bug-gnu-emacs
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

bug#35224: [PATCH] include indentation size in .dir-locals


From: Basil L. Contovounesios
Subject: bug#35224: [PATCH] include indentation size in .dir-locals
Date: Thu, 11 Apr 2019 16:22:54 +0100
User-agent: Gnus/5.13 (Gnus v5.13) Emacs/27.0.50 (gnu/linux)

Konstantin Kharlamov <hi-angel@yandex.ru> writes:

> On Чт, Apr 11, 2019 at 11:58, Basil L. Contovounesios <contovob@tcd.ie> wrote:
>> Konstantin Kharlamov <hi-angel@yandex.ru> writes:
>>
>>>  On Чт, Apr 11, 2019 at 11:26, Basil L. Contovounesios <contovob@tcd.ie>
>>> wrote:
>>>
>>>>  Globally setting c-basic-offset overrides the cc-mode style inheritance
>>>>  system.  There are various ways to customise this variable in a more
>>>>  granular way, such as defining a custom style which inherits from a
>>>>  built-in style, or setting c-basic-offset locally in a mode hook.
>>>>  For details, see (info "(ccmode) Config Basics").
>>>
>>>  Oh, okay, so this works as expected. But since googling "change indentation
>>>  size" always gives using specifically c-basic-offset (e.g.
>>>  
>>> https://stackoverflow.com/questions/14939608/how-to-change-emacs-struct-indents-from-4-to-2-spaces),
>>>  I guess it's safe to assume most of Emacs users has it changed. So it's
>>> useful
>>>  to add this to dir-locals anyway.
>>
>> IMO, that's a bug in the configuration of those users because they're
>> overriding how cc-mode works, and the cc-mode manual describes in detail
>> the various ways to customise indentation.  So I don't see a need to
>> change the Emacs dir-locals-file to accommodate this use-case.
>
> Okay, but we can't ignore the fact that such usecase exists and is widely
> employed (if needed, I can throw more links to these advices on stackexchange
> site).

The widespread dissemination of synoptic, not universally applicable, or
bad advise in the wild doesn't necessarily validate any particular
use-case.

> So let me ask: is there a downside to making this change? Because if there's
> none then the change is an improvement, it's that simple.

The downside is that this is an invasive change in that it affects all
people working on the Emacs sources.  For many it may have no obvious
effect, for others it may mask problems/inconsistencies in their
configuration, and yet others it may end up annoying due to unforeseen
consequences.  A net loss, potentially.

There is nothing obviously wrong with the current Emacs dir-locals-file
in this respect (if it ain't broke don't fix it), whereas the user
customisations you refer to are arguably problematic (if it is broke fix
it).

Again, just one opinion; I won't mind if others welcome this change.

-- 
Basil





reply via email to

[Prev in Thread] Current Thread [Next in Thread]