Class PoiExtractionTools

java.lang.Object

ai.redouble.demo.poi.PoiExtractionTools

All Implemented Interfaces:

ExtractionTools

public class PoiExtractionTools
extends Object
implements ExtractionTools

The Apache POI / PDFBox reader, chosen over the engine's toolkit-free default whenever this module is on the classpath (a higher priority). It reads PDFs locally, page by page, so a PDF with a text layer never costs a model call. A native build that leaves this module off falls back to the toolkit-free reader, which is why the module exists separately.

Since:

0.1 (2026-09-22)

Author:

Andrey Santrosyan

Constructor Summary

Constructors

Constructor	Description
PoiExtractionTools()

Method Summary

Modifier and Type	Method	Description
AbstractTool<FileRef, Extraction>	deterministic(Identifiable parent, FileRef input)	The no-model reader for text and Office documents (and, when ExtractionTools.readsPdfLocally(), PDFs).
int	priority()	Ties break toward the higher number; the engine's toolkit-free default is 0.
boolean	readsPdfLocally()	True when this reader opens PDFs itself (POI/PDFBox); false when a PDF goes whole to the model tier.
AbstractTool<FileRef, Extraction>	vision(Identifiable parent, FileRef input)	The model reader for images (and PDFs when ExtractionTools.readsPdfLocally() is false, or a PDF's scanned pages when true).

Methods inherited from class Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Constructor Details

PoiExtractionTools

public PoiExtractionTools()

Method Details

priority

public int priority()

Description copied from interface: ExtractionTools

Ties break toward the higher number; the engine's toolkit-free default is 0.

Specified by:

priority in interface ExtractionTools

readsPdfLocally

public boolean readsPdfLocally()

Description copied from interface: ExtractionTools

True when this reader opens PDFs itself (POI/PDFBox); false when a PDF goes whole to the model tier.

Specified by:

readsPdfLocally in interface ExtractionTools

deterministic

public AbstractTool<FileRef, Extraction> deterministic(Identifiable parent,
 FileRef input)

Description copied from interface: ExtractionTools

The no-model reader for text and Office documents (and, when ExtractionTools.readsPdfLocally(), PDFs).

Specified by:

deterministic in interface ExtractionTools

vision

public AbstractTool<FileRef, Extraction> vision(Identifiable parent,
 FileRef input)

Description copied from interface: ExtractionTools

The model reader for images (and PDFs when ExtractionTools.readsPdfLocally() is false, or a PDF's scanned pages when true).

Specified by:

vision in interface ExtractionTools