[Top][All Lists]

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

Re: Compilation warnings in mouse.el

From: Eli Zaretskii
Subject: Re: Compilation warnings in mouse.el
Date: Wed, 13 Jul 2016 17:30:04 +0300

> Cc: address@hidden
> From: Dmitry Gutov <address@hidden>
> Date: Wed, 13 Jul 2016 01:56:30 +0300
> On 07/12/2016 08:04 AM, Eli Zaretskii wrote:
> > Unlike Stefan, I think we should add :group to every defcustom.
> The additions to each file just have to be consistent with what's 
> already there. The presently discussed patch, wasn't.

But if each defcustom has its :group, then the need for consistency is
no longer a requirement, is it?

> Speaking of personal preference, however, I'd rather using the `mouse' 
> group didn't work, in the absence of a defgroup somewhere.

If we can do something to that effect, I don't think I will object.

> > Having them mysteriously missing from some of them is a time bomb:
> > remove or add enough defcustoms in the same file, and you have a bug.
> How would removing a defcustom, or merely adding one, lead to a bug?

Well, not bug, a compilation warning that suddenly appears out of
nowhere for code that was there for a long time.

> The new defcustom might have :group unspecified, and if there's no 
> defgroup in the current file, it would become a problem, but that's 
> unrelated to any defcustom's already present in the file.

AFAIU, extant defcustoms could start causing warnings, given some
changes in other defcustom's.  Am I mistaken?

In short, I think having one defcustom depend on another is bad for

reply via email to

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