How to use the pdf link extractor
- Choose a PDF or drop it on the box. The file is read on your device, and it is not uploaded.
- Read the counts: links to addresses, different addresses, links that jump inside the file and pages with links.
- Study the table. Each link is listed with its page, a kind label (web, email, phone, file or other) and the full address as plain text.
- Copy the list of addresses, one per line without repeats, or download a CSV with the page, kind and address of every link.
What counts as a link
A PDF can store a link as a clickable area on a page with an address attached. This tool lists those areas. Text that looks like an address but has no link area behind it is not listed, and neither are links inside the document that jump to another page, which are counted instead. A single click area can show different words from the address it points to, so the table gives the real address.
Safe handling
The tool never opens, loads or follows an address. Each one is shown as plain text, with a label for the kind of address, so you can look before you go. Addresses that start with file, or that point to a folder on a computer, are labeled so they are easy to notice. Before you click any link in a document you did not write, check where it goes. You can use the list to check links in your own documents too, for example before you send a report.
How the file is read
The PDF is opened with the PDF.js library inside your browser, and the page annotations are read one page at a time. PDFs with a password cannot be opened, and the tool does not try to guess or remove passwords. Files up to 25 MB and 500 pages are accepted. A test file with four links on three pages gave the same four addresses, in page order, as a count of the link actions in the raw file.
More tools for reading inside a PDF
To look inside a PDF in other ways, try the PDF Attachment Extractor.
Assumptions and limits
- The PDF is not password protected and has no more than 500 pages.
- Only link areas with an address are listed as links; internal jumps are only counted.
- Addresses are shown as text exactly as stored, with control characters turned into spaces.
- Nothing is opened, visited or sent anywhere, and the file stays on your device.
- The list is cut off at 20,000 links to keep the page responsive.
Frequently asked questions
Does the tool check that the links work?
No. Checking a link would mean sending a request to the address, which this tool never does. It only lists what the file contains.
Why is a web address in my PDF missing from the list?
It may be plain text with no link area behind it, or the link may be in an image. Only link areas with an address attached are listed.
What does the kind label mean?
It is taken from the start of the address: https and http are web addresses, mailto is an email, tel is a phone number and file points to a file on a computer. Anything else is labeled other.
Can I list the links of a scanned document?
A scan is a picture of pages and has no link areas, so the list will be empty. Text recognition tools may add them, but this tool only reads what the file stores.