers + + extra_query: Add additional query parameters to the request + + extra_body: Add additional JSON properties to the request + + timeout: Override the client-level default timeout for this request, in seconds + """ + if not vector_store_id: + raise ValueError(f"Expected a non-empty value for `vector_store_id` but received {vector_store_id!r}") + if not file_id: + raise ValueError(f"Expected a non-empty value for `file_id` but received {file_id!r}") + extra_headers = {"OpenAI-Beta": "assistants=v2", **(extra_headers or {})} + return self._post( + f"/vector_stores/{vector_store_id}/files/{file_id}", + body=maybe_transform({"attributes": attributes}, file_update_params.FileUpdateParams), + options=make_request_options( + extra_headers=extra_headers, extra_query=extra_query, extra_body=extra_body, timeout=timeout + ), + cast_to=VectorStoreFile, + ) + + def list( + self, + vector_store_id: str, + *, + after: str | NotGiven = NOT_GIVEN, + before: str | NotGiven = NOT_GIVEN, + filter: Literal["in_progress", "completed", "failed", "cancelled"] | NotGiven = NOT_GIVEN, + limit: int | NotGiven = NOT_GIVEN, + order: Literal["asc", "desc"] | 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, + extra_query: Query | None = None, + extra_body: Body | None = None, + timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN, + ) -> SyncCursorPage[VectorStoreFile]: + """ + Returns a list of vector store files. + + Args: + after: A cursor for use in pagination. `after` is an object ID that defines your place + in the list. For instance, if you make a list request and receive 100 objects, + ending with obj_foo, your subsequent call can include after=obj_foo in order to + fetch the next page of the list. + + before: A cursor for use in pagination. `before` is an object ID that defines your place + in the list. For instance, if you make a list request and receive 100 objects, + starting with obj_foo, your subsequent call can include before=obj_foo in order + to fetch the previous page of the list. + + filter: Filter by file status. One of `in_progress`, `completed`, `failed`, `cancelled`. + + limit: A limit on the number of objects to be returned. Limit can range between 1 and + 100, and the default is 20. + + order: Sort order by the `created_at` timestamp of the objects. `asc` for ascending + order and `desc` for descending order. + + extra_headers: Send extra headers + + extra_query: Add additional query parameters to the request + + extra_body: Add additional JSON properties to the request + + timeout: Override the client-level default timeout for this request, in seconds + """ + if not vector_store_id: + raise ValueError(f"Expected a non-empty value for `vector_store_id` but received {vector_store_id!r}") + extra_headers = {"OpenAI-Beta": "assistants=v2", **(extra_headers or {})} + return self._get_api_list( + f"/vector_stores/{vector_store_id}/files", + page=SyncCursorPage[VectorStoreFile], + options=make_request_options( + extra_headers=extra_headers, + extra_query=extra_query, + extra_body=extra_body, + timeout=timeout, + query=maybe_transform( + { + "after": after, + "before": before, + "filter": filter, + "limit": limit, + "order": order, + }, + file_list_params.FileListParams, + ), + ), + model=VectorStoreFile, + ) + + def delete( + self, + file_id: str, + *, + vector_store_id: str, + # 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, + extra_query: Query | None = None, + extra_body: Body | None = None, + timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN, + ) -> VectorStoreFileDeleted: + """Delete a vector store file. + + This will remove the file from the vector store but + the file itself will not be deleted. To delete the file, use the + [delete file](https://platform.openai.com/docs/api-reference/files/delete) + endpoint. + + Args: + extra_headers: Send extra headers + + extra_query: Add additional query parameters to the request + + extra_body: Add additional JSON properties to the request + + timeout: Override the client-level default timeout for this request, in seconds + """ + if not vector_store_id: + raise ValueError(f"Expected a non-empty value for `vector_store_id` but received {vector_store_id!r}") + if not file_id: + raise ValueError(f"Expected a non-empty value for `file_id` but received {file_id!r}") + extra_headers = {"OpenAI-Beta": "assistants=v2", **(extra_headers or {})} + return self._delete( + f"/vector_stores/{vector_store_id}/files/{file_id}", + options=make_request_options( + extra_headers=extra_headers, extra_query=extra_query, extra_body=extra_body, timeout=timeout + ), + cast_to=VectorStoreFileDeleted, + ) + + def create_and_poll( + self, + file_id: str, + *, + vector_store_id: str, + poll_interval_ms: int | 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) + + return self.poll( + file_id, + vector_store_id=vector_store_id, + poll_interval_ms=poll_interval_ms, + ) + + def poll( + self, + file_id: str, + *, + vector_store_id: str, + poll_interval_ms: int | NotGiven = NOT_GIVEN, + ) -> VectorStoreFile: + """Wait for the vector store file to finish processing. + + Note: this will return even if the file failed to process, you need to check + file.last_error and file.status to handle these cases + """ + headers: dict[str, str] = {"X-Stainless-Poll-Helper": "true"} + if is_given(poll_interval_ms): + headers["X-Stainless-Custom-Poll-Interval"] = str(poll_interval_ms) + + while True: + response = self.with_raw_response.retrieve( + file_id, + vector_store_id=vector_store_id, + extra_headers=headers, + ) + + file = response.parse() + if file.status == "in_progress": + if not is_given(poll_interval_ms): + from_header = response.headers.get("openai-poll-after-ms") + if from_header is not None: + poll_interval_ms = int(from_header) + else: + poll_interval_ms = 1000 + + self._sleep(poll_interval_ms / 1000) + elif file.status == "cancelled" or file.status == "completed" or file.status == "failed": + return file + else: + if TYPE_CHECKING: # type: ignore[unreachable] + assert_never(file.status) + else: + return file + + def upload( + self, + *, + vector_store_id: str, + file: FileTypes, + chunking_strategy: FileChunkingStrategyParam | NotGiven = NOT_GIVEN, + ) -> VectorStoreFile: + """Upload a file to the `files` API and then attach it to the given vector store. + + Note the file will be asynchronously processed (you can use the alternative + polling helper method to wait for processing to complete). + """ + file_obj = self._client.files.create(file=file, purpose="assistants") + return self.create(vector_store_id=vector_store_id, file_id=file_obj.id, chunking_strategy=chunking_strategy) + + def upload_and_poll( + self, + *, + vector_store_id: str, + file: FileTypes, + poll_interval_ms: int | 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") + return self.create_and_poll( + vector_store_id=vector_store_id, + file_id=file_obj.id, + chunking_strategy=chunking_strategy, + poll_interval_ms=poll_interval_ms, + ) + + def content( + self, + file_id: str, + *, + vector_store_id: str, + # 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, + extra_query: Query | None = None, + extra_body: Body | None = None, + timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN, + ) -> SyncPage[FileContentResponse]: + """ + Retrieve the parsed contents of a vector store file. + + Args: + extra_headers: Send extra headers + + extra_query: Add additional query parameters to the request + + extra_body: Add additional JSON properties to the request + + timeout: Override the client-level default timeout for this request, in seconds + """ + if not vector_store_id: + raise ValueError(f"Expected a non-empty value for `vector_store_id` but received {vector_store_id!r}") + if not file_id: + raise ValueError(f"Expected a non-empty value for `file_id` but received {file_id!r}") + extra_headers = {"OpenAI-Beta": "assistants=v2", **(extra_headers or {})} + return self._get_api_list( + f"/vector_stores/{vector_store_id}/files/{file_id}/content", + page=SyncPage[FileContentResponse], + options=make_request_options( + extra_headers=extra_headers, extra_query=extra_query, extra_body=extra_body, timeout=timeout + ), + model=FileContentResponse, + ) + + +class AsyncFiles(AsyncAPIResource): + @cached_property + def with_raw_response(self) -> AsyncFilesWithRawResponse: + """ + This property can be used as a prefix for any HTTP method call to return + the raw response object instead of the parsed content. + + For more information, see https://www.github.com/openai/openai-python#accessing-raw-response-data-eg-headers + """ + return AsyncFilesWithRawResponse(self) + + @cached_property + def with_streaming_response(self) -> AsyncFilesWithStreamingResponse: + """ + An alternative to `.with_raw_response` that doesn't eagerly read the response body. + + For more information, see https://www.github.com/openai/openai-python#with_streaming_response + """ + return AsyncFilesWithStreamingResponse(self) + + async def create( + self, + vector_store_id: str, + *, + file_id: str, + attributes: Optional[Dict[str, Union[str, float, bool]]] | 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, + extra_query: Query | None = None, + extra_body: Body | None = None, + timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN, + ) -> VectorStoreFile: + """ + Create a vector store file by attaching a + [File](https://platform.openai.com/docs/api-reference/files) to a + [vector store](https://platform.openai.com/docs/api-reference/vector-stores/object). + + Args: + file_id: A [File](https://platform.openai.com/docs/api-reference/files) ID that the + vector store should use. Useful for tools like `file_search` that can access + files. + + attributes: Set of 16 key-value pairs that can be attached to an object. This can be useful + for storing additional information about the object in a structured format, and + querying for objects via API or the dashboard. Keys are strings with a maximum + length of 64 characters. Values are strings with a maximum length of 512 + characters, booleans, or numbers. + + chunking_strategy: The chunking strategy used to chunk the file(s). If not set, will use the `auto` + strategy. Only applicable if `file_ids` is non-empty. + + extra_headers: Send extra headers + + extra_query: Add additional query parameters to the request + + extra_body: Add additional JSON properties to the request + + timeout: Override the client-level default timeout for this request, in seconds + """ + if not vector_store_id: + raise ValueError(f"Expected a non-empty value for `vector_store_id` but received {vector_store_id!r}") + extra_headers = {"OpenAI-Beta": "assistants=v2", **(extra_headers or {})} + return await self._post( + f"/vector_stores/{vector_store_id}/files", + body=await async_maybe_transform( + { + "file_id": file_id, + "attributes": attributes, + "chunking_strategy": chunking_strategy, + }, + file_create_params.FileCreateParams, + ), + options=make_request_options( + extra_headers=extra_headers, extra_query=extra_query, extra_body=extra_body, timeout=timeout + ), + cast_to=VectorStoreFile, + ) + + async def retrieve( + self, + file_id: str, + *, + vector_store_id: str, + # 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, + extra_query: Query | None = None, + extra_body: Body | None = None, + timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN, + ) -> VectorStoreFile: + """ + Retrieves a vector store file. + + Args: + extra_headers: Send extra headers + + extra_query: Add additional query parameters to the request + + extra_body: Add additional JSON properties to the request + + timeout: Override the client-level default timeout for this request, in seconds + """ + if not vector_store_id: + raise ValueError(f"Expected a non-empty value for `vector_store_id` but received {vector_store_id!r}") + if not file_id: + raise ValueError(f"Expected a non-empty value for `file_id` but received {file_id!r}") + extra_headers = {"OpenAI-Beta": "assistants=v2", **(extra_headers or {})} + return await self._get( + f"/vector_stores/{vector_store_id}/files/{file_id}", + options=make_request_options( + extra_headers=extra_headers, extra_query=extra_query, extra_body=extra_body, timeout=timeout + ), + cast_to=VectorStoreFile, + ) + + async def update( + self, + file_id: str, + *, + vector_store_id: str, + attributes: Optional[Dict[str, Union[str, float, bool]]], + # 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, + extra_query: Query | None = None, + extra_body: Body | None = None, + timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN, + ) -> VectorStoreFile: + """ + Update attributes on a vector store file. + + Args: + attributes: Set of 16 key-value pairs that can be attached to an object. This can be useful + for storing additional information about the object in a structured format, and + querying for objects via API or the dashboard. Keys are strings with a maximum + length of 64 characters. Values are strings with a maximum length of 512 + characters, booleans, or numbers. + + extra_headers: Send extra headers + + extra_query: Add additional query parameters to the request + + extra_body: Add additional JSON properties to the request + + timeout: Override the client-level default timeout for this request, in seconds + """ + if not vector_store_id: + raise ValueError(f"Expected a non-empty value for `vector_store_id` but received {vector_store_id!r}") + if not file_id: + raise ValueError(f"Expected a non-empty value for `file_id` but received {file_id!r}") + extra_headers = {"OpenAI-Beta": "assistants=v2", **(extra_headers or {})} + return await self._post( + f"/vector_stores/{vector_store_id}/files/{file_id}", + body=await async_maybe_transform({"attributes": attributes}, file_update_params.FileUpdateParams), + options=make_request_options( + extra_headers=extra_headers, extra_query=extra_query, extra_body=extra_body, timeout=timeout + ), + cast_to=VectorStoreFile, + ) + + def list( + self, + vector_store_id: str, + *, + after: str | NotGiven = NOT_GIVEN, + before: str | NotGiven = NOT_GIVEN, + filter: Literal["in_progress", "completed", "failed", "cancelled"] | NotGiven = NOT_GIVEN, + limit: int | NotGiven = NOT_GIVEN, + order: Literal["asc", "desc"] | 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, + extra_query: Query | None = None, + extra_body: Body | None = None, + timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN, + ) -> AsyncPaginator[VectorStoreFile, AsyncCursorPage[VectorStoreFile]]: + """ + Returns a list of vector store files. + + Args: + after: A cursor for use in pagination. `after` is an object ID that defines your place + in the list. For instance, if you make a list request and receive 100 objects, + ending with obj_foo, your subsequent call can include after=obj_foo in order to + fetch the next page of the list. + + before: A cursor for use in pagination. `before` is an object ID that defines your place + in the list. For instance, if you make a list request and receive 100 objects, + starting with obj_foo, your subsequent call can include before=obj_foo in order + to fetch the previous page of the list. + + filter: Filter by file status. One of `in_progress`, `completed`, `failed`, `cancelled`. + + limit: A limit on the number of objects to be returned. Limit can range between 1 and + 100, and the default is 20. + + order: Sort order by the `created_at` timestamp of the objects. `asc` for ascending + order and `desc` for descending order. + + extra_headers: Send extra headers + + extra_query: Add additional query parameters to the request + + extra_body: Add additional JSON properties to the request + + timeout: Override the client-level default timeout for this request, in seconds + """ + if not vector_store_id: + raise ValueError(f"Expected a non-empty value for `vector_store_id` but received {vector_store_id!r}") + extra_headers = {"OpenAI-Beta": "assistants=v2", **(extra_headers or {})} + return self._get_api_list( + f"/vector_stores/{vector_store_id}/files", + page=AsyncCursorPage[VectorStoreFile], + options=make_request_options( + extra_headers=extra_headers, + extra_query=extra_query, + extra_body=extra_body, + timeout=timeout, + query=maybe_transform( + { + "after": after, + "before": before, + "filter": filter, + "limit": limit, + "order": order, + }, + file_list_params.FileListParams, + ), + ), + model=VectorStoreFile, + ) + + async def delete( + self, + file_id: str, + *, + vector_store_id: str, + # 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, + extra_query: Query | None = None, + extra_body: Body | None = None, + timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN, + ) -> VectorStoreFileDeleted: + """Delete a vector store file. + + This will remove the file from the vector store but + the file itself will not be deleted. To delete the file, use the + [delete file](https://platform.openai.com/docs/api-reference/files/delete) + endpoint. + + Args: + extra_headers: Send extra headers + + extra_query: Add additional query parameters to the request + + extra_body: Add additional JSON properties to the request + + timeout: Override the client-level default timeout for this request, in seconds + """ + if not vector_store_id: + raise ValueError(f"Expected a non-empty value for `vector_store_id` but received {vector_store_id!r}") + if not file_id: + raise ValueError(f"Expected a non-empty value for `file_id` but received {file_id!r}") + extra_headers = {"OpenAI-Beta": "assistants=v2", **(extra_headers or {})} + return await self._delete( + f"/vector_stores/{vector_store_id}/files/{file_id}", + options=make_request_options( + extra_headers=extra_headers, extra_query=extra_query, extra_body=extra_body, timeout=timeout + ), + cast_to=VectorStoreFileDeleted, + ) + + async def create_and_poll( + self, + file_id: str, + *, + vector_store_id: str, + poll_interval_ms: int | 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) + + return await self.poll( + file_id, + vector_store_id=vector_store_id, + poll_interval_ms=poll_interval_ms, + ) + + async def poll( + self, + file_id: str, + *, + vector_store_id: str, + poll_interval_ms: int | NotGiven = NOT_GIVEN, + ) -> VectorStoreFile: + """Wait for the vector store file to finish processing. + + Note: this will return even if the file failed to process, you need to check + file.last_error and file.status to handle these cases + """ + headers: dict[str, str] = {"X-Stainless-Poll-Helper": "true"} + if is_given(poll_interval_ms): + headers["X-Stainless-Custom-Poll-Interval"] = str(poll_interval_ms) + + while True: + response = await self.with_raw_response.retrieve( + file_id, + vector_store_id=vector_store_id, + extra_headers=headers, + ) + + file = response.parse() + if file.status == "in_progress": + if not is_given(poll_interval_ms): + from_header = response.headers.get("openai-poll-after-ms") + if from_header is not None: + poll_interval_ms = int(from_header) + else: + poll_interval_ms = 1000 + + await self._sleep(poll_interval_ms / 1000) + elif file.status == "cancelled" or file.status == "completed" or file.status == "failed": + return file + else: + if TYPE_CHECKING: # type: ignore[unreachable] + assert_never(file.status) + else: + return file + + async def upload( + self, + *, + vector_store_id: str, + file: FileTypes, + chunking_strategy: FileChunkingStrategyParam | NotGiven = NOT_GIVEN, + ) -> VectorStoreFile: + """Upload a file to the `files` API and then attach it to the given vector store. + + Note the file will be asynchronously processed (you can use the alternative + polling helper method to wait for processing to complete). + """ + file_obj = await self._client.files.create(file=file, purpose="assistants") + return await self.create( + vector_store_id=vector_store_id, file_id=file_obj.id, chunking_strategy=chunking_strategy + ) + + async def upload_and_poll( + self, + *, + vector_store_id: str, + file: FileTypes, + poll_interval_ms: int | 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") + return await self.create_and_poll( + vector_store_id=vector_store_id, + file_id=file_obj.id, + poll_interval_ms=poll_interval_ms, + chunking_strategy=chunking_strategy, + ) + + def content( + self, + file_id: str, + *, + vector_store_id: str, + # 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, + extra_query: Query | None = None, + extra_body: Body | None = None, + timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN, + ) -> AsyncPaginator[FileContentResponse, AsyncPage[FileContentResponse]]: + """ + Retrieve the parsed contents of a vector store file. + + Args: + extra_headers: Send extra headers + + extra_query: Add additional query parameters to the request + + extra_body: Add additional JSON properties to the request + + timeout: Override the client-level default timeout for this request, in seconds + """ + if not vector_store_id: + raise ValueError(f"Expected a non-empty value for `vector_store_id` but received {vector_store_id!r}") + if not file_id: + raise ValueError(f"Expected a non-empty value for `file_id` but received {file_id!r}") + extra_headers = {"OpenAI-Beta": "assistants=v2", **(extra_headers or {})} + return self._get_api_list( + f"/vector_stores/{vector_store_id}/files/{file_id}/content", + page=AsyncPage[FileContentResponse], + options=make_request_options( + extra_headers=extra_headers, extra_query=extra_query, extra_body=extra_body, timeout=timeout + ), + model=FileContentResponse, + ) + + +class FilesWithRawResponse: + def __init__(self, files: Files) -> None: + self._files = files + + self.create = _legacy_response.to_raw_response_wrapper( + files.create, + ) + self.retrieve = _legacy_response.to_raw_response_wrapper( + files.retrieve, + ) + self.update = _legacy_response.to_raw_response_wrapper( + files.update, + ) + self.list = _legacy_response.to_raw_response_wrapper( + files.list, + ) + self.delete = _legacy_response.to_raw_response_wrapper( + files.delete, + ) + self.content = _legacy_response.to_raw_response_wrapper( + files.content, + ) + + +class AsyncFilesWithRawResponse: + def __init__(self, files: AsyncFiles) -> None: + self._files = files + + self.create = _legacy_response.async_to_raw_response_wrapper( + files.create, + ) + self.retrieve = _legacy_response.async_to_raw_response_wrapper( + files.retrieve, + ) + self.update = _legacy_response.async_to_raw_response_wrapper( + files.update, + ) + self.list = _legacy_response.async_to_raw_response_wrapper( + files.list, + ) + self.delete = _legacy_response.async_to_raw_response_wrapper( + files.delete, + ) + self.content = _legacy_response.async_to_raw_response_wrapper( + files.content, + ) + + +class FilesWithStreamingResponse: + def __init__(self, files: Files) -> None: + self._files = files + + self.create = to_streamed_response_wrapper( + files.create, + ) + self.retrieve = to_streamed_response_wrapper( + files.retrieve, + ) + self.update = to_streamed_response_wrapper( + files.update, + ) + self.list = to_streamed_response_wrapper( + files.list, + ) + self.delete = to_streamed_response_wrapper( + files.delete, + ) + self.content = to_streamed_response_wrapper( + files.content, + ) + + +class AsyncFilesWithStreamingResponse: + def __init__(self, files: AsyncFiles) -> None: + self._files = files + + self.create = async_to_streamed_response_wrapper( + files.create, + ) + self.retrieve = async_to_streamed_response_wrapper( + files.retrieve, + ) + self.update = async_to_streamed_response_wrapper( + files.update, + ) + self.list = async_to_streamed_response_wrapper( + files.list, + ) + self.delete = async_to_streamed_response_wrapper( + files.delete, + ) + self.content = async_to_streamed_response_wrapper( + files.content, + ) diff --git a/.venv/lib/python3.12/site-packages/openai/resources/vector_stores/vector_stores.py b/.venv/lib/python3.12/site-packages/openai/resources/vector_stores/vector_stores.py new file mode 100644 index 00000000..aaa6ed27 --- /dev/null +++ b/.venv/lib/python3.12/site-packages/openai/resources/vector_stores/vector_stores.py @@ -0,0 +1,868 @@ +# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details. + +from __future__ import annotations + +from typing import List, Union, Optional +from typing_extensions import Literal + +import httpx + +from ... import _legacy_response +from .files import ( + Files, + AsyncFiles, + FilesWithRawResponse, + AsyncFilesWithRawResponse, + FilesWithStreamingResponse, + AsyncFilesWithStreamingResponse, +) +from ...types import ( + FileChunkingStrategyParam, + vector_store_list_params, + vector_store_create_params, + vector_store_search_params, + vector_store_update_params, +) +from ..._types import NOT_GIVEN, Body, Query, Headers, NotGiven +from ..._utils import ( + maybe_transform, + async_maybe_transform, +) +from ..._compat import cached_property +from ..._resource import SyncAPIResource, AsyncAPIResource +from ..._response import to_streamed_response_wrapper, async_to_streamed_response_wrapper +from ...pagination import SyncPage, AsyncPage, SyncCursorPage, AsyncCursorPage +from .file_batches import ( + FileBatches, + AsyncFileBatches, + FileBatchesWithRawResponse, + AsyncFileBatchesWithRawResponse, + FileBatchesWithStreamingResponse, + AsyncFileBatchesWithStreamingResponse, +) +from ..._base_client import AsyncPaginator, make_request_options +from ...types.vector_store import VectorStore +from ...types.vector_store_deleted import VectorStoreDeleted +from ...types.shared_params.metadata import Metadata +from ...types.file_chunking_strategy_param import FileChunkingStrategyParam +from ...types.vector_store_search_response import VectorStoreSearchResponse + +__all__ = ["VectorStores", "AsyncVectorStores"] + + +class VectorStores(SyncAPIResource): + @cached_property + def files(self) -> Files: + return Files(self._client) + + @cached_property + def file_batches(self) -> FileBatches: + return FileBatches(self._client) + + @cached_property + def with_raw_response(self) -> VectorStoresWithRawResponse: + """ + This property can be used as a prefix for any HTTP method call to return + the raw response object instead of the parsed content. + + For more information, see https://www.github.com/openai/openai-python#accessing-raw-response-data-eg-headers + """ + return VectorStoresWithRawResponse(self) + + @cached_property + def with_streaming_response(self) -> VectorStoresWithStreamingResponse: + """ + An alternative to `.with_raw_response` that doesn't eagerly read the response body. + + For more information, see https://www.github.com/openai/openai-python#with_streaming_response + """ + return VectorStoresWithStreamingResponse(self) + + def create( + self, + *, + 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[Metadata] | NotGiven = NOT_GIVEN, + name: str | 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, + extra_query: Query | None = None, + extra_body: Body | None = None, + timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN, + ) -> VectorStore: + """ + Create a vector store. + + Args: + chunking_strategy: The chunking strategy used to chunk the file(s). If not set, will use the `auto` + strategy. Only applicable if `file_ids` is non-empty. + + expires_after: The expiration policy for a vector store. + + file_ids: A list of [File](https://platform.openai.com/docs/api-reference/files) IDs that + the vector store should use. Useful for tools like `file_search` that can access + files. + + metadata: Set of 16 key-value pairs that can be attached to an object. This can be useful + for storing additional information about the object in a structured format, and + querying for objects via API or the dashboard. + + Keys are strings with a maximum length of 64 characters. Values are strings with + a maximum length of 512 characters. + + name: The name of the vector store. + + extra_headers: Send extra headers + + extra_query: Add additional query parameters to the request + + extra_body: Add additional JSON properties to the request + + timeout: Override the client-level default timeout for this request, in seconds + """ + extra_headers = {"OpenAI-Beta": "assistants=v2", **(extra_headers or {})} + return self._post( + "/vector_stores", + body=maybe_transform( + { + "chunking_strategy": chunking_strategy, + "expires_after": expires_after, + "file_ids": file_ids, + "metadata": metadata, + "name": name, + }, + vector_store_create_params.VectorStoreCreateParams, + ), + options=make_request_options( + extra_headers=extra_headers, extra_query=extra_query, extra_body=extra_body, timeout=timeout + ), + cast_to=VectorStore, + ) + + def retrieve( + self, + vector_store_id: str, + *, + # 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, + extra_query: Query | None = None, + extra_body: Body | None = None, + timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN, + ) -> VectorStore: + """ + Retrieves a vector store. + + Args: + extra_headers: Send extra headers + + extra_query: Add additional query parameters to the request + + extra_body: Add additional JSON properties to the request + + timeout: Override the client-level default timeout for this request, in seconds + """ + if not vector_store_id: + raise ValueError(f"Expected a non-empty value for `vector_store_id` but received {vector_store_id!r}") + extra_headers = {"OpenAI-Beta": "assistants=v2", **(extra_headers or {})} + return self._get( + f"/vector_stores/{vector_store_id}", + options=make_request_options( + extra_headers=extra_headers, extra_query=extra_query, extra_body=extra_body, timeout=timeout + ), + cast_to=VectorStore, + ) + + def update( + self, + vector_store_id: str, + *, + expires_after: Optional[vector_store_update_params.ExpiresAfter] | NotGiven = NOT_GIVEN, + metadata: Optional[Metadata] | NotGiven = NOT_GIVEN, + name: Optional[str] | 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, + extra_query: Query | None = None, + extra_body: Body | None = None, + timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN, + ) -> VectorStore: + """ + Modifies a vector store. + + Args: + expires_after: The expiration policy for a vector store. + + metadata: Set of 16 key-value pairs that can be attached to an object. This can be useful + for storing additional information about the object in a structured format, and + querying for objects via API or the dashboard. + + Keys are strings with a maximum length of 64 characters. Values are strings with + a maximum length of 512 characters. + + name: The name of the vector store. + + extra_headers: Send extra headers + + extra_query: Add additional query parameters to the request + + extra_body: Add additional JSON properties to the request + + timeout: Override the client-level default timeout for this request, in seconds + """ + if not vector_store_id: + raise ValueError(f"Expected a non-empty value for `vector_store_id` but received {vector_store_id!r}") + extra_headers = {"OpenAI-Beta": "assistants=v2", **(extra_headers or {})} + return self._post( + f"/vector_stores/{vector_store_id}", + body=maybe_transform( + { + "expires_after": expires_after, + "metadata": metadata, + "name": name, + }, + vector_store_update_params.VectorStoreUpdateParams, + ), + options=make_request_options( + extra_headers=extra_headers, extra_query=extra_query, extra_body=extra_body, timeout=timeout + ), + cast_to=VectorStore, + ) + + def list( + self, + *, + after: str | NotGiven = NOT_GIVEN, + before: str | NotGiven = NOT_GIVEN, + limit: int | NotGiven = NOT_GIVEN, + order: Literal["asc", "desc"] | 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, + extra_query: Query | None = None, + extra_body: Body | None = None, + timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN, + ) -> SyncCursorPage[VectorStore]: + """Returns a list of vector stores. + + Args: + after: A cursor for use in pagination. + + `after` is an object ID that defines your place + in the list. For instance, if you make a list request and receive 100 objects, + ending with obj_foo, your subsequent call can include after=obj_foo in order to + fetch the next page of the list. + + before: A cursor for use in pagination. `before` is an object ID that defines your place + in the list. For instance, if you make a list request and receive 100 objects, + starting with obj_foo, your subsequent call can include before=obj_foo in order + to fetch the previous page of the list. + + limit: A limit on the number of objects to be returned. Limit can range between 1 and + 100, and the default is 20. + + order: Sort order by the `created_at` timestamp of the objects. `asc` for ascending + order and `desc` for descending order. + + extra_headers: Send extra headers + + extra_query: Add additional query parameters to the request + + extra_body: Add additional JSON properties to the request + + timeout: Override the client-level default timeout for this request, in seconds + """ + extra_headers = {"OpenAI-Beta": "assistants=v2", **(extra_headers or {})} + return self._get_api_list( + "/vector_stores", + page=SyncCursorPage[VectorStore], + options=make_request_options( + extra_headers=extra_headers, + extra_query=extra_query, + extra_body=extra_body, + timeout=timeout, + query=maybe_transform( + { + "after": after, + "before": before, + "limit": limit, + "order": order, + }, + vector_store_list_params.VectorStoreListParams, + ), + ), + model=VectorStore, + ) + + def delete( + self, + vector_store_id: str, + *, + # 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, + extra_query: Query | None = None, + extra_body: Body | None = None, + timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN, + ) -> VectorStoreDeleted: + """ + Delete a vector store. + + Args: + extra_headers: Send extra headers + + extra_query: Add additional query parameters to the request + + extra_body: Add additional JSON properties to the request + + timeout: Override the client-level default timeout for this request, in seconds + """ + if not vector_store_id: + raise ValueError(f"Expected a non-empty value for `vector_store_id` but received {vector_store_id!r}") + extra_headers = {"OpenAI-Beta": "assistants=v2", **(extra_headers or {})} + return self._delete( + f"/vector_stores/{vector_store_id}", + options=make_request_options( + extra_headers=extra_headers, extra_query=extra_query, extra_body=extra_body, timeout=timeout + ), + cast_to=VectorStoreDeleted, + ) + + def search( + self, + vector_store_id: str, + *, + query: Union[str, List[str]], + filters: vector_store_search_params.Filters | NotGiven = NOT_GIVEN, + max_num_results: int | NotGiven = NOT_GIVEN, + ranking_options: vector_store_search_params.RankingOptions | NotGiven = NOT_GIVEN, + rewrite_query: bool | 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, + extra_query: Query | None = None, + extra_body: Body | None = None, + timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN, + ) -> SyncPage[VectorStoreSearchResponse]: + """ + Search a vector store for relevant chunks based on a query and file attributes + filter. + + Args: + query: A query string for a search + + filters: A filter to apply based on file attributes. + + max_num_results: The maximum number of results to return. This number should be between 1 and 50 + inclusive. + + ranking_options: Ranking options for search. + + rewrite_query: Whether to rewrite the natural language query for vector search. + + extra_headers: Send extra headers + + extra_query: Add additional query parameters to the request + + extra_body: Add additional JSON properties to the request + + timeout: Override the client-level default timeout for this request, in seconds + """ + if not vector_store_id: + raise ValueError(f"Expected a non-empty value for `vector_store_id` but received {vector_store_id!r}") + extra_headers = {"OpenAI-Beta": "assistants=v2", **(extra_headers or {})} + return self._get_api_list( + f"/vector_stores/{vector_store_id}/search", + page=SyncPage[VectorStoreSearchResponse], + body=maybe_transform( + { + "query": query, + "filters": filters, + "max_num_results": max_num_results, + "ranking_options": ranking_options, + "rewrite_query": rewrite_query, + }, + vector_store_search_params.VectorStoreSearchParams, + ), + options=make_request_options( + extra_headers=extra_headers, extra_query=extra_query, extra_body=extra_body, timeout=timeout + ), + model=VectorStoreSearchResponse, + method="post", + ) + + +class AsyncVectorStores(AsyncAPIResource): + @cached_property + def files(self) -> AsyncFiles: + return AsyncFiles(self._client) + + @cached_property + def file_batches(self) -> AsyncFileBatches: + return AsyncFileBatches(self._client) + + @cached_property + def with_raw_response(self) -> AsyncVectorStoresWithRawResponse: + """ + This property can be used as a prefix for any HTTP method call to return + the raw response object instead of the parsed content. + + For more information, see https://www.github.com/openai/openai-python#accessing-raw-response-data-eg-headers + """ + return AsyncVectorStoresWithRawResponse(self) + + @cached_property + def with_streaming_response(self) -> AsyncVectorStoresWithStreamingResponse: + """ + An alternative to `.with_raw_response` that doesn't eagerly read the response body. + + For more information, see https://www.github.com/openai/openai-python#with_streaming_response + """ + return AsyncVectorStoresWithStreamingResponse(self) + + async def create( + self, + *, + 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[Metadata] | NotGiven = NOT_GIVEN, + name: str | 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, + extra_query: Query | None = None, + extra_body: Body | None = None, + timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN, + ) -> VectorStore: + """ + Create a vector store. + + Args: + chunking_strategy: The chunking strategy used to chunk the file(s). If not set, will use the `auto` + strategy. Only applicable if `file_ids` is non-empty. + + expires_after: The expiration policy for a vector store. + + file_ids: A list of [File](https://platform.openai.com/docs/api-reference/files) IDs that + the vector store should use. Useful for tools like `file_search` that can access + files. + + metadata: Set of 16 key-value pairs that can be attached to an object. This can be useful + for storing additional information about the object in a structured format, and + querying for objects via API or the dashboard. + + Keys are strings with a maximum length of 64 characters. Values are strings with + a maximum length of 512 characters. + + name: The name of the vector store. + + extra_headers: Send extra headers + + extra_query: Add additional query parameters to the request + + extra_body: Add additional JSON properties to the request + + timeout: Override the client-level default timeout for this request, in seconds + """ + extra_headers = {"OpenAI-Beta": "assistants=v2", **(extra_headers or {})} + return await self._post( + "/vector_stores", + body=await async_maybe_transform( + { + "chunking_strategy": chunking_strategy, + "expires_after": expires_after, + "file_ids": file_ids, + "metadata": metadata, + "name": name, + }, + vector_store_create_params.VectorStoreCreateParams, + ), + options=make_request_options( + extra_headers=extra_headers, extra_query=extra_query, extra_body=extra_body, timeout=timeout + ), + cast_to=VectorStore, + ) + + async def retrieve( + self, + vector_store_id: str, + *, + # 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, + extra_query: Query | None = None, + extra_body: Body | None = None, + timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN, + ) -> VectorStore: + """ + Retrieves a vector store. + + Args: + extra_headers: Send extra headers + + extra_query: Add additional query parameters to the request + + extra_body: Add additional JSON properties to the request + + timeout: Override the client-level default timeout for this request, in seconds + """ + if not vector_store_id: + raise ValueError(f"Expected a non-empty value for `vector_store_id` but received {vector_store_id!r}") + extra_headers = {"OpenAI-Beta": "assistants=v2", **(extra_headers or {})} + return await self._get( + f"/vector_stores/{vector_store_id}", + options=make_request_options( + extra_headers=extra_headers, extra_query=extra_query, extra_body=extra_body, timeout=timeout + ), + cast_to=VectorStore, + ) + + async def update( + self, + vector_store_id: str, + *, + expires_after: Optional[vector_store_update_params.ExpiresAfter] | NotGiven = NOT_GIVEN, + metadata: Optional[Metadata] | NotGiven = NOT_GIVEN, + name: Optional[str] | 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, + extra_query: Query | None = None, + extra_body: Body | None = None, + timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN, + ) -> VectorStore: + """ + Modifies a vector store. + + Args: + expires_after: The expiration policy for a vector store. + + metadata: Set of 16 key-value pairs that can be attached to an object. This can be useful + for storing additional information about the object in a structured format, and + querying for objects via API or the dashboard. + + Keys are strings with a maximum length of 64 characters. Values are strings with + a maximum length of 512 characters. + + name: The name of the vector store. + + extra_headers: Send extra headers + + extra_query: Add additional query parameters to the request + + extra_body: Add additional JSON properties to the request + + timeout: Override the client-level default timeout for this request, in seconds + """ + if not vector_store_id: + raise ValueError(f"Expected a non-empty value for `vector_store_id` but received {vector_store_id!r}") + extra_headers = {"OpenAI-Beta": "assistants=v2", **(extra_headers or {})} + return await self._post( + f"/vector_stores/{vector_store_id}", + body=await async_maybe_transform( + { + "expires_after": expires_after, + "metadata": metadata, + "name": name, + }, + vector_store_update_params.VectorStoreUpdateParams, + ), + options=make_request_options( + extra_headers=extra_headers, extra_query=extra_query, extra_body=extra_body, timeout=timeout + ), + cast_to=VectorStore, + ) + + def list( + self, + *, + after: str | NotGiven = NOT_GIVEN, + before: str | NotGiven = NOT_GIVEN, + limit: int | NotGiven = NOT_GIVEN, + order: Literal["asc", "desc"] | 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, + extra_query: Query | None = None, + extra_body: Body | None = None, + timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN, + ) -> AsyncPaginator[VectorStore, AsyncCursorPage[VectorStore]]: + """Returns a list of vector stores. + + Args: + after: A cursor for use in pagination. + + `after` is an object ID that defines your place + in the list. For instance, if you make a list request and receive 100 objects, + ending with obj_foo, your subsequent call can include after=obj_foo in order to + fetch the next page of the list. + + before: A cursor for use in pagination. `before` is an object ID that defines your place + in the list. For instance, if you make a list request and receive 100 objects, + starting with obj_foo, your subsequent call can include before=obj_foo in order + to fetch the previous page of the list. + + limit: A limit on the number of objects to be returned. Limit can range between 1 and + 100, and the default is 20. + + order: Sort order by the `created_at` timestamp of the objects. `asc` for ascending + order and `desc` for descending order. + + extra_headers: Send extra headers + + extra_query: Add additional query parameters to the request + + extra_body: Add additional JSON properties to the request + + timeout: Override the client-level default timeout for this request, in seconds + """ + extra_headers = {"OpenAI-Beta": "assistants=v2", **(extra_headers or {})} + return self._get_api_list( + "/vector_stores", + page=AsyncCursorPage[VectorStore], + options=make_request_options( + extra_headers=extra_headers, + extra_query=extra_query, + extra_body=extra_body, + timeout=timeout, + query=maybe_transform( + { + "after": after, + "before": before, + "limit": limit, + "order": order, + }, + vector_store_list_params.VectorStoreListParams, + ), + ), + model=VectorStore, + ) + + async def delete( + self, + vector_store_id: str, + *, + # 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, + extra_query: Query | None = None, + extra_body: Body | None = None, + timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN, + ) -> VectorStoreDeleted: + """ + Delete a vector store. + + Args: + extra_headers: Send extra headers + + extra_query: Add additional query parameters to the request + + extra_body: Add additional JSON properties to the request + + timeout: Override the client-level default timeout for this request, in seconds + """ + if not vector_store_id: + raise ValueError(f"Expected a non-empty value for `vector_store_id` but received {vector_store_id!r}") + extra_headers = {"OpenAI-Beta": "assistants=v2", **(extra_headers or {})} + return await self._delete( + f"/vector_stores/{vector_store_id}", + options=make_request_options( + extra_headers=extra_headers, extra_query=extra_query, extra_body=extra_body, timeout=timeout + ), + cast_to=VectorStoreDeleted, + ) + + def search( + self, + vector_store_id: str, + *, + query: Union[str, List[str]], + filters: vector_store_search_params.Filters | NotGiven = NOT_GIVEN, + max_num_results: int | NotGiven = NOT_GIVEN, + ranking_options: vector_store_search_params.RankingOptions | NotGiven = NOT_GIVEN, + rewrite_query: bool | 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, + extra_query: Query | None = None, + extra_body: Body | None = None, + timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN, + ) -> AsyncPaginator[VectorStoreSearchResponse, AsyncPage[VectorStoreSearchResponse]]: + """ + Search a vector store for relevant chunks based on a query and file attributes + filter. + + Args: + query: A query string for a search + + filters: A filter to apply based on file attributes. + + max_num_results: The maximum number of results to return. This number should be between 1 and 50 + inclusive. + + ranking_options: Ranking options for search. + + rewrite_query: Whether to rewrite the natural language query for vector search. + + extra_headers: Send extra headers + + extra_query: Add additional query parameters to the request + + extra_body: Add additional JSON properties to the request + + timeout: Override the client-level default timeout for this request, in seconds + """ + if not vector_store_id: + raise ValueError(f"Expected a non-empty value for `vector_store_id` but received {vector_store_id!r}") + extra_headers = {"OpenAI-Beta": "assistants=v2", **(extra_headers or {})} + return self._get_api_list( + f"/vector_stores/{vector_store_id}/search", + page=AsyncPage[VectorStoreSearchResponse], + body=maybe_transform( + { + "query": query, + "filters": filters, + "max_num_results": max_num_results, + "ranking_options": ranking_options, + "rewrite_query": rewrite_query, + }, + vector_store_search_params.VectorStoreSearchParams, + ), + options=make_request_options( + extra_headers=extra_headers, extra_query=extra_query, extra_body=extra_body, timeout=timeout + ), + model=VectorStoreSearchResponse, + method="post", + ) + + +class VectorStoresWithRawResponse: + def __init__(self, vector_stores: VectorStores) -> None: + self._vector_stores = vector_stores + + self.create = _legacy_response.to_raw_response_wrapper( + vector_stores.create, + ) + self.retrieve = _legacy_response.to_raw_response_wrapper( + vector_stores.retrieve, + ) + self.update = _legacy_response.to_raw_response_wrapper( + vector_stores.update, + ) + self.list = _legacy_response.to_raw_response_wrapper( + vector_stores.list, + ) + self.delete = _legacy_response.to_raw_response_wrapper( + vector_stores.delete, + ) + self.search = _legacy_response.to_raw_response_wrapper( + vector_stores.search, + ) + + @cached_property + def files(self) -> FilesWithRawResponse: + return FilesWithRawResponse(self._vector_stores.files) + + @cached_property + def file_batches(self) -> FileBatchesWithRawResponse: + return FileBatchesWithRawResponse(self._vector_stores.file_batches) + + +class AsyncVectorStoresWithRawResponse: + def __init__(self, vector_stores: AsyncVectorStores) -> None: + self._vector_stores = vector_stores + + self.create = _legacy_response.async_to_raw_response_wrapper( + vector_stores.create, + ) + self.retrieve = _legacy_response.async_to_raw_response_wrapper( + vector_stores.retrieve, + ) + self.update = _legacy_response.async_to_raw_response_wrapper( + vector_stores.update, + ) + self.list = _legacy_response.async_to_raw_response_wrapper( + vector_stores.list, + ) + self.delete = _legacy_response.async_to_raw_response_wrapper( + vector_stores.delete, + ) + self.search = _legacy_response.async_to_raw_response_wrapper( + vector_stores.search, + ) + + @cached_property + def files(self) -> AsyncFilesWithRawResponse: + return AsyncFilesWithRawResponse(self._vector_stores.files) + + @cached_property + def file_batches(self) -> AsyncFileBatchesWithRawResponse: + return AsyncFileBatchesWithRawResponse(self._vector_stores.file_batches) + + +class VectorStoresWithStreamingResponse: + def __init__(self, vector_stores: VectorStores) -> None: + self._vector_stores = vector_stores + + self.create = to_streamed_response_wrapper( + vector_stores.create, + ) + self.retrieve = to_streamed_response_wrapper( + vector_stores.retrieve, + ) + self.update = to_streamed_response_wrapper( + vector_stores.update, + ) + self.list = to_streamed_response_wrapper( + vector_stores.list, + ) + self.delete = to_streamed_response_wrapper( + vector_stores.delete, + ) + self.search = to_streamed_response_wrapper( + vector_stores.search, + ) + + @cached_property + def files(self) -> FilesWithStreamingResponse: + return FilesWithStreamingResponse(self._vector_stores.files) + + @cached_property + def file_batches(self) -> FileBatchesWithStreamingResponse: + return FileBatchesWithStreamingResponse(self._vector_stores.file_batches) + + +class AsyncVectorStoresWithStreamingResponse: + def __init__(self, vector_stores: AsyncVectorStores) -> None: + self._vector_stores = vector_stores + + self.create = async_to_streamed_response_wrapper( + vector_stores.create, + ) + self.retrieve = async_to_streamed_response_wrapper( + vector_stores.retrieve, + ) + self.update = async_to_streamed_response_wrapper( + vector_stores.update, + ) + self.list = async_to_streamed_response_wrapper( + vector_stores.list, + ) + self.delete = async_to_streamed_response_wrapper( + vector_stores.delete, + ) + self.search = async_to_streamed_response_wrapper( + vector_stores.search, + ) + + @cached_property + def files(self) -> AsyncFilesWithStreamingResponse: + return AsyncFilesWithStreamingResponse(self._vector_stores.files) + + @cached_property + def file_batches(self) -> AsyncFileBatchesWithStreamingResponse: + return AsyncFileBatchesWithStreamingResponse(self._vector_stores.file_batches) |