Subject: Re: scan_ffs from OpenBSD ported and improved.
To: None <tech-userlevel@NetBSD.org>
From: der Mouse <mouse@Rodents.Montreal.QC.CA>
List: tech-userlevel
Date: 06/06/2005 23:27:16
> I leave them as is because I think they are better to understand IMHO
> (der mouse said so too).

That's not quite what Mouse said, nor, even if it were, does it
necessarily mean you should do as I suggest.

First, I didn't say anything about your particular code.  I said that
there are cases where breaking the rules produces clearer code than
following them, with the associated comment that turning the guidelines
into straitjackets is a Bad Thing; I haven't looked at your code, so I
don't have an opinion as to whether it's such an instance.

Second, I am not authoritative about this stuff with respect to what
goes into NetBSD, not by any stretch of the imagination.

There are a number of points on which I disagree with
/usr/share/misc/style; dogmatism about whitespace was just the one I
happened to mention.  There are only two points on which I really think
it is out-and-out wrong: (1) the use of old-style definitions, since
they are incompatible, in general, with prototyped declarations (I
think gcc is wrong to allow a prototype to rewrite an old-style
definition into the related prototype definition...it's an invitation
to write broken code by making it appear to work) and (2) encouraging
the use of NULL, which my code-reading experience indicates is
significantly harder to use correctly than bare 0.

If anyone really cares, I can do a full ripping-apart of .../style, but
this probably is not the thread to do so in - indeed, I'm not sure it
should be done on any of the lists at all.

/~\ The ASCII				der Mouse
\ / Ribbon Campaign
 X  Against HTML	       mouse@rodents.montreal.qc.ca
/ \ Email!	     7D C8 61 52 5D E7 2D 39  4E F1 31 3E E8 B3 27 4B