#Text Extraction

All page numbers are one-based.

const firstPageText = pdf.page(1).text();

Use document.text(...) for multi-page extraction:

const text = pdf.text({
  maxPages: 5,
  pages: [1, 3, 4],
  maxChars: 200_000,
});

When pages is provided without maxPages, the explicit page list is used as is. Invalid page numbers throw PdfPageRangeError.

#Limits

text(...) stops after maxPages effective pages. Text output is capped by maxChars, default 200_000, so huge PDFs cannot accidentally become enormous prompt payloads.

#Text Shape

PDF text extraction preserves PDFium's text order. It is good for search, summaries, and model context, but PDFs do not always encode text in visual reading order.