Extract Text From a PDF Without Uploading the File

Reads the text layer of a PDF with pdf.js and puts it in a copy-ready box on the page, with a .txt download alongside it. Your browser opens the document; no server ever receives it.

About the PDF Text Extractor

This tool asks pdf.js for the characters a PDF actually stores. It walks every page in order, collects the text items, reassembles them into lines, and prints the result into a read-only text box with a Copy text button above the download. There is no page-range field — the extractor always reads the document from the first page to the last, and separates each one with a line reading ----- Page break ----- so you can see where one page ended.

One thing to understand before you start: a PDF either has a text layer or it does not. A file produced by Word, LaTeX, Google Docs or a browser print-to-PDF stores real characters, and everything comes out. A file produced by a flatbed scanner or a phone camera stores a photograph of a page, and there are no characters in it to find. If the box comes back empty or nearly empty, that is what has happened — the tool does not run OCR, so it cannot invent text that the file never contained.

Key Features

  • Reads characters, not pixels: Calls the pdf.js getTextContent() API, so you get the exact glyphs embedded in the document rather than a guess made from an image.
  • Copy in one click: The Copy text button writes the whole extraction to your clipboard through the Clipboard API, ready to paste into an email, a note or a search box.
  • Download as well as display: Every run also produces a .txt file named after the source PDF, so you can keep the extraction without selecting anything by hand.
  • Page boundaries preserved: Pages are joined with a visible break marker, which matters when you need to cite where a sentence appeared.
  • Whole document in one pass: No page picker to fill in and no cap on length — a 400-page report is read the same way a two-page letter is.
  • Self-contained: pdf.js and its worker are served from /assets/vendor/js/ on this domain, and the site Content-Security-Policy refuses every outbound request, so there is no path by which the document could be transmitted.

How to Use the PDF Text Extractor

  1. Choose the PDF you want to read, or drop it onto the tool.
  2. Press Extract Text and watch the progress line count through the pages.
  3. Read the result in the box that appears below the tool.
  4. Press Copy text to send the whole extraction to your clipboard, or use the download link for the .txt file.
  5. Close the tab when you are finished — nothing was stored.

How It Works

pdf.js returns each page as a list of text items, and every item carries a transformation matrix describing where on the page it sits. The extractor tracks the vertical component of that matrix and starts a new line whenever the position shifts by more than two units, which is how a flat list of fragments becomes readable paragraphs.

That heuristic is honest but simple, and it has a known consequence: on a two-column layout, a line from the left column and a line from the right column sit at the same height, so they are treated as one line and end up concatenated. Academic papers, newsletters and magazine-style PDFs will therefore need tidying. Single-column documents — contracts, letters, reports, manuscripts — reconstruct cleanly.

When You Need It

A compliance officer receives a signed vendor policy as a PDF and needs to check whether a specific clause survived the last revision. Pasting a confidential supplier agreement into a web converter to make it searchable is exactly the disclosure the compliance function exists to prevent, so the extraction happens in the tab instead and the text goes straight into a local search.

A grant writer works from a funder's guidance document that is marked not for circulation. Pulling the eligibility wording out to quote it accurately in an application is routine work, but it should not require handing the funder's private guidance to a third party first.

It is also the fastest way to answer a question you may not have known you had: if you extract a PDF and nothing comes back, you have learned that the file is a scan, and that any tool promising to convert it to Word will only ever give you a picture in a document.

FAQ

Frequently Asked Questions

Why did the extraction come back empty?
The PDF has no text layer. It is almost certainly a scan or a photograph saved as a PDF, meaning each page is an image and there are no characters stored anywhere in the file. This tool reads what is there rather than recognising shapes, so an image-only page yields nothing. Optical character recognition would be needed, and that is a different kind of process entirely.
Does the text come out in the correct order on a two-column page?
Not reliably. Lines are grouped by their vertical position, and in a two-column layout both columns share the same vertical positions, so the left and right lines get merged into one. Expect to reorder the output by hand for journal papers and newsletters. Single-column documents come out in true reading order.
Can it open a PDF that asks for a password?
If the PDF has an owner password only — the kind that restricts printing or copying but opens without prompting — pdf.js will load it and the text will extract. If it has a user password, so that a reader must type something before the file will display at all, loading fails and you will see an error. Remove the password first with the PDF Unlock tool.
Is the document transmitted anywhere while it is being read?
No. The PDF is handed to pdf.js inside the page you already have open, and pdf.js is hosted on this domain. There is no upload step in the code, and the Content-Security-Policy header would block a request to an outside host even if one existed. You can confirm it: open your browser's network panel, run an extraction, and watch that no request is made.
How is this different from the PDF to TXT converter?
The extraction itself is identical. This tool shows the result on screen so you can read, check and copy it immediately, and gives you the .txt as a secondary option. The TXT converter skips the display and hands you the file directly, which is what you want when the text is destined for a script rather than your eyes.
Are tables preserved?
No. A table in a PDF is not stored as a table — it is stored as text positioned at particular coordinates, with the ruled lines drawn separately as graphics. Extraction returns the cell contents as lines of text with the structure gone. For tabular data, the PDF to Excel converter is the better starting point.
Keep going

Related Tools

PDF to TXT ConverterPDF to HTML ConverterWord Counter for PDF