2011.01.11 03:17 "Re: [Tiff] [libtiff v3.9.4] TIFFClientOpen() & 'w' mode", by IMoL
I have been informed by The Doctor that it's in bugzilla [didn't see that there was a bugzilla for libtiff to check]:
Apparently fixed in repo on 2010-12-06.
On Mon, Jan 10, 2011 at 12:53 PM, Olivier Paquet <firstname.lastname@example.org> wrote:
On Mon, Jan 10, 2011 at 11:47 AM, IMoL <email@example.com> wrote:
If I'm reading the code correctly, tif->tif_mode & O_TRUNC in the if statement will always be false. that be m & O_TRUNC? So shouldn't
It certainly looks like that was the intent. As it is written now, there is a bogus attempt to read the header from the truncated file which should fail for reasonable implementations of the read proc. I don't think there would be any harm in "fixing" it but I'd appreciate another set of eyes to look that over.