Jun 2 2020
Checked it the source of the problem is QmitkAutocropAction of the multisegmentation plugin. It generates an mitk::Image instead of a mitk::LabelSetImage as a result. As a consequence a precondition of the QmitkMultiLabelSegmentationView fails with an assertion.
Jun 1 2020
@sharonYu Thanks for the update. It seems to realy be an so far unkown problem. Would it be possible to provide an anonymized data set to debug the situation and see what is happening in code?
We scan the raw DCE-MR image directory using the MITK DICOM Browser and got the correct Dicom information in the list. However, the error message showed when pressing the View bottom.
May 31 2020
Poor man's backup of the settings: 😄
Note that it's only on Linux as it is the only client where I activated tests. I will disable them for now again and we can decide on Wednesday how we want to proceed.
May 30 2020
For mitk::UIDGenerator I have decided to base it internally on boost::uuid (name_generator).
Damn. As long as the test fail we cannot activate the tests, as reviews will never leave draft state.
@kompan : This should be clarified now and is more a documentation issue? Isn't it.
But I think it would not hurt to add a test for that feature into the perfusion checklist. Thanks.
May 29 2020
Seems to work, but mitkPluginGeneratorBuildTest-Release fails.
http://e071-jenkins:8080/job/MITK/job/Differential/job/Linux/220/consoleFull
Activated tests on Linux Differential clients. Buildable is currently running. If it works as expected, I will also activate the tests for macOS and Windows.
{B605}
It is still the problem explained in https://phabricator.mitk.org/T24766#131411.
As it is not @maleike's patch, it must have been the upgrade from VTK 8.0 to 8.1 or the depth peeling reactivation in general. Checking with the latest version of VTK...
Oof, was able to boil it down to a VTK bug when depth peeling is enabled. Evaluating different options, like if it has something to do with @maleike's VTK patch, or if it is already fixed in a newer version...
Only segmentations are affected. There is no relevant change in the mapper code between the last release and now, though. I found out, that it has something to do with opacity. In particular when setting the opacity of the top segmentation to 1, it works, but masks the segmentation below with its own extent. When setting both opacities to 1, everything looks like it should.
Works in v2018.04.2. Parent image has layer 4, first segmentation 5, second 6, just as one would expect it to be. Drag & dropping the lower segmentation above the upper one, interchanges the layer.
During (and only during) drawing on any segmentation (after first mouse move), both segmentations are also visible.
Pushed new branch bugfix/T27436-HideInterpWidgetsInMultilabelSegView.
May 28 2020
It seems that all the test data (dce_2CXM_ValidationDataSet) load successfully.
From my side, the migration is completed now. As discussed in the meeting on Monday, the name of a task in a single-task data set is optional. I created test cases for the plot functions that return plot objects, separate issues are created for those who don't. The extraction of top performing algorithms and subsets of tasks from rankings is also migrated and tested. I calculated the package test coverage to identify code that is never executed and deleted what seemed obsolete to me. There are still files with unused code where I'm not sure whether they are needed in other scenarios that are not well documented:
- select.R
- benchmarkUtils.R
- winner.R
- S3.R
- second.R
- extract.workflow.R
- compareRanks.R
It seems that all the test data (dce_2CXM_ValidationDataSet) load successfully.
Closed now but might be related to T27085.
Some things may have changed. So If I have time during the next 4 weeks I would like to give it a second look, finish the clean-up and merge / close this branch.
When using Qt 5.13 on Ubuntu 20.04 I had no issues with our third party libraries, only a lot of deprecated warnings as errors in MITK and other minor issues.
May 27 2020
Sigh, CTK with Qt 5.12.6 fails with the same linker error (definitely worked in the past). I guess it is VS 2019 16.6.0 then.
Our current version of VTK does not work with Qt 5.15. I'm just going back to 5.12.6 for now, as I didn't plan to pump so much time into this issue.
Qt 5.15 is not working with qwt 6.1, though.