The "Xtract" input system is used to read batches that have been exported to the file system by the IRISXtract application. Unlike other file-based input systems, xSuite Interface does not generate a separate status file for Xtract batches. Instead, xSuite Interface uses the status file already created by Xtract and changes the file extension.

The file extension reflects the internal processing state of xSuite Interface:

By default, only the page-related image files are read in as attachments to the documents in a batch. The various types of accompanying index files are not read in by default. If any of these files are also to be included, specify this explicitly in the .IncludeIndexFile[] property below. The following files are generally supported:

Batch- and transaction-related files are redundantly attached to each document in the batch.

Several of the page-related files (P{page}.txt and P{page}.dat) might exist, all of which are assigned to the document in question. In order to make the files prop.txt and para.dat, which have the same name at different levels, distinguishable for later processing, these files are implicitly given a name prefix. The name prefix identifies the level: batch_, unit_, doc_ and page_.

The content of the index files must be read downstream with a suitable index data reader:

Page-level index data refers to a file attachment, but the xSuite Interface data model provides for document-level index data only. However, by using the .ReadPageIndex[] property, this index data can be attached to file attachments in the form of free metadata.