Re: patch to change behavior of g_strsplit on empty string back to 1.2 behavior

[Joel Becker <jlbec evilplan org>]

On Wed, Jul 18, 2001 at 05:19:34PM -0700, Darin Adler wrote:
> On Wednesday, July 18, 2001, at 05:11  PM, Joel Becker wrote:
> 
> > 	AFAICT, stable behaved like perl.  If it doesn't, I'm amazed I
> > haven't seen bugs.  I've been using it like perl for the longest time.
> 
> Definitely not. I wrote a test program and everything. Stable definitely 
> turns g_strsplit("FOO=bar=baz=bletch", "=", 2) into "Foo", "bar", 
> "baz=bletch", NULL.

	Very, very eek.  From glib 1.2.8 glib.h:

/* NULL terminated string arrays.
 * g_strsplit() splits up string into max_tokens tokens at delim and
 * returns a newly allocated string array.

	That means that the returned value will have at most max_tokens
non-NULL elements.  This is what I've always gone by wrt behavior.  It
is also in CVS glib glib/strfuncs.h.
	Ok, testing shows that glib 1.2.8 as well as 1.2.10 have the
max_tokens=2 means 3 behavior.  I suspect I now have bugs :-(

Joel

Btw, It would appear that Apple Mail (or your outgoing mailservers)
don't generate Message-IDs.  That's so annoying, because being CC'd I
get two copies of the message, with different Message-IDs (gnome.org
and my local spool).  Random thought :-)

-- 

"There is no sincerer love than the love of food."  
         - George Bernard Shaw 

			http://www.jlbec.org/
			jlbec evilplan org