Re: master 1b71c995da: Avoid binding mouse-1 in xref when mouse-1 doesn't follow links

Re: master 1b71c995da: Avoid binding mouse-1 in xref when mouse-1 doesn't follow links

[Stefan Monnier]

> @@ -965,7 +965,9 @@ beginning of the line."
>  
>  (defvar xref--button-map
>    (let ((map (make-sparse-keymap)))
> -    (define-key map [mouse-1] #'xref-goto-xref)
> +    (when mouse-1-click-follows-link
> +      (define-key map [mouse-1] #'xref-goto-xref))
> +    (define-key map [follow-link] 'mouse-face)
>      (define-key map [mouse-2] #'xref-select-and-show-xref)
>      map))

Hmm... this won't adjust if `mouse-1-click-follows-link` is changed
later on.  Why do we need this `mouse-1` binding?


        Stefan

Re: master 1b71c995da: Avoid binding mouse-1 in xref when mouse-1 doesn't follow links

[Lars Ingebrigtsen]

Stefan Monnier <monnier@iro.umontreal.ca> writes:

> Hmm... this won't adjust if `mouse-1-click-follows-link` is changed
> later on.  Why do we need this `mouse-1` binding?

Dmitry seems to like the two different actions (which only differ in
whether the other buffer is selected or not).  I'd be fine with removing
the mouse-1 binding.

-- 
(domestic pets only, the antidote for overdose, milk.)
   bloggy blog: http://lars.ingebrigtsen.no

RE: [External] : Re: master 1b71c995da: Avoid binding mouse-1 in xref when mouse-1 doesn't follow links

[Drew Adams]

> > Hmm... this won't adjust if `mouse-1-click-follows-link` is changed
> > later on.  Why do we need this `mouse-1` binding?
> 
> Dmitry seems to like the two different actions (which only differ in
> whether the other buffer is selected or not).  I'd be fine with
> removing the mouse-1 binding.

Bug #35353:

 I use `mouse-1-click-follows-link' = nil.
 (I use `mouse-2', not `mouse-1', to follow
 clicked links, buttons, etc.)  But this
 setting seems to have no effect in buffer `*xref*'.

 Except by clicking on a file line (it seems),
 I find it impossible to click `mouse-1' without
 having Emacs follow a link; impossible to set
 point in the buffer using `mouse-1'; no way to
 just click buffer text to select its frame.
 What's that all about?

For which you've said:

 Lars>> we don't honor mouse-1-click-follows-link.
 Eli> FWIW, I don't see that as a serious issue in
      this case.  Feel free to close as wontfix,
      if you want.

But you said today that you fixed (this part of)
the reported bug, for Emacs 29.  I hope so.

Re: master 1b71c995da: Avoid binding mouse-1 in xref when mouse-1 doesn't follow links

[Dmitry Gutov]

On 29.04.2022 16:16, Lars Ingebrigtsen wrote:
> Stefan Monnier<monnier@iro.umontreal.ca>  writes:
> 
>> Hmm... this won't adjust if `mouse-1-click-follows-link` is changed
>> later on.  Why do we need this `mouse-1` binding?
> Dmitry seems to like the two different actions (which only differ in
> whether the other buffer is selected or not).  I'd be fine with removing
> the mouse-1 binding.

TBH, I don't really mind either way. I don't use the mouse in these 
buffers (and rarely do that at all in Emacs).

I proposed a slightly different change in the past 
(https://debbugs.gnu.org/cgi/bugreport.cgi?bug=35353#77), but if other 
similar modes (like Grep and Occur) behave differently and consistently 
between themselves, it probably makes sense to follow their example.

Re: master 1b71c995da: Avoid binding mouse-1 in xref when mouse-1 doesn't follow links

[Lars Ingebrigtsen]

Dmitry Gutov <dgutov@yandex.ru> writes:

>> Dmitry seems to like the two different actions (which only differ in
>> whether the other buffer is selected or not).  I'd be fine with removing
>> the mouse-1 binding.
>
> TBH, I don't really mind either way. I don't use the mouse in these
> buffers (and rarely do that at all in Emacs).
>
> I proposed a slightly different change in the past
> (https://debbugs.gnu.org/cgi/bugreport.cgi?bug=35353#77), but if other
> similar modes (like Grep and Occur) behave differently and
> consistently between themselves, it probably makes sense to follow
> their example.

So I've now removed the mouse-1 binding to make this map work like most
other buttons.  (I have no opinion as to whether to keep xref-goto-xref
or xref-select-and-show-xref as the binding, so I kept the mouse-2
binding.)

-- 
(domestic pets only, the antidote for overdose, milk.)
   bloggy blog: http://lars.ingebrigtsen.no

Re: master 1b71c995da: Avoid binding mouse-1 in xref when mouse-1 doesn't follow links

[Dmitry Gutov]

On 30.04.2022 14:39, Lars Ingebrigtsen wrote:
> (I have no opinion as to whether to keep xref-goto-xref
> or xref-select-and-show-xref as the binding, so I kept the mouse-2
> binding.)

If our goal is consistency with Grep, I suppose we should pick the 
command that's more similar to its behavior as well?

AFAICS mouse click in Grep selects the location's window in the end. 
That's the behavior of 'xref-goto-xref'.

Re: master 1b71c995da: Avoid binding mouse-1 in xref when mouse-1 doesn't follow links

[Lars Ingebrigtsen]

Dmitry Gutov <dgutov@yandex.ru> writes:

> If our goal is consistency with Grep, I suppose we should pick the
> command that's more similar to its behavior as well?
>
> AFAICS mouse click in Grep selects the location's window in the
> end. That's the behavior of 'xref-goto-xref'.

It makes sense to be consistent here, so I've switched the xref mouse
action to xref-goto-xref, then.

-- 
(domestic pets only, the antidote for overdose, milk.)
   bloggy blog: http://lars.ingebrigtsen.no

Re: master 1b71c995da: Avoid binding mouse-1 in xref when mouse-1 doesn't follow links

[Stefan Monnier]

Lars Ingebrigtsen [2022-04-29 15:16:07] wrote:
> Stefan Monnier <monnier@iro.umontreal.ca> writes:
>> Hmm... this won't adjust if `mouse-1-click-follows-link` is changed
>> later on.  Why do we need this `mouse-1` binding?
> Dmitry seems to like the two different actions (which only differ in
> whether the other buffer is selected or not).  I'd be fine with removing
> the mouse-1 binding.

I think the benefit of using the "standard" behavior (i.e. only have
a mouse-2 binding and rely on `mouse-1-click-follows-link` to make it
available on `mouse-1`) outweighs the benefit of choosing between those
two slightly different behaviors by opting for one binding or the other.


        Stefan