Skip to main content

Sema4AI Document Intelligence API client

Project description

Document Intelligence Client

This Python package offers a robust client for seamless interaction with the Document Intelligence API. It enables you to efficiently retrieve, manage, and process document data within your workspace.

Installation

To install the package directly from GitHub using pip, run the following command:

pip install sema4ai-di-client

Usage

After installing the package, you can import it and start using the DocumentIntelligenceClient class.

Importing the Package

from sema4ai.di_client import DocumentIntelligenceClient

Getting a Document Work Item

To retrieve a document work item, make sure you've set the required environment variables before initializing the client. Specifically, ensure that DOCUMENT_INTELLIGENCE_SERVICE_URL and AGENTS_EVENTS_SERVICE_URL is set.

When running in Sema4.ai Control Room these are all handled by the platform.

  • Environment Variables: Make sure the required environment variables (DOCUMENT_INTELLIGENCE_SERVICE_URL, AGENTS_EVENTS_SERVICE_URL) are set in your environment before running the code.
  • Workspace ID: If you are developing on local, then make sure to set workspace_id in DocumentIntelligenceClient(workspace_id='<your_workspace_id>'), which is optional and deduced on non-local environments from the URL passed.
  • Initialize the Client: The client will automatically read the environment variables and initialize the connection.
  • Error Handling: The example includes error handling and ensures the client is properly closed at the end.

Here's an example:

from sema4ai.di_client import DocumentIntelligenceClient

# Ensure environment variables are set for the service URLs
# Example:
# export DOCUMENT_INTELLIGENCE_SERVICE_URL='https://api.yourdomain.com'
# export AGENTS_EVENTS_SERVICE_URL='https://agents.yourdomain.com'

# Initialize the client
client = DocumentIntelligenceClient()

# Specify the document ID you want to retrieve
document_id = 'your_document_id'

# Get the document work item
try:
    document_work_item = client.get_document_work_item(document_id)
    if document_work_item:
        print("Document Work Item:")
        print(document_work_item)
    else:
        print("No document work item found for the given document ID.")
except Exception as e:
    print(f"An error occurred: {e}")
finally:
    client.close()  # Make sure to close the client connection

Available Methods

The DocumentIntelligenceClient offers several methods to interact with the Document Intelligence API and manage work items. Below are examples of the available operations:

  • Get Document Type: Retrieve details about a specific document type.

    document_type = client.get_document_type(document_type_name)
    
  • Get Document Format: Fetch the format of a document based on its type and class.

    document_format = client.get_document_format(document_type_name, document_class_name)
    
  • Store Extracted Content: Store extracted content after processing a document.

    client.store_extracted_content(extracted_content)
    
  • Store Transformed Content: Save content that has been transformed by a process.

    client.store_transformed_content(transformed_content)
    
  • Store Computed Content: Submit content that has been computed after analysis.

    client.store_computed_content(computed_content)
    
  • Get Document Content: Retrieve document content in various states, such as raw, extracted, transformed, or computed.

    content = client.get_document_content(document_id, content_state)
    
  • Remove Document Content: Delete content for a document in a specific state.

    client.remove_document_content(document_id, content_state)
    
  • Complete Work Item Stage: Mark a work item’s current stage as complete and move to the next stage.

    response = client.work_items_complete_stage(
        work_item_id='your_work_item_id',
        status='SUCCESS',  # or 'FAILURE'
        status_reason='optional_reason',  # Optional
        log_details_path='optional_log_path'  # Optional
    )
    

Dependencies

The package requires the following dependencies:

  • urllib3 >= 1.25.3, < 2.1.0
  • python-dateutil
  • pydantic >= 2
  • typing-extensions >= 4.7.1

These should be installed automatically when you install the package via pip.

Example Code Snippet

Below is an example code snippet if you are testing on prod.

from sema4ai.di_client import DocumentIntelligenceClient

client = DocumentIntelligenceClient()

# Fetch and print the document type
try:
    document_type = client.get_document_type("CounterParty Reconciliation")
    print(document_type)
except Exception as e:
    print(f"An error occurred: {e}")
finally:
    client.close()  # Ensure proper resource cleanup

Below is an example code snippet if you are testing on local.

from sema4ai.di_client import DocumentIntelligenceClient
import os

# Set the required environment variables within the Python code
os.environ['DOCUMENT_INTELLIGENCE_SERVICE_URL'] = 'http://127.0.0.1:9080'
os.environ['AGENTS_EVENTS_SERVICE_URL'] = 'http://127.0.0.1:9080'

workspace_id = "<your Sema4.ai Control Room Workspace ID>"
client = DocumentIntelligenceClient(workspace_id=workspace_id)

# Fetch and print the document type
try:
    document_type = client.get_document_type("CounterParty Reconciliation")
    print(document_type)
except Exception as e:
    print(f"An error occurred: {e}")
finally:
    client.close()  # Ensure proper resource cleanup

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

sema4ai_di_client-1.0.11.tar.gz (71.2 kB view details)

Uploaded Source

Built Distribution

sema4ai_di_client-1.0.11-py3-none-any.whl (139.5 kB view details)

Uploaded Python 3

File details

Details for the file sema4ai_di_client-1.0.11.tar.gz.

File metadata

  • Download URL: sema4ai_di_client-1.0.11.tar.gz
  • Upload date:
  • Size: 71.2 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: poetry/1.8.4 CPython/3.10.12 Linux/6.5.0-1025-azure

File hashes

Hashes for sema4ai_di_client-1.0.11.tar.gz
Algorithm Hash digest
SHA256 9390a92e5477d0bb7178e523f5e30848ffd33912347d615d28398c79f2f03c36
MD5 97273194dc710d56deb40b65877a4c14
BLAKE2b-256 9167e7047cd01d388be481db1376003b11051090cff421ea05dec36210159f14

See more details on using hashes here.

File details

Details for the file sema4ai_di_client-1.0.11-py3-none-any.whl.

File metadata

  • Download URL: sema4ai_di_client-1.0.11-py3-none-any.whl
  • Upload date:
  • Size: 139.5 kB
  • Tags: Python 3
  • Uploaded using Trusted Publishing? No
  • Uploaded via: poetry/1.8.4 CPython/3.10.12 Linux/6.5.0-1025-azure

File hashes

Hashes for sema4ai_di_client-1.0.11-py3-none-any.whl
Algorithm Hash digest
SHA256 cea377b341431b465b04339ee736fe1d7922b0d84746fefb7a8f4ba816f9a7eb
MD5 4f13cb7056b4786400de4f123b2a1ba3
BLAKE2b-256 41a1424642d5446c91fefc7b152391bf8a8d079d8e1122d398b39ae6b9e702cb

See more details on using hashes here.

Supported by

AWS AWS Cloud computing and Security Sponsor Datadog Datadog Monitoring Fastly Fastly CDN Google Google Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page