Re: mouse-1-click-follows-link is hard to use

[Kim F. Storm]

Miles Bader <address@hidden> writes:

> On Tue, 08 Mar 2005 21:55:48 +0100, Kim F. Storm <address@hidden> wrote:
>> The reason for the delay is to be able to detect a double mouse-1 click.
>> 
>> Perhaps it shouldn't do that ...   WDOT?
>
> Is it different than Emacs normally deals with double clicks?

Yes, a double click is normally treated as two commands:

a single click action followed by the double click action.

Now, if the single click (mouse-1 => mouse-2 mapping) changes buffer,
the double click will happen in the wrong buffer...

The solution could be to somehow ignore a subsequent double click
after following the link, but I think that requires C-level support
(to inhibit the "mouse double click" detection).  Not a big deal
of course.


>
> BTW (my real reason for posting to this thread :-) I like
> mouse-1-follow-link generally (so don't want to disable it), but there
> are some cases where it is annoying.
>
> For instance, in a Gnus *Group* buffer, clicking with mouse-1 enters
> the group.  

If you click the group name (which has mouse-face and a hand pointer),
yes.   Just click anywhere else in the line to navigate in the buffer
itself (or use long click)...

>             Sometimes this is OK, but often I want to merely position
> point there so I can use one of the alternate entry commands (e.g.,
> `=' or `C-u RET').  A similar thing happens in Gnus summary buffers.

Likewise, click on the subject rather than the name.  The mouse-face
clearly indicates when mouse-1 will do more than set point...

> I'm not sure what the thing to do here is, whether Gnus is
> intentionally enabling mouse-1-follows-link, whether it's inadvertant,
> or what.

It is enabled intentionally (on mouse-face links only).

>
> If it's being enabled automagically by Emacs, perhaps that mechanism
> needs adjustment (or Gnus needs to tweak it).  I guess the reasons why
> m1fl is inappropriate in these places are (1) there are many commands
> one can use (and are commonly uesd) on each "link", and (2) they don't
> look like links anyway.

Why isn't mouse-face and hand pointer enough to differentiate them?

-- 
Kim F. Storm <address@hidden> http://www.cua.dk