PDF to Text

Extract PDF text in three steps

1. Choose one PDF up to 20 MiB and wait for its page count and supported profile to be checked. Encrypted and XFA documents are not supported.
2. Choose all pages or enter physical page numbers and ranges, such as 1,3-5. At most 200 pages can be selected; repeated pages are extracted only once in document order.
3. Select Extract text. Inspect individual page text, then choose Copy all text or Download TXT for the complete current selection.

What the extracted text represents

The tool reads the PDF's existing text layer, not the pixels of a scan. Font mappings, native spacing and text order affect the result; it may differ from the visible page.

Text items stay in the engine's returned order. An engine end-of-line adds a newline. Between selected pages, the tool inserts a newline, a form-feed character and another newline.

The engine normalizes some compatibility characters and whitespace. The tool does not reconstruct columns or tables, join broken words, or guarantee the original Unicode sequence.

The page preview shows one selected page at a time. Copy all text and Download TXT include the whole selection, with page separators, in UTF-8 without an added byte-order mark.

Selections and empty pages

For a five-page document, entering 5,1,3-5 selects pages 1,3,4,5 once each, in document order. Printed page numbers and viewer labels do not change these indices.

If one selected page has no extracted text but another does, the empty page's place is retained in the combined text through page separators.

If no selected page yields non-whitespace text, the tool reports No text extracted. The document might be scanned, blank, glyph-only or missing usable text mappings; no OCR runs automatically.

Supported scope and limits

This is text-layer extraction, not OCR or a layout-preserving converter. Comments, interactive forms, attachments and structure-tree reading order are not promised as page text.

Hidden, clipped or covered text can be included. Extraction does not certify redaction, remove sensitive content or prove that the text matches what is visible. Review the result.

The PDF may contain up to 500 pages, with at most 200 selected. Text-item, per-page, total text and time limits are enforced; a failed page or exceeded limit produces no partial export.

A dedicated core worker supports cancellation, but browser scheduling and native allocations are not hard real-time or memory guarantees. A timeout does not prove that the PDF is invalid.

Local document handling

The tool adds no document upload, input history or content-bearing URL. It may download fixed same-origin engine support files. Copy and TXT download happen only when you choose them; existing site services are not certified by this local workflow.