emacs-devel
[Top][All Lists]
Advanced

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

Re: Emacs rewrite in a maintainable language


From: Andy Moreton
Subject: Re: Emacs rewrite in a maintainable language
Date: Wed, 14 Oct 2015 14:28:59 +0100
User-agent: Gnus/5.13 (Gnus v5.13) Emacs/25.0.50 (windows-nt)

On Wed 14 Oct 2015, Oleh Krehel wrote:

> Andy Moreton <address@hidden> writes:
>
>> On Tue 13 Oct 2015, Oleh Krehel wrote:
>>
>>> Paul Eggert <address@hidden> writes:
>>>
>>>> Oleh Krehel wrote:
>>>>> I'd like to switch all const
>>>>> pointers to a single "const char *" style.
>>>>
>>>> Let's not. It's more consistent to put 'const' after the type it
>>>> modifies, and if we're going to have a "standard" style, that's the
>>>> one we should have.
>>>
>>> "const char *" style is used in 85.9% of the cases. If you want to
>>> revert those uses to "char const *" I'll support you, since I want the
>>> consistency of a single style.
>>
>> Consistency is helpful, but pointless churn makes version history less
>> useful, so please don't make this kind of change unless you are also
>> making a semantic change to that area of the code.
>
> One commit is pointless churn?
>
>>>> But really, we have better things to do.
>>>
>>> I don't. I have no problem spending time to replace "char const *" into
>>> "const char *". Browsing through a code base that I want to get familiar
>>> with and fixing a small annoyance doesn't sound bad to me.
>>
>> Pointless churn. Find something that fixes an bug, or adds a useful new
>> feature, and work on that. You will find it more rewarding, and other
>> users and developers will find your contribution more helpful.
>
> I don't care about pleasing the user or developer gods. I want to do
> what's helpful for me, and I think that change would be. I'm offering to
> do the work for free, and currently I'm not offering any other work as
> an alternative. You either want the improvement or not. The concerns for
> opportunity cost are irrelevant here.

Changes of this nature often turn into bikeshedding (where developers
disagree about the one true style), or an endless series of small
changes to modify code style (if style checker tools are not used).

All of this make it harder to follow the history when trying to find how
the behaviour has changed across versions. That is a cost that comes
with an otherwise harmless change.

> Very often I see /harmless/ suggestions turned down as not worth it,
> both on this list and on debbugs. This isn't a good way to move forward
> if we want to expand the base of developers and bug-reporting users.

It is not cost-free, as noted above. By all means update the style to be
consistent across the codebase, as long as you introduce a style checker
that can automatically enforce the agreed style.

As there have been dissenting voices in this thread, it seems your
preferred style does not meet with universal approval.

    AndyM




reply via email to

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