Re: GConf and bonobo-conf are now obsolete



On 31 Jul 2001 10:54:34 +0800, James Henstridge wrote:
> On 30 Jul 2001, Seth Nickell wrote:
> 
> [snip]
> > char foobar[512];
> ...
> > sprintf (&foobar, "%s%s%s%s\n", g_get_home_dir(), 	          ".foo/",
> > "bar/foo", "bar");
> 
> won't this result in a segfault? (or overwriting random memory).  &foobar
> is a (char **) -- you should just be passing in foobar.  With code like
> this, the user's preferences will get lost, which we don't want.

You only need to test code if it wasn't perfect in the first place. I do
not test my code because it is perfect. Therefore my code is perfect,
and does not need to be tested.

James, need I remind you that sometimes the user is wrong? Its a
valuable feature to ignore some of their so called "preferences". An RMS
GUADEC quote comes to mind...

-Seth






[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]