How to use it
- Choose or drop a PDF file.
- Decide whether each page should start with a heading such as “--- Page 3 ---”.
- Select Extract text and wait for all pages to be read.
- Copy the text or download it as a UTF-8 .txt file.
How it works
PDF files that were created from a word processor, a website or most other software contain a text layer: the characters and their positions. The tool reads that layer page by page with pdf.js and turns it into lines of plain text. Fonts, sizes, colours, images and tables are not kept; table cells come out as text separated by spaces or line breaks.
The reading order follows the order stored in the PDF, which is usually the visual order. Multi-column layouts, footnotes and text boxes can come out in an unexpected sequence, and hyphenated words at line ends stay split.
Scanned documents and photos saved as PDF contain only images of text and have no text layer, so nothing can be extracted. The tool detects this and points you to OCR (optical character recognition), which reads text from images. Text in some PDFs uses fonts without a proper character map; such text may come out as unreadable symbols.
PDFs that need a password to open cannot be read. PDFs with only copy restrictions are read, so respect the rights of the document's owner.
Examples
Quoting a report
Extract the text of a 40-page PDF report, search it in the text box and copy the paragraph you need without fighting the PDF viewer's selection.
Scanned letter
A letter scanned to PDF produces no text. The tool says so and links to the OCR tool, which can recognise the text from an image of the page.
Frequently asked questions
Why is no text found in my PDF?
The PDF most likely contains scanned images rather than text. Run the pages through OCR to recognise the text.
Why are words or lines in the wrong order?
The tool uses the order in which text is stored in the file. Some programs store columns, headers or text boxes in a different order from how they appear.
Why does the text contain strange symbols?
Some PDFs use fonts without information about which character each glyph represents. The visible letters cannot be mapped back to text; OCR is the only option for such files.
Is formatting such as bold or tables kept?
No. The output is plain text with line breaks. Table cells and columns become text on lines.
Privacy
The PDF is opened and read in your browser with pdf.js. The file and the extracted text are not uploaded, stored or included in analytics; only an anonymous event that the tool was used is counted.