/unstructured-python-client

A Python client for the Unstructured hosted API

Primary LanguagePythonMIT LicenseMIT

Python SDK for the Unstructured API

NOTE: This README is for the 0.26.0-beta version. The current published SDK, 0.25.5 can be found here.

This is a Python client for the Unstructured API.

Please refer to the Unstructured docs for a full guide to using the client.

SDK Installation

PIP

pip install unstructured-client

Poetry

poetry add unstructured-client

SDK Example Usage

Example

import os

import unstructured_client
from unstructured_client.models import operations, shared

client = unstructured_client.UnstructuredClient(
    api_key_auth=os.getenv("UNSTRUCTURED_API_KEY"),
    server_url=os.getenv("UNSTRUCTURED_API_URL"),
)

filename = "PATH_TO_FILE"
with open(filename, "rb") as f:
    data = f.read()

req = operations.PartitionRequest(
    partition_parameters=shared.PartitionParameters(
        files=shared.Files(
            content=data,
            file_name=filename,
        ),
        # --- Other partition parameters ---
        strategy=shared.Strategy.AUTO,
        languages=['eng'],
    ),
)

try:
    res = client.general.partition(request=req)
    print(res.elements[0])
except Exception as e:
    print(e)

Refer to the API parameters page for all available parameters.

Configuration

Splitting PDF by pages

See page splitting for more details.

In order to speed up processing of large PDF files, the client splits up PDFs into smaller files, sends these to the API concurrently, and recombines the results. split_pdf_page can be set to False to disable this.

The amount of workers utilized for splitting PDFs is dictated by the split_pdf_concurrency_level parameter, with a default of 5 and a maximum of 15 to keep resource usage and costs in check. The splitting process leverages asyncio to manage concurrency effectively. The size of each batch of pages (ranging from 2 to 20) is internally determined based on the concurrency level and the total number of pages in the document. Because the splitting process uses asyncio the client can encouter event loop issues if it is nested in another async runner, like running in a gevent spawned task. Instead, this is safe to run in multiprocessing workers (e.g., using multiprocessing.Pool with fork context).

Example:

req = shared.PartitionParameters(
    files=files,
    strategy="fast",
    languages=["eng"],
    split_pdf_concurrency_level=8
)

Sending specific page ranges

When split_pdf_page=True (the default), you can optionally specify a page range to send only a portion of your PDF to be extracted. The parameter takes a list of two integers to specify the range, inclusive. A ValueError is thrown if the page range is invalid.

Example:

req = shared.PartitionParameters(
    files=files,
    strategy="fast",
    languages=["eng"],
    split_pdf_page_range=[10,15],
)

Splitting PDF by pages - strict mode

When split_pdf_allow_failed=False (the default), any errors encountered during sending parallel request will break the process and raise an exception. When split_pdf_allow_failed=True, the process will continue even if some requests fail, and the results will be combined at the end (the output from the errored pages will not be included).

Example:

req = shared.PartitionParameters(
    files=files,
    strategy="fast",
    languages=["eng"],
    split_pdf_allow_failed=True,
)

Retries

Some of the endpoints in this SDK support retries. If you use the SDK without any configuration, it will fall back to the default retry strategy provided by the API. However, the default retry strategy can be overridden on a per-operation basis, or across the entire SDK.

To change the default retry strategy for a single API call, simply provide a RetryConfig object to the call:

from unstructured_client import UnstructuredClient
from unstructured_client.models import shared
from unstructured_client.utils import BackoffStrategy, RetryConfig

s = UnstructuredClient(
    api_key_auth="YOUR_API_KEY",
)


res = s.general.partition(request={
    "partition_parameters": {
        "files": {
            "content": open("<file_path>", "rb"),
            "file_name": "your_file_here",
        },
        "chunking_strategy": shared.ChunkingStrategy.BY_TITLE,
        "split_pdf_page_range": [
            1,
            10,
        ],
        "strategy": shared.Strategy.HI_RES,
    },
},
    RetryConfig("backoff", BackoffStrategy(1, 50, 1.1, 100), False))

if res.elements is not None:
    # handle response
    pass

If you'd like to override the default retry strategy for all operations that support retries, you can use the retry_config optional parameter when initializing the SDK:

from unstructured_client import UnstructuredClient
from unstructured_client.models import shared
from unstructured_client.utils import BackoffStrategy, RetryConfig

s = UnstructuredClient(
    retry_config=RetryConfig("backoff", BackoffStrategy(1, 50, 1.1, 100), False),
    api_key_auth="YOUR_API_KEY",
)


res = s.general.partition(request={
    "partition_parameters": {
        "files": {
            "content": open("<file_path>", "rb"),
            "file_name": "your_file_here",
        },
        "chunking_strategy": shared.ChunkingStrategy.BY_TITLE,
        "split_pdf_page_range": [
            1,
            10,
        ],
        "strategy": shared.Strategy.HI_RES,
    },
})

if res.elements is not None:
    # handle response
    pass

Custom HTTP Client

The Python SDK makes API calls using the httpx HTTP library. In order to provide a convenient way to configure timeouts, cookies, proxies, custom headers, and other low-level configuration, you can initialize the SDK client with your own HTTP client instance. Depending on whether you are using the sync or async version of the SDK, you can pass an instance of HttpClient or AsyncHttpClient respectively, which are Protocol's ensuring that the client has the necessary methods to make API calls. This allows you to wrap the client with your own custom logic, such as adding custom headers, logging, or error handling, or you can just pass an instance of httpx.Client or httpx.AsyncClient directly.

For example, you could specify a header for every request that this sdk makes as follows:

from unstructured_client import UnstructuredClient
import httpx

http_client = httpx.Client(headers={"x-custom-header": "someValue"})
s = UnstructuredClient(client=http_client)

or you could wrap the client with your own custom logic:

from unstructured_client import UnstructuredClient
from unstructured_client.httpclient import AsyncHttpClient
import httpx

class CustomClient(AsyncHttpClient):
    client: AsyncHttpClient

    def __init__(self, client: AsyncHttpClient):
        self.client = client

    async def send(
        self,
        request: httpx.Request,
        *,
        stream: bool = False,
        auth: Union[
            httpx._types.AuthTypes, httpx._client.UseClientDefault, None
        ] = httpx.USE_CLIENT_DEFAULT,
        follow_redirects: Union[
            bool, httpx._client.UseClientDefault
        ] = httpx.USE_CLIENT_DEFAULT,
    ) -> httpx.Response:
        request.headers["Client-Level-Header"] = "added by client"

        return await self.client.send(
            request, stream=stream, auth=auth, follow_redirects=follow_redirects
        )

    def build_request(
        self,
        method: str,
        url: httpx._types.URLTypes,
        *,
        content: Optional[httpx._types.RequestContent] = None,
        data: Optional[httpx._types.RequestData] = None,
        files: Optional[httpx._types.RequestFiles] = None,
        json: Optional[Any] = None,
        params: Optional[httpx._types.QueryParamTypes] = None,
        headers: Optional[httpx._types.HeaderTypes] = None,
        cookies: Optional[httpx._types.CookieTypes] = None,
        timeout: Union[
            httpx._types.TimeoutTypes, httpx._client.UseClientDefault
        ] = httpx.USE_CLIENT_DEFAULT,
        extensions: Optional[httpx._types.RequestExtensions] = None,
    ) -> httpx.Request:
        return self.client.build_request(
            method,
            url,
            content=content,
            data=data,
            files=files,
            json=json,
            params=params,
            headers=headers,
            cookies=cookies,
            timeout=timeout,
            extensions=extensions,
        )

s = UnstructuredClient(async_client=CustomClient(httpx.AsyncClient()))

File uploads

Certain SDK methods accept file objects as part of a request body or multi-part request. It is possible and typically recommended to upload files as a stream rather than reading the entire contents into memory. This avoids excessive memory consumption and potentially crashing with out-of-memory errors when working with very large files. The following example demonstrates how to attach a file stream to a request.

Tip

For endpoints that handle file uploads bytes arrays can also be used. However, using streams is recommended for large files.

from unstructured_client import UnstructuredClient
from unstructured_client.models import shared

s = UnstructuredClient(
    api_key_auth="YOUR_API_KEY",
)


res = s.general.partition(request={
    "partition_parameters": {
        "files": {
            "content": open("<file_path>", "rb"),
            "file_name": "your_file_here",
        },
        "chunking_strategy": shared.ChunkingStrategy.BY_TITLE,
        "split_pdf_page_range": [
            1,
            10,
        ],
        "strategy": shared.Strategy.HI_RES,
    },
})

if res.elements is not None:
    # handle response
    pass

Maturity

This SDK is in beta, and there may be breaking changes between versions without a major version update. Therefore, we recommend pinning usage to a specific package version. This way, you can install the same version each time without breaking changes unless you are intentionally looking for the latest version.

Installation Instructions for Local Development

The following instructions are intended to help you get up and running with unstructured-python-client locally if you are planning to contribute to the project.

  • Using pyenv to manage virtualenv's is recommended but not necessary

    • Mac install instructions. See here for more detailed instructions.
      • brew install pyenv-virtualenv
      • pyenv install 3.10
    • Linux instructions are available here.

  • Create a virtualenv to work in and activate it, e.g. for one named unstructured-python-client:

    pyenv virtualenv 3.10 unstructured-python-client pyenv activate unstructured-python-client

  • Run make install and make test

Contributions

While we value open-source contributions to this SDK, this library is generated programmatically by Speakeasy. In order to start working with this repo, you need to:

  1. Install Speakeasy client locally https://github.com/speakeasy-api/speakeasy#installation
  2. Run speakeasy auth login
  3. Run make client-generate. This allows to iterate development with python client.

There are two important files used by make client-generate:

  1. openapi.json which is actually not stored here, but fetched from unstructured-api, represents the API that is supported on backend.
  2. overlay_client.yaml is a handcrafted diff that when applied over above, produces openapi_client.json which is used to generate SDK.

Once PR with changes is merged, Github CI will autogenerate the Speakeasy client in a new PR, using the openapi.json and overlay_client.yaml You will have to manually bring back the human created lines in it.

Feel free to open a PR or a Github issue as a proof of concept and we'll do our best to include it in a future release!

SDK Created by Speakeasy