ToolwebHow to use this tool
- Choose one unlocked PDF of up to 25 MiB and 200 pages, or load the example.
- Enter physical page positions such as 1-3, 5 and select Extract text. Leave the range blank for all pages.
- Find a word, copy the result or download a UTF-8 TXT file. Compare the extracted characters and reading order with the original.
Retrieve body text from selected pages
- Input and settings
- Five-page text PDF; range 2-3, 5; include page headings
- Result
- A UTF-8 TXT file containing text from pages 2, 3 and 5 with page headings
Numbers refer to physical positions in the file, not printed page labels. Repeated numbers are extracted once.
The PDF must contain actual text
This tool reads text items stored inside a document using PDF.js. It does not perform OCR on photographs or scanned pages. A page may show visible lettering but yield an empty result when that lettering is stored only as an image.
An empty result is explained separately from a processing error. If stored text produces incorrect characters, the original PDF may have an incomplete or incorrect font character map. Consider exporting another copy with text from the application that created it.
Check line breaks and reading order against the source
Extraction follows the order of stored text items. Line-ending information and changes in vertical position help reconstruct line breaks. It does not semantically rebuild newspaper columns, complex tables, equations or vertical writing. Compare extracted passages with the original before quoting or reusing them.
Turn off page headings to join body text without the added page labels. Find performs a case-insensitive literal search, and Next match selects the corresponding part of the result. The extracted result is read-only; use a separate editor to modify the text afterward.
Process long documents in smaller page ranges
The input limit is one PDF of 25 MiB and up to 200 pages. Extraction is capped at 150,000 characters per page, 2 million total characters, 200,000 text items and a 60-second task budget. Selecting fewer pages can reduce browser memory use and help retrieve the text you need sooner.
The file and extracted content are not sent to a server or URL and are not written to persistent browser storage. They remain in tab memory until cleared, reloaded or closed. Back navigation can restore tab state, while TXT files already downloaded remain on your device.
Common questions
Why is there no text from my scanned contract?
A PDF made of scanned images may contain no searchable text. This tool does not include OCR, so a separate text-recognition tool is needed.
Can I open a locked PDF if I know its password?
This tool does not collect passwords. If you are authorized to use the document, create an unlocked copy with the original application and select that copy.
Does the TXT preserve tables and fonts?
No. It saves plain UTF-8 text without the original fonts, colors, table formatting or images. Reading order in tables or columns also needs to be checked against the original.