Page Link Extractor: all links, anchor text, nofollow

Paste in a list of page URLs and you get back every link on each page. Each link comes with its anchor text, its rel values and a flag for internal or external.

Each page is fetched once with redirects followed, and the HTML is read as the server sends it. Scripts are not run, so a link that JavaScript adds after the page loads won't show up.

What you get for each link

• linkUrl, the absolute URL, resolved against the page address or its <base href>, with the raw href kept beside it
• anchorText. For a link wrapped around an image with no text, the image's alt text is used, and isImageLink and imageAlt are set.
• linkType: internal when the link's host matches the page's host (a leading www. is ignored), external otherwise, and other for mailto, tel and similar links when you ask for them
• rel as a list, plus nofollow, sponsored and ugc as true or false
• target, the title attribute, the link host and scheme, whether it came from an <a> or an image map <area>, and its position on the page

Google lets a site mark a link with sponsored, ugc or nofollow and accepts several values in one attribute, separated by spaces or commas [1]. Both forms are read here.

Every row also carries the page it came from: pageUrl, finalPageUrl after redirects, the HTTP status, and that page's totals for links, internal, external, nofollow and unique URLs.

Input

• Page URLs (required). An address without http:// or https:// is read as https://.
• Links to return: all, internal only or external only.
• Output format. One row per link is the default and suits a spreadsheet. One row per page puts the links in a nested links list.
• One row per unique URL on a page, off by default. When on, repeated links to the same URL merge into one row with an occurrences count.
• Count subdomains as internal, off by default. With it on, blog.example.com counts as internal on example.com.
• Remove #fragments from link URLs (on), skip links to a spot on the same page such as #top (on), include mailto and tel links (off).
• Timeout per page (20 seconds), pages read at once (10) and the user agent.

javascript: and data: links are always left out.

Output

The dataset has the rows described above. The Overview view shows one line per link and the Pages view shows the page totals. A page that answers 400 or above, isn't HTML, can't be reached or isn't a valid URL gets one row with ok false and the reason in error, and the run goes on to the next page. A page that has no links of the types you picked gets one row with linkUrl empty and a note.

The key value store holds an OUTPUT record with pages processed, pages failed, links found and whether the run stopped at your spending limit.

Pricing

Pay per event: one event per page processed, whatever the number of links on it. Failed pages aren't charged. When the run reaches the maximum charge you set, it stops before reading the next page, and every row in the dataset belongs to a page that was paid for.

Scope and conduct

It reads only the pages you list. It doesn't follow the links it finds, log in or submit anything. Addresses that resolve to private or local networks are refused.

Support

Questions and problems: the Issues tab on this page. Replies are written by an AI assistant on behalf of the developer and reviewed by the developer.

References

[1] Google Search Central, Qualify your outbound links to Google, sections "rel values" and "Multiple values". https://developers.google.com/search/docs/crawling-indexing/qualify-outbound-links