FR: Replace with canonical URL

[harmtemolder]

Describe a related problem (optional)

No response

Describe your suggested feature

Webpages may define a canonical URL of their content (<link rel="canonical" href="https://…" />, see https://en.wikipedia.org/wiki/Canonical_link_element). I know of at least one blog that uses a unique ID in their URL paths to identify the user, but all users' versions of a single article all point to the same canonical URL. Sharing/opening that instead will effectively block this form of tracking. This would probably involve requesting the page and parsing its HTML to extract the <link> element.

Describe alternatives you've considered for your suggested feature

No response

Other details

No response

Acknowledgements

• I have searched the existing issues and this is a new ticket, NOT a duplicate or related to another open issue.
• I have written a short but informative title.
• I will fill out all of the requested information in this form.

[TrianguloY]

The Status code module only detects redirections based on the request headers, redirections based on the content (either by html link or by javascript) are not. Originally the app even performed a HEAD, which according to the documentation should be 'the same as a GET but only returning the headers'. Some (a lot of) servers don't follow the documentation, so I had to change it to really do a GET. In any case the actual body is not downloaded.

Adding redirections based on javascript would probably be disastrous (I won't mind adding it with a disabled-by-default toggle with a BIG warning), but redirections from headers should not be as problematic, so I think could be a nice addition. Not sure if as a separate module or inside the Status module...probably as separate.

[harmtemolder]

Sounds good. Let me know if I can help define or test or anything.

To clarify: the canonical link tag by itself does not trigger a redirect. It's a signal for search engine spiders to avoid duplicates of the same page, to focus all value of all URL variants of a page to a single canonical URL. A user will not notice anything in their address bar or anywhere else.

[TrianguloY]

Oh, I see.

In that case I think this could be merged with the idea of having a module to display the metadata tags (the image, text, etc). A good first easy implementation would be a module that just requests the data, extracts the http headers (and only those) and displays them.

[Ultrabenosaurus]

Just an FYI that link[rel=canonical] is also used by content thief platforms like MSN news and Yahoo news, and userscripts can redirect you to the real source - though this does rely on you visiting the "bad" link in your browser first.

I've been using one for MSN news and recently forked it to add Yahoo news support, which I'll send back upstream after some more rigorous testing.

https://github.com/pr0xim1ty/msnredirect

https://github.com/Ultrabenosaurus/msnredirect/blob/patch-1/msnredirect.user.js

Won't help with adding this feature to URLCheck, but at least provides a stopgap solution for some users.