cvs commit: src/include regex.h

Alexey Dokuchaev danfe at nsu.ru
Fri Dec 19 18:51:06 PST 2003


On Thu, Dec 18, 2003 at 03:49:19PM +1100, Bruce Evans wrote:
> On Tue, 16 Dec 2003, Nate Lawson wrote:
> 
> > On Tue, 16 Dec 2003, Jordan K. Hubbard wrote:
> > >   Modified files:
> > >     include              regex.h
> > >   Log:
> > >   Conformance: Add REG_ENOSYS (see http://www.opengroup.org/onlinepubs/007904975/basedefs/regex.h.html)
> > >
> > >   Revision  Changes    Path
> > >   1.7       +1 -0      src/include/regex.h
> > >
> > >
> > > Index: src/include/regex.h
> > > diff -u src/include/regex.h:1.6 src/include/regex.h:1.7
> > > --- src/include/regex.h:1.6	Thu Oct  3 11:12:03 2002
> > > +++ src/include/regex.h	Tue Dec 16 18:56:28 2003
> > > @@ -75,6 +75,7 @@
> > >  #define	REG_DUMP	0200
> > >
> > >  /* regerror() flags */
> > > +#define REG_ENOSYS	-1	/* Reserved */
> > >  #define	REG_NOMATCH	 1
> > >  #define	REG_BADPAT	 2
> > >  #define	REG_ECOLLATE	 3
> >
> > Same.
> 
> Not quite the same :-).  This commit shares the following other bugs with
> the corresponding commit to fnmatch.h:
> 
> - missing parentheses around macro.
> - abnormal comment style.  The comment style in this file was perfectly
>   abnormal (unlike in fnmatch.h where it was normal).  The new comment
>   matches neither, since it begins the sentence fragment with a capital
>   letter (like the normal style but unlike the rest of the file) and
>   it doesn't end the sentence fragment with a "." (unlike the normal
>   style but like the rest of the file).
> 
> ... but doesn't share the following:
> 
> - namespace pollution.  There is none here since REG_ENOSYS is only
>   obsolescent; it is not an XSI extension.
> - new macro in a section all by itself.  Here it is even numerically sorted.

What's really cool about Bruce is this in-depth style(9) picking.  Oh I
just love to see people care for perfection.  8-) 8-)

./danfe

P.S.  And this is actually one of the reasons FreeBSD *is* better than
<censored/>.



More information about the cvs-src mailing list