Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Bug: Raster masks broken #396

Closed
2 tasks done
Annabel2 opened this issue Jan 8, 2025 · 5 comments
Closed
2 tasks done

Bug: Raster masks broken #396

Annabel2 opened this issue Jan 8, 2025 · 5 comments

Comments

@Annabel2
Copy link

Annabel2 commented Jan 8, 2025

Code of Conduct

  • I have read this document and I accept the conditions

Search previous report

  • I have searched my issue

Computer's operating system

Linux

What GPU do you have ?

No GPU

Ansel version

Ansel-0.0.0+980~ga329ee63-x86_64.AppImage

Type of binary

Pre-built packages provided by the Ansel project (.AppImage, .EXE)

What happened?

I've been working on a photo with Exposure masks which are re-used later in the pipe. There were appreciable problems which seemed resolved by appimage 980, which is great. But on adding a second Color Calibration module with a drawn mask, and re-using it in a Color balance rgb module, errors occurred. Using the option to clear all pipeline cashes, the image in the darktable sorted itself out. On pressing Escape however, the lighttable image was still wrong. On exporting, it was the incorrect image which was created as jpeg. The attached screen images show the change, the only difference is pressing Escape.

When I output I'm getting this in the Terminal window -
(AppRun.wrapped:17322): GLib-CRITICAL **: 17:18:27.553: g_date_time_unref: assertion 'datetime != NULL' failed

If I export from the darktable (correct image) rather than lighttable, unfortunately the jpeg is still wrong.

Screenshot from 2025-01-08 17-12-46--
Screenshot from 2025-01-08 17-12-40

Does your problem occur with OpenCL enabled ?

I don't know

Does your problem occur with OpenCL disabled ?

Yes

Previous Ansel version that use to work

No response

Output of ansel -d all

No response

Backtrace

No response

@aurelienpierre
Copy link
Collaborator

Could you send the picture + XMP please ?

@Annabel2
Copy link
Author

Annabel2 commented Jan 8, 2025

Uploaded to your private cloud, files-for-issue-396.zip

@aurelienpierre
Copy link
Collaborator

Reproduced

aurelienpierre added a commit that referenced this issue Jan 10, 2025
Remove inline loops and complicated branching to make the sequence more legible. Layout the ground work to troubleshoot #396 and possibly share code between GUI pipelines.
@aurelienpierre
Copy link
Collaborator

Ok the issue is more simple. Raster masks are broken.

@aurelienpierre aurelienpierre changed the title Bug: Jpeg export is wrong, mask error Bug: Raster masks broken Jan 10, 2025
@aurelienpierre aurelienpierre added this to the 0.1 milestone Jan 10, 2025
aurelienpierre added a commit that referenced this issue Jan 11, 2025
…masks from DB

It's confusing because (parametric) blending operations are fetched in the same SQL query as module params, but masks are not. They have their own method and their own SQL query that needs to update the existing history stack.

So we just need to commit blendops (including raster masks) after that step of history fetching.

Fix Bug: Raster masks broken #396
@aurelienpierre
Copy link
Collaborator

Fixed by cf52835

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Projects
Development

No branches or pull requests

2 participants