unofficial mirror of help-gnu-emacs@gnu.org
 help / color / mirror / Atom feed
* Update point in visible but non-selected buffer/window
@ 2016-10-21  8:39 Joost Kremers
  2016-10-21 15:16 ` Eli Zaretskii
  0 siblings, 1 reply; 3+ messages in thread
From: Joost Kremers @ 2016-10-21  8:39 UTC (permalink / raw)
  To: Help Gnu Emacs mailing list

Hi all,

I just found out the hard way that changing point programmatically 
(i.e., from within an Elisp program) in a buffer that is visible 
in a non-selected window doesn't (necessarily?) update the 
position of the cursor, and when the window is selected again, 
point is reset to the cursor position.

To be more explicit:

Suppose I have two buffers, A and B, being displayed in two 
windows, W_a and W_b. W_a is selected, point is at the beginning 
of the buffer. Now I select W_b. In W_a, the cursor is still 
visible as a hollow box. In W_b, I execute some command that moves 
point in buffer A (specifically, by using `re-search-forward' 
inside a `with-current-buffer'). The cursor in W_a does not move, 
however. Still, as long as I do not select W_a, point is where 
`re-search-forward' left it (and thus does not correspond to the 
visible cursor position). For example, when I do something like:

(with-current-buffer "A" (point))

I get the result I expect, which is not the position where I see 
the (hollow) cursor.

However, if I then select window W_a, the hollow cursor is changed 
back into a box, but it remains at the position where it was all 
the time, i.e., at the beginning of the buffer, and point is 
adjusted accordingly.

I can see why this behaviour makes sense from a user's 
perspective, and I figured out that I can call `select-window' 
inside `with-current-buffer' to force the cursor position to 
change as well as point. But I haven't been able to find any info 
on this behaviour in the Elisp manual, so I was wondering if this 
has been described somewhere and whether the solution I stumbled 
upon (using `select-window') is the best way to achieve the effect 
I want, or whether there is a better, more canonical way.

TIA

Joost




-- 
Joost Kremers
Life has its moments



^ permalink raw reply	[flat|nested] 3+ messages in thread

* Re: Update point in visible but non-selected buffer/window
  2016-10-21  8:39 Update point in visible but non-selected buffer/window Joost Kremers
@ 2016-10-21 15:16 ` Eli Zaretskii
  2016-10-21 21:37   ` Joost Kremers
  0 siblings, 1 reply; 3+ messages in thread
From: Eli Zaretskii @ 2016-10-21 15:16 UTC (permalink / raw)
  To: help-gnu-emacs

> From: Joost Kremers <joostkremers@fastmail.fm>
> Date: Fri, 21 Oct 2016 10:39:37 +0200
> 
> I just found out the hard way that changing point programmatically 
> (i.e., from within an Elisp program) in a buffer that is visible 
> in a non-selected window doesn't (necessarily?) update the 
> position of the cursor, and when the window is selected again, 
> point is reset to the cursor position.

This is by design.  That's what window-point is about.  When you
select a window, the buffer shown in that window gets its point reset
to the place of window-point, and the cursor is positioned
accordingly.

> I can see why this behaviour makes sense from a user's 
> perspective, and I figured out that I can call `select-window' 
> inside `with-current-buffer' to force the cursor position to 
> change as well as point. But I haven't been able to find any info 
> on this behaviour in the Elisp manual, so I was wondering if this 
> has been described somewhere and whether the solution I stumbled 
> upon (using `select-window') is the best way to achieve the effect 
> I want, or whether there is a better, more canonical way.

See window-point and set-window-point.



^ permalink raw reply	[flat|nested] 3+ messages in thread

* Re: Update point in visible but non-selected buffer/window
  2016-10-21 15:16 ` Eli Zaretskii
@ 2016-10-21 21:37   ` Joost Kremers
  0 siblings, 0 replies; 3+ messages in thread
From: Joost Kremers @ 2016-10-21 21:37 UTC (permalink / raw)
  To: Eli Zaretskii; +Cc: help-gnu-emacs


On Fri, Oct 21 2016, Eli Zaretskii wrote:
>> From: Joost Kremers <joostkremers@fastmail.fm>
>> Date: Fri, 21 Oct 2016 10:39:37 +0200
>> 
>> I just found out the hard way that changing point 
>> programmatically 
>> (i.e., from within an Elisp program) in a buffer that is 
>> visible 
>> in a non-selected window doesn't (necessarily?) update the 
>> position of the cursor, and when the window is selected again, 
>> point is reset to the cursor position.
>
> This is by design.  That's what window-point is about.  When you
> select a window, the buffer shown in that window gets its point 
> reset
> to the place of window-point, and the cursor is positioned
> accordingly.

Yeah, I assumed it was by design, because it makes sense.

>> I can see why this behaviour makes sense from a user's 
>> perspective, and I figured out that I can call `select-window' 
>> inside `with-current-buffer' to force the cursor position to 
>> change as well as point. But I haven't been able to find any 
>> info 
>> on this behaviour in the Elisp manual, so I was wondering if 
>> this 
>> has been described somewhere and whether the solution I 
>> stumbled 
>> upon (using `select-window') is the best way to achieve the 
>> effect 
>> I want, or whether there is a better, more canonical way.
>
> See window-point and set-window-point.

Thanks. I missed that.

-- 
Joost Kremers
Life has its moments



^ permalink raw reply	[flat|nested] 3+ messages in thread

end of thread, other threads:[~2016-10-21 21:37 UTC | newest]

Thread overview: 3+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2016-10-21  8:39 Update point in visible but non-selected buffer/window Joost Kremers
2016-10-21 15:16 ` Eli Zaretskii
2016-10-21 21:37   ` Joost Kremers

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).