detect URLs better for "Open Link"




2 years ago
11 months ago


(Reporter: Dan Jacobson, Unassigned)


48 Branch

Firefox Tracking Flags

(Not tracked)




2 years ago
User Agent: Mozilla/5.0 (X11; Linux i686; rv:48.0) Gecko/20100101 Firefox/48.0
Build ID: 20160610001751

Steps to reproduce:

and right clicked.

Actual results:

No "Open Link" choices.

Expected results:

Should have acted just like it does when one highlights .

By the way, please highlight the following entire block:

and right click and then choose Open Link. Well it tries to take you to a URL composed of all five... clearly wrong.

And before you click Open Link, you don't know where it is exactly planning to go... maybe the first one? you think.

Anyway you should detect that this block of five is no longer a (single) URL, and not offer Open Link.


2 years ago
Component: Untriaged → Menus

Comment 1

2 years ago
OK, in addition please try in the comment text box below, pasting or typing
e.g., or just etc. and then highlight it, and right click (all before pressing "Save Changes", which don't press).

As you see, if the URL is part of text entry, then it is completely undetected, for no good reason! We have to press Save Changes below to turn it into a read-only part of a web page before URL detection is turned on!

Comment 2

2 years ago
By the way, the "Paste and Go" menu item in the URL bar knows that both and are both URLs, so maybe just reuse that code.

However Paste and Go also thinks our big block of five, along with its newlines even, is also an URL...


2 years ago
Priority: -- → P4

Comment 3

11 months ago
I'm having the same issue. Text links which don't have an http:// at the beginning and do have a path after the domain don't work. For example:

But domains only, do work:

And http:// with a path also works:

It would also be great if some more "stranger" cases were supported. Some websites (forums for example) purposely "obfuscate" URLs so that they are not automatically linkified. For example:
e x a m p l e . c o m

In the first case, all spaces in the domain part should be ignored, and in the second, if the protocol can't be recognised, default to http. This is basically what Fetch Text URL extension does:
You need to log in before you can comment on or make changes to this bug.