This is the mail archive of the
glibc-bugs@sourceware.org
mailing list for the glibc project.
[Bug stdio/6530] *printf() and incomplete multibyte sequences returns -1 bogusly
- From: "jrnieder at gmail dot com" <sourceware-bugzilla at sourceware dot org>
- To: glibc-bugs at sources dot redhat dot com
- Date: Sat, 23 Jun 2012 04:22:10 +0000
- Subject: [Bug stdio/6530] *printf() and incomplete multibyte sequences returns -1 bogusly
- Auto-submitted: auto-generated
- References: <bug-6530-131@http.sourceware.org/bugzilla/>
http://sourceware.org/bugzilla/show_bug.cgi?id=6530
--- Comment #18 from Jonathan Nieder <jrnieder at gmail dot com> 2012-06-23 04:22:10 UTC ---
(In reply to comment #17)
> Apologies for jumping on your reply. I did read the whole thing but didn't get
> what you intended the message to be, and the whole history of this bug has been
> folks misquoting text that has nothing to do with the non-long %s format in the
> non-wide printf functions to justify the buggy behavior... :-(
I had been assuming that the bug was introduced by copying the wprintf()
implementation for reuse in printf() --- i.e., no misunderstanding of the
standard but just simple operator error.
But actually the patch introducing the regression (40c014b3, "Correct
handling of multibyte character strings in %s format with precision",
2000-07-22) only affected vsnprintf() and co. The changelog comment
makes no sense. There is no explanation on libc-alpha@ or libc-hacker@
from around that time. It looks like a simple, unilateral, bad decision.
There's a patch identical to yours complete with changelog and copyright
assignment at <http://sourceware.org/ml/libc-hacker/2010-08/msg00009.html>.
It just reverts 40c014b3. It was ignored. A story from a worse time.
Thanks for fixing it.
--
Configure bugmail: http://sourceware.org/bugzilla/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.