Project

General

Profile

Bug #11611

Segfault when soft proofing against Linear XYZ

Added by Ċ½ilvinas Ċ½altiena about 2 years ago. Updated about 2 years ago.

Status:
Fixed
Priority:
Low
Assignee:
-
Category:
Darkroom
Target version:
Start date:
05/13/2017
Due date:
% Done:

100%

Affected Version:
2.2.4
System:
all
bitness:
64-bit
hardware architecture:
amd64/x86

Description

Steps to reproduce
1) put allow_lab_output=TRUE in darktablerc (it turns not only lab output, but linear XYZ too)
2) choose linear XYZ as soft proofing profile
3) turn soft proofing on.

Reaction: segfault with message "Couldn't link the profiles" (in console).

Associated revisions

Revision 5abc16c9
Added by Tobias Ellinghaus about 2 years ago

Support XYZ as prrofing profile

This fixes #11611.

Revision 580ff840
Added by Tobias Ellinghaus about 2 years ago

Support XYZ as prrofing profile

This fixes #11611.

(cherry picked from commit 5abc16c95fefd9e8cb1491948ef0290b701ff7b3)

History

#1 Updated by Tobias Ellinghaus about 2 years ago

  • System changed from other GNU/Linux to all
  • % Done changed from 10 to 20
  • Status changed from Confirmed to Triaged

For some reason lcms2 doesn't like to make a transform when the softproofing profile is linear XYZ. It works with Lab though. Ideally we want to have a way to filter out unsuited profiles and not add them to the proofing profiles list. And as a 2nd step we want to make sure that commit_params in colorout always provides a valid way to transform the image for process(). And last but not least, process() should never crash. Maybe just turning all pixels pink would be a good fallback measure. It's obviously the wrong result so people won't be tricked to assume that what they see is actually the correct result. Thoughts & comments?

#2 Updated by Tobias Ellinghaus about 2 years ago

  • % Done changed from 20 to 100
  • Status changed from Triaged to Fixed

#3 Updated by Roman Lebedev about 2 years ago

  • Target version set to 2.4.0

Also available in: Atom PDF