Skip site navigation (1)Skip section navigation (2)
Date:      Sun, 6 Jul 2003 21:20:37 -0400
From:      Garance A Drosihn <drosih@rpi.edu>
To:        Alfred Perlstein <bright@mu.org>
Cc:        Philippe Charnier <charnier@FreeBSD.org>
Subject:   Re: cvs commit: src/usr.sbin/lpr/lpd lpd.c ...
Message-ID:  <p05210650bb2e737f369a@[128.113.24.47]>
In-Reply-To: <20030707003216.GI72093@elvis.mu.org>
References:  <200307061244.h66CiBqg079835@repoman.freebsd.org> <p0521064ebb2e5c8dd5e7@[128.113.24.47]> <20030707003216.GI72093@elvis.mu.org>

next in thread | previous in thread | raw e-mail | index | archive | help
At 5:32 PM -0700 7/6/03, Alfred Perlstein wrote:
>* Garance A Drosihn <drosih@rpi.edu> [030706 16:17] wrote:
>  >
>>  Speaking only for lpd, I wish you would not have added FBSDID.
>>  I had started to add FBSDID to all the source files of
>>  lpr & friends, and found that it was just annoying when I
>>  then went to build lpr on other platforms.  So, I switched
>  > back to just plain 'rcsid[] ='.
>>
>
>Isn't there a common header where you can just add:
>
>#ifndef __FBSDID
>#define __FBSDID(s)     struct __hack
>#endif
>
>?

Well, any change to use __FBSDID() has to #include <sys/cdefs.h>
to get FBSDID() to work.  Of course, /usr/include/sys/cdefs.h
does not exist on all platforms either (and is certainly not
likely to define __FBSDID even if the file does exist), so that's
just more of an annoyance when I'm doing cross-platform work.
Other than the #include for sys/cdefs.h, there *is* no include
before the reference to __FBSDID in any source file, due to the
standard ordering of lines in a source file.

I remember asking about this with a few people before I decided
to stick with rcsid instead of FBSDID(), but I don't remember
if that was in a mailing list or in private email.  It looks
like I had only committed a change to use FBSDID() in one file,
lpr/chkprintcap/skimprintcap.c, and the comment when I switched
that back to rcsid says "discussed with bde and obrien", so it
was probably just private email.

I thought I had switched more source files to use FBSDID, but
it looks like that was the only file where I had made a commit
that I later backed out.  I was probably in the middle of an
update to switch all the rest of the files when I started to
realize what an added hassle it was going to be for me.

  [ ignoring that question for the moment, why would I want
   to define __FBSDID as 'struct __hack'? ]

Yes, there are ways I can get around all of these minor issues,
but at the time it seemed like a lot of extra futzing around
for very little gain.

-- 
Garance Alistair Drosehn            =   gad@gilead.netel.rpi.edu
Senior Systems Programmer           or  gad@freebsd.org
Rensselaer Polytechnic Institute    or  drosih@rpi.edu



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?p05210650bb2e737f369a>