See D884 for fixes.

Can reproduce. I agree with your explanation. When group 1 is empty, it can't actually be selected. There's the dotted line, but group 2 is still in focus (which is stated in the popup dialog). I guess after group 2 is deleted the "focus" is not set correctly.
I guess we could either ensure that the next available label is automatically selected, since usually there is always a label selected, or the "Delete group" should be respective to the dotted-line selection of a group.
I'm not sure if this contradicts previous discussions we had about the topic, but I would actually prefer the second option. Generally, it feels weird that one is able to kind of select a group and interact with its context menu, but the buttons above still refer to the group of the focused label. For example, it is impossible to add a label to an empty group with the buttons (except for the first one).

Deleted branch from rMITK MITK: release/T30248-2023-Week-46.

Pushed new branch to rMITK MITK: release/T30248-2023-Week-46.

New checkbox "Allow all models" added to Preferences. Changes included in D875

I would lean to adding a preference "allow unsupported models" (default ==false). If people activate it, they can see all models (that are not blacklisted) and not just the white listed (verified) once.
If the risk is communicated and users are willing to take it, then I would not actively prevent people from using models (as long as they don't crash MITK or so).

Deleted branch from rMITK MITK: bugfix/T30230-FixCppMicroServicesTestDriver.

Pushed new branch to rMITK MITK: bugfix/T30230-FixCppMicroServicesTestDriver.

At the moment it is not very elegant to generate plugin-dependencies just for the documentation. And this solution also does't allow for having references in two-ways between plugins (refer plugin B in plugin A and vice versa). A different approach could be an additional entry in the manifest_header.cmake just for documentation dependecies or to take all references into account for all plugins.

This task can be potentially closed after T30149 is handled (that one should be done first).

Discussion result: Should be fixed with an exact calculation instead of an approximation based on histogram.

Deleted branch from rMITK MITK: bugfix/T30214-MergeAFewGitHubPullRequests.

• Merged GitHub pull request https://github.com/MITK/MITK/pull/273 (Workbench crash on simple tree model update)
• Merged GitHub pull request https://github.com/MITK/MITK/pull/272 (Do not include uic generated file in header).

Merged GitHub pull request https://github.com/MITK/MITK/pull/265 (Fix typos in Plugins subdirectory).

Merged T22990: SlicedGeometry3D crash when volume is rotated by 45 degrees (GitHub pull request: https://github.com/MITK/MITK/pull/203).

Pushed new branch to rMITK MITK: bugfix/T30214-MergeAFewGitHubPullRequests.

Discussion result: Use permissive mode always switched on. No preferences needed.

Discussion result: Feature is useful but not as plugin on its own but instead as part of the statistics plugin (as an additional button like copy to clipboard).

In T30186#253599, @kislinsk wrote:

Is there some additional functionality involved or could we just add it as another MIME type for images instead of a dedicated plugin?

Yes there is. As only (potentially) the pixel values of image parts with an mask should be exported. And you can select multiple image (as long as they have the same geometry), there pixel values will be exported into the same table.

@full Is this still an issue (with the current release)?

As far as I understand the plugin's purpose it is basically an export of an image as CSV file? And the request is to add it to MITK instead of eMITK? Is there some additional functionality involved or could we just add it as another MIME type for images instead of a dedicated plugin?

Deleted branch from rMITK MITK: feature/T30177-ContourModelSetToSegmentationApp.

Deleted branch from rMITK MITK: feature/T30176-FindTask.

After we have moved to Qt6, we then have QtCharts to our disposal and migrate to that.

After we have moved to Qt6, we then have QtCharts to our disposal and migrate to that.

After we have moved to Qt6, we then have QtCharts to our disposal and migrate to that.

Should be done after(!) we have moved to Qt6 as we then have QtCharts to our disposal and migrate to that.

Should be done after(!) we have moved to Qt6 as we then have QtCharts to our disposal and migrate to that.

We want to have:

• a slick minimum Workbench version
• a fuller / bigger version. Exact contents and name still need to be compiled (e.g. Perfusion, Fitting, ...)

The app is now called MitkContoursToImage. By default it outputs classic binary images for each contour set in the input. Use the --format/-f flag (defaults to binary) to change the output format to label or multilabel. The former outputs single-label segmentations for each contour set in the input and the latter outputs a single multi-label segmentation with labels for each contour set. Note that each label will be in its own layer/group to allow for overlapping labels.

Pushed new branch to rMITK MITK: feature/T30177-ContourModelSetToSegmentationApp.

Pushed new branch to rMITK MITK: feature/T30176-FindTask.

I just did the Pic3D/brain test mentioned above and it works with v2023.04.

We could set up a test case to check if it is still an issue in the context of segmentation task lists. A segmentation task list with two tasks, one task referencing Pic3D.nrrd, the other task referencing the tilted brain.nrrd image. If this is working with the latest release it should be a proof that is not an issue anymore.

Thanks for the report. We recently removed this functionality from MITK so I would like to close this task as Wontfix if you are fine with the workaround you mentioned?