On Wed, 26.09.07 08:37, Padraig O'Briain ([EMAIL PROTECTED]) wrote: > > Sure. But couldn't you guys just fix Solaris' printf? Maybe as part of > > this project "Indiana" thing? I mean, this is not the first time Sun > > guys report an issue like this against a developed-on-Linux project, > > is it? > > > > No. This has been discussed before. See > http://mail.opensolaris.org/pipermail/opensolaris-code/2006-July/002972.html.
Wow, once in a while Jörg Schilling is right! The position of the Solaris devs on this is stupid. Setting a pointer to NULL is the official way how C people set something to a "disabled", "deactivated" state. Like the NULL value in SQL. As such it is clearly not a programming error if I set something to NULL. Those people on the Solaris ML repeatedly say that a string set to NULL is a programming error. This is just stupid. Setting a string to NULL is one of the most common and most of the time correct things to do. Being unable to print it is just a unnecessary PITA. The purpose of printf() is to show the state of the program to the user. It is *not* to crash on valid program states. Also, checking for NULL strings is trivial. Costs almost nothing. So, Sun now wants to make solaris as sexy as Linux for hacking Free Software. How about starting with making it a sane programming environment? Lennart -- Lennart Poettering Red Hat, Inc. lennart [at] poettering [dot] net ICQ# 11060553 http://0pointer.net/lennart/ GnuPG 0x1A015CC4 _______________________________________________ avahi mailing list [email protected] http://lists.freedesktop.org/mailman/listinfo/avahi
