Skip to main content

​
  Source connectors

Pipelines Transform Source connector operations connect workflows to remotely hosted source locations. These locations hold the documents and semi-structured data that Unstructured Pipelines processes. Learn more about source connectors.

​
  Destination connectors

Pipelines Transform Destination connector operations connect workflows to remotely hosted blob storage, databases, and vector stores. Pipelines delivers output results to these destinations. Learn more about destination connectors.

​
  Workflows

Pipelines Transform Workflow operations define and manage sequences of activities in Unstructured Pipelines. These activities ingest remotely hosted documents and semi-structured data, process them, and deliver the results to remotely hosted blob storage, databases, and vector stores. Learn more about workflows.

​
  Templates

Pipelines Transform Template operations return information about common Unstructured Pipelines workflow patterns. You can apply a pattern to your workflows for predictable, consistent results.

​
  Jobs

Pipelines Transform For Unstructured Pipelines, specific job operations run and manage workflows in Unstructured Pipelines. Learn more about job operations for Pipelines. For Unstructured Transform, specific job operations send locally hosted files to the Unstructured Transform service for processing. Learn more about job operations for Transform.

​
  Notifications

Pipelines Transform Notification operations alert recipients when an Unstructured Pipelines workflow changes state. States include in progress, completed, stopped, and failed. Learn about webhooks and email notifications.

​
See also

Retry failed API calls

Pipelines Transform
Python SDK REST clients
For the Unstructured Python SDK, if an API request fails, you can programmatically retry the request with an exponential backoff strategy.

Choose from available models

Pipelines Transform
Python SDK REST clients
For Pipelines workflows, you can pick a specific model to use in some kinds of workflow nodes. Some of these models might not be available, depending on your organization’s Unstructured deployment type.

Handle API errors

Pipelines Transform
Python SDK REST clients
For the Unstructured Python SDK, you can programmatically detect and react to errors.

Migrate from previous API versions

Pipelines Transform
Python SDK REST clients
For Pipelines, move your existing code to the latest Unstructured Python SDK version.