Foxit Developer

Text extraction ​

Foxit PDF SDK for Web can extract document text, single-page text, text in a region, and per-character information. Related APIs include:

  • PDFDoc.getText()
  • PDFPage.getText()
  • PDFPage.getTextInRect()
  • PDFPage.getCharCount()
  • PDFPage.getCharInfoByIndex()
  • PDFPage.getCharRange()
  • PDFPage.getCharInfoAtPoint().

To find keywords in a document and obtain match rectangles, use PDFDoc.searchText() or getTextSearch() in Text search. This page covers sequential text readout, region selection, and character-level geometry.

Text extraction suits content analysis, proofreading, region pick-up, batch processing, or syncing text with business systems.

Extract document text ​

PDFDoc.getText(pages) extracts text over a page range. It returns TaskProgress; use onProgress for progress and segment text (field names depend on the callback). You do not await a single full string directly.

javascript
const pdfDoc = pdfViewer.getCurrentPDFDoc();
const progress = pdfDoc.getText([[0, 2]]);
const removeListener = progress.onProgress((data) => {
    console.log(data.percent);
    if (data.content !== undefined) {
        console.log(data.content);
    }
});
// Unsubscribe when done and end the task if needed
removeListener();

NOTE

PDFDoc.getText() does not support XFA documents (the implementation throws on XFA). For XFA forms, use form APIs. Extraction requires document extract permission.

Extract single-page text ​

javascript
const page = await pdfDoc.getPageByIndex(0);
const text = await page.getText();

console.log(text);

For structured line breaks with options such as minimum width, use PDFPage.getTexts() (see API Reference).

Extract text in a region ​

PDFPage.getTextInRect(rect, type) extracts text inside a PDF-coordinate rectangle. Optional type: 0 page and field text, 1 page only, 2 fields only; default 0.

The return value is { pageText, fieldsText } (fieldsText is an array of field entries), not a plain string.

javascript
const page = await pdfDoc.getPageByIndex(0);

const { pageText, fieldsText } = await page.getTextInRect({
    left: 100,
    right: 300,
    top: 720,
    bottom: 680
}, 0);

NOTE

  • getTextInRect() uses PDF coordinates, not Viewer device pixels—the same system as markRedactAnnot in Redaction.
  • Throws on static XFA (similar to the XFA limit on PDFDoc.getText()).
  • Requires document extract permission.

Per-character text information ​

From 11.1.0, PDFPage exposes character indexing—useful for proofreading, annotation anchors, and pairing with TextSearchMatch from Text search (e.g. getCharInfoByIndex after a hit).

javascript
const page = await pdfDoc.getPageByIndex(0);
const charCount = await page.getCharCount();

for (let i = 0; i < charCount; i++) {
    const charInfo = await page.getCharInfoByIndex(i);
    console.log(charInfo);
}

Character index ranges in a region ({ start, end }[]):

javascript
const charRange = await page.getCharRange({
    left: 100,
    right: 300,
    top: 720,
    bottom: 680
});

At a PDF point, use PDFPage.getCharInfoAtPoint(point, tolerance) with point as [x, y] and tolerance e.g. 0–30.

Notes ​

  • Page index starts at 0.
  • Results depend on extractable text in the PDF; scanned pages usually need OCR first.
  • Regions and character geometry use PDF coordinates, not device pixels used for snapshots or rendering.
  • For large documents, process by page or page range; PDFDoc.getText() supports progress callbacks.
  • Character and region APIs need extract or extract for accessibility permission as implemented.