Something causes multi_name to go to wrong history items.
1. clean library, clean image (no sidecar)
2. load image into darktable
3. add some spot correction
4. add new instance of spot correction, add some more correction
5. exit darktable
6. load image into darktable
There will be two "spot correction", both without any MI name, and flip and sharpen will both have MI name of "1"
If this is yet another regression of mine, i guess i will mostly step away from dt, for some time.
Use a more consistent XMP schema for history
This breaks backwards compatibility of sidecar files. It has the big
benefit that things like empty multi_name (or even empty params) would
be supported. Next step is to add support for human readable params in
XMP using module introspection.
Fixes #11124 and probably others.
#4 Updated by Tobias Ellinghaus over 3 years ago
- % Done changed from 20 to 100
- Status changed from Triaged to Fixed
Applied in changeset darktable|829fb7d7fdd03b56d96e32ba9562fe0fedd403ba.