Slug: document-upload Version: 1.0.0 Infer: Yes

Overview

Upload Document Model

The Upload Document model enables the addition of new documents to existing document families within Kodexa. This model is particularly useful for multi-document workflows, creating document collections, and triggering data extraction using specified data definitions.

How It Works

  1. The model takes an input document from your processing pipeline
  2. It associates this document with the current document family
  3. Optionally, it applies specified data definitions (taxonomies) to the document
  4. Data can be extracted from the document and stored in a specified data store
  5. The document can be redirected to a different document store if needed

Options Configuration

OptionDescription
data_storeThe target data store for extracted information from the document
taxonomiesList of data definitions (taxonomies) to apply for data extraction
document_storeOptional alternative document store where the document should be saved
pathOptional new path to assign to the document in the document store

Process Flow

Use Cases

The Upload Document model is particularly useful for:

  • Document Collections: Adding related documents to a document family
  • Multi-Document Processing: Building document sets that need to be processed together
  • Document Versioning: Adding new versions of documents to existing families
  • Automated Data Extraction: Triggering extraction using specified data definitions
  • Document Routing: Redirecting documents to appropriate document stores
  • Document Organization: Building logical document groupings

Example Scenarios

Basic Document Upload

To simply add a document to the current document family:

# No additional configuration required

Document Upload with Data Extraction

To upload a document and extract data based on an invoice taxonomy:

data_store: "org/invoice-data:1.0.0"
taxonomies:
  - "org/invoice-taxonomy:1.0.0"

Document Upload with Redirection

To upload a document to a different document store with a specific path:

document_store: "org/archive-documents:1.0.0"
path: "completed/invoice-123.pdf"

Integration Considerations

  • The model preserves the original document’s content and metadata
  • Document families can contain multiple documents in different formats
  • Data extraction is performed asynchronously if taxonomies are specified
  • The model works seamlessly with other document processing steps in a pipeline
  • Permissions are respected when adding documents to document stores

Inference Options

The following options can be configured when using this model for inference:

NameLabelTypeDescriptionDefaultRequired
data_storeData StoretableStoreN/A-No
taxonomiesData DefinitionslistN/A-No
document_storeTarget Document StoredocumentStoreN/A-No
pathNew PathstringN/A-No

Model Details

  • Provider: Kodexa AI