Extract PDF Links & Add Links

Export PDF links to CSV or make URLs and page references clickable

Export PDF links to CSV or make URLs and page references clickable

1Choose file2Settings3Process4Save

Choose a PDF and its contents will be analyzed and shown here.

Files are processed only in this browser. Nothing is uploaded to a server.

Export URLs, link text and page numbers from a PDF to CSV

Continue in another tool
Task settings
Extraction settings

Click a row to highlight the link area on its page.

PageLink textTypeTargetDomain
Preview

Remove the password from an encrypted PDF first. Processing focuses on page content, so bookmarks, attachments, forms and signatures may not be preserved. Keep the original and check the result.

About Extract PDF Links & Add Links

A link in a PDF isn't the blue underlined text. It is an invisible rectangle, a link annotation, laid over the page. The rectangle carries an action such as “open this URL” or “go to page 7”, and inside the file it has no connection to the words underneath it. That is why a PDF can show one address in the text and send the click somewhere else entirely.

The Extract tab reads every link annotation and exports a table of page, link text, type, target and domain (as CSV, a Markdown table, or a TXT list of URLs only). The Add tab finds URLs, email addresses and page references like “p. 12” in the text, turns them into real links, and checks existing links for broken targets.

The link types worth a second look

Links are classed as web address, email, internal page jump, open external document (GoToR), launch external file (Launch), JavaScript, and other. When you're vetting a PDF someone sent you, check the last three first. Launch and JavaScript actions can start a program or run a script with one click, so finding them in an ordinary document is a reason for suspicion. For the same reason, look for rows where the link text and the domain column don't match.

The link text is collected from text whose centre falls inside the rectangle. In PDFs where the link area is narrower than the words or slightly offset, the cell may be empty or pick up a neighbouring word. Turning on plain-text URLs also gathers addresses that are only typed in the body, and the tool tries to rejoin URLs that wrap at a line end. An address that already has a link annotation on the same page isn't listed twice.

Add tab: page references and the offset

Printed page numbers rarely match PDF page positions, because covers and front matter come first. If body page 1 follows a cover and three contents pages, enter an offset of 4 and “p. 12” links to the 16th page of the PDF. Candidates whose target falls outside the document get a red outline, which usually means the offset is wrong.

Candidates that overlap an existing link by half or more are skipped. If the area you want isn't detected, type a target (URL, email or page number) and drag a box on the preview. The broken-link check confirms that internal targets exist, including named destinations, but for external URLs it can only check the format: browser security (CORS) prevents testing whether the address actually opens.

Worked examples

Listing the reference links in a paper

Input
A 20-page paper with 35 link annotations; each DOI is linked twice, once in the text and once in the references; “Merge duplicates” on
Output
One row per target, with every page it appears on in the Page column

Merging collapses identical URLs into one row and records how many times each occurred. Adding “Exclude internal page links” drops footnote and contents jumps so only outside sources remain.

How to use it

  1. Add the PDF; its link annotations appear in a table, and clicking a row highlights the link area on its page.
  2. Choose whether to merge duplicates, drop internal links or include plain-text URLs, then pick CSV, Markdown or TXT.
  3. Save the list. To create links instead, pick candidates in the Add tab and run it.

Frequently asked questions

Will the CSV open correctly in Excel?

Yes. The file is written as UTF-8 with a byte-order mark, which Excel uses to detect the encoding, so accented and non-Latin characters display correctly.

Does it find URLs in scanned PDFs?

It finds link annotations in any PDF, but plain-text URLs can only be found when the page has a text layer. Run OCR on a scan first.

Do added links get underlined?

With “Underline links” on, the link annotation's border is drawn as an underline. The text on the page itself is not changed.

Notes

Why annotations live as objects separate from the page content is explained in the guide “How PDF Files Work”.

Sources

Last reviewed:

Privacy and how data is handled