GNU bug report logs - #74361
[PATCH] New option xref-navigation-display-window-action

Previous Next

Full log

Message #47 received at 74361 <at> debbugs.gnu.org (full text, mbox):

From: Dmitry Gutov <dmitry <at> gutov.dev>
To: Eli Zaretskii <eliz <at> gnu.org>
Cc: rudalics <at> gmx.at, 74361 <at> debbugs.gnu.org, juri <at> linkov.net
Subject: Re: bug#74361: [PATCH] New option
 xref-navigation-display-window-action
Date: Tue, 19 Nov 2024 03:21:26 +0200

On 18/11/2024 19:03, Eli Zaretskii wrote:

>> IIUC your suggestion was:
>>
>>     If non-nil, the `display-buffer' action for showing results of Xref
>> commands.
>>
>> And those are commands belonging to the package (for the moment), and as
>> such fitting the description.
> 
> Not in my book (they are replacement commands), but feel free to find
> better text.

Okay, if the phrasing still makes sense to you with the above details, 
let's use this version.

>>>> Also, there is a nuance: when the Xref buffer itself is shown (i.e. when
>>>> there are multiple locations matching a xref-find-definitions search),
>>>> we're not going to pass (category . xref) to display-buffer either - it
>>>> is reserved for displaying the buffers of destination locations.
>>>
>>> And this is a separate issue, not related to the doc string?
>>
>> Still about the doc string.
> 
> But not important enough to affect the first line of it.
> 
>> If we say "showing results of Xref
>> commands", then the Xref buffer with the list of locations also matches
>> that description, doesn't it? Or rather it might be the first thing a
>> user would think of - but our customization wouldn't apply to it
>> immediately.
> 
> The rest of the doc string could explain this subtlety.

Fair point, thanks.

Next step is to find the appropriate place for it. Maybe not a docstring 
if the user option to customize will be the global one 
(display-buffer-alist), but either in Commentary or somewhere in the manual.

Speaking of the reference that we used for this solution, it seems that

  (category . comint)

is only mentioned in NEWS and in the obsoletion message for 
'display-comint-buffer-action'. There is also a hit for it in the 
manual, but only in the form "If the caller display-buffer passes a 
category as a symbol ...", not saying exactly that comint modes do or 
should do that. Maybe that's something to be fixed as well.

Previous Next

GNU bug tracking system
Copyright (C) 1999 Darren O. Benham, 1997,2003 nCipher Corporation Ltd, 1994-97 Ian Jackson.