Node Configuration

Owned by Brian Banet
Last updated: Sept 30, 2024
3 min read

The ArticulateML Parser Node has a number of required and optional components. Complete profile definitions are defined in the node itself.

Connection Settings

  • A Release Channel (stable, labs) is required. Unless otherwise directed, customers are encouraged to use the “stable” channel.

  • A License Key is required. If you have not been provided a license key for ArticulateML, requests into the service will fail.

These settings can be applied literally in the node, or they may be applied using Process Field Notation ({p_LicenseKey}).

image-20240328-130141.png

Page Selection

Directly beneath the connection settings, a checkbox option for Process all pages can be seen. By default, ArticulateML will only ever process Page 1 of a document. If your use case demands multi-page processing, this option should be checked.

image-20240328-130547.png

It is not advised to enable Process all pages on all documents. Doing so will likely introduce extra/incorrect data points into the process, and can results in additional costs as well. Only process all pages when the specific use case demands it.

In use cases where some but not all processing might need to span pages of a document, customers are strong encouraged to fork their processes so multiple ArticulateML nodes may be used with the appropriate settings for each specific scenario..

Headers / Footers

This section is leveraged for managing Key / Value responses from the OCR step. You are not required to set header fields.

  1. The Process Field dropdown will contain a list of all process fields added to the workflow. Select the process field you wish to assign from the list. If the list is empty, note you must first add fields to the workflow.

  2. AI Field Expressions is used to store the parsing expression for the process field. Do note, the expression assigned here will set a single value to a “header” field type. This section of the node does not interface with Table fields.

  3. Add a new Header Mapping if you care to set data from the AI response to additional fields.

  4. If you have a mapping that isn’t needed, click the Delete link to the right of that mapping.

Table Fields

Handling table fields is slightly different than handling header fields. To start, some additional information is required in the form of (1) the table field to target and (2) the “table of interest” from the AI response.

  1. Set the name of the table field that will be used to collect data. The is the name of a GlobalCapture process field that is of type “Table”.

  2. Set the table of interest / assign $toi. When table expressions are evaluated, any occurrence of the string $toi will be replaced with the expressions provided in the Table AI Table of Interest field. This essentially allows for shorthand notation to reference that part of the expression without having to rekey it, potentially multiple times, in more complex expressions and / or functions.

    When mapping values to table fields, with a properly formed Table of Interest expressing the rest of the statement can be significantly simpler and less error prone. Use of this replacement is highly encouraged.

  3. Table Field Cell is the Column Field’s name in the table you wish to target. This is the field name, as it is presented in the table.

  4. AI Field Expressions is used to store the parsing expression for the process field. As with Header/Footer, the expression provided here sets the value of the field from the AI response.