Full-text search (POC)

Overview

This document provides technical support information for the new module "Full-text search (POC)" which includes the plugin "OCR (POC)". The module is designed to calculate Optical Character Recognition (OCR) for objects during the ingest process, storing the results in the record metadata to enable searching on the text of a file.

Activation

1. Create a field definition Dynamic.PocOcr of type TextField

   • Also, mark the field definition as global and indexed such that the end-user experiences a full-text search using the global search

2. Link the module FULL_TEXT_SEARCH_POC to the customer’s organisation. See the REST documentation for information on how to do that.

Once activated, OCR will be carried out for all files that are newly ingested via the ingest 1.5 and 2.0 flows for that organisation. The extracted text will be saved to the Dynamic.PocOcr field.

Supported file types

The following table contains file types that are confirmed to work.

Theoretically all supported Tika formats are supported: Apache Tika – Supported Document Formats. However, this is not guaranteed for file formats not listed in the table below.