Bug#947182: libjpeg-turbo-progs: jpegexiforient fails if APP1 doesn't occur immediately after JFIF header

2020-08-29 Thread Akkana Peck
I know this bug was closed as wontfix, but it should be noted that this means that exifautotran can no longer be used on images that were resized with ImageMagick, GraphicsMagick, or PIL (Python Imaging Library): they all produce files that exifautotrans silently ignores, breaking scripts that

Bug#947182: libjpeg-turbo-progs: jpegexiforient fails if APP1 doesn't occur immediately after JFIF header

2019-12-22 Thread Mike Gabriel
Control: tag -1 wontfix Hi all, On So 22 Dez 2019 17:57:27 CET, Ondřej Surý wrote: Hi Anselm, see https://datatracker.ietf.org/doc/draft-iab-protocol-maintenance/ why this is a very bad idea. I am not going to apply the patch to the Debian, nor should it be applied upstream. The

Bug#947182: libjpeg-turbo-progs: jpegexiforient fails if APP1 doesn't occur immediately after JFIF header

2019-12-22 Thread Ondřej Surý
Hi Anselm, see https://datatracker.ietf.org/doc/draft-iab-protocol-maintenance/ why this is a very bad idea. I am not going to apply the patch to the Debian, nor should it be applied upstream. The program violation specification (Digikam) should be fixed instead. Applying robustness

Bug#947182: libjpeg-turbo-progs: jpegexiforient fails if APP1 doesn't occur immediately after JFIF header

2019-12-22 Thread Anselm Lingnau
Package: libjpeg-turbo-progs Version: 1:1.5.2-2+b1 Severity: normal Tags: patch Dear Maintainer, according to the EXIF standard, the APP1 segment (which contains the EXIF data) must occur immediately after the SOI (start of image) marker at the beginning of a JFIF file. The jpegexiforient