Date: Tue, 9 Oct 2001 12:49:50 -0400
Message-Id: <200110091649.f99Gnow11464@envy.delorie.com>
From: DJ Delorie <dj AT delorie DOT com>
To: djgpp-workers AT delorie DOT com
In-reply-to: <007601c150df$50c4fa20$3e8224d5@zephyr>
	(ebotcazou AT libertysurf DOT fr)
Subject: Re: _findfirst() patch
References: <200110091438 DOT QAA23042 AT lws256 DOT lu DOT erisoft DOT se> <007601c150df$50c4fa20$3e8224d5 AT zephyr>
Reply-To: djgpp-workers AT delorie DOT com
Errors-To: nobody AT delorie DOT com
X-Mailing-List: djgpp-workers AT delorie DOT com
X-Unsubscribes-To: listserv AT delorie DOT com
Precedence: bulk


> Aren't there people who use their WAP mobile phone to read the list ? On the
> Allegro Developers list, big patches are not welcomed. But ok, now I know...

If my mail server rejects your mail as being too big, *that* is the
limit.  It's currently 150k.  Since part of the purpose of this list
is to review patches, you can assume that we are OK with patches being
posted in a form that is most convenient for review.

> Afaics, src/libc/dos/errno/doserr2e.c:__doserr_to_errno() is not very
> documented either. What is the rule ?

If it's in a public header, it should be documented.  If it's a global
symbol, it probably should be documented also.  Exceptions for
functions that just aren't useful to apps, or that we wish for apps to
avoid using.  Otherwise, the intentionally vague rule is "do what
makes sense, and is best for the end user."

> > The formal arguments in prototypes should have "_" prepended.
> 
> I had initially intended to put the underscore but then I remarked it was
> not really mandatory (chsize(), snprintf(), vsnprintf(), ...).

It is mandatory in declarations in public headers (anything in
include/...).  If you find cases that are missing leading underscores,
feel free to fix them.  We've had problems with arguments in headers
matching #define tokens and causing parse errors.