Two options in order to integrate Epiphan were discussed:
- Queries
- All Stories
- Search
- Advanced Search
- Transactions
- Transaction Logs
All Stories
Oct 23 2020
Deleted branch feature/T27789-ClippingPlanePlugin.
Pushed new branch feature/T27789-ClippingPlanePlugin.
Oct 22 2020
In T26382 the recent developments are summarized.
Oct 21 2020
Flags to check whether the "Next" button has being activated have been added. Needs to be adapted due to a bug:
Oct 20 2020
I got this error message on SS FSE CEST, ERROR: Image(0000027E06632400): A spacing of 0 is not allowed: Spacing is [0.31, 0.31, 0]
Ups ;) Replace with - or _?
Also, in the zip folder for example, we suddenly have nested folders: MITK-snapshots/2020-09-25-123-gabcdefab-windows-x86_64.
Deleted branch bugfix/T27900-TypoInReference.
Pushed new branch bugfix/T27900-TypoInReference.
Oct 19 2020
This notification can be omitted, since there is already a "generating report" notification on the background.
Agree, but we should definitely package Python instead of relying on system python in any case as there are more strings attached. Even between minor Python updates there are imcompatibilities especially when SWIG is involved. Also we rely on certain Python packages like numpy or SimpleITK. So the plan is:
I was just about to add a note that there is the general question how to handle python in installers. I guess for just having the basic runtime available it would be ok to just load libpython3.so or libpython3.8.so and let the system take the right one from the right path. But on Windows I guess most people don't install Python in system locations. Anyway I think this could be fixed even with the current packaging if we need it for 2020, but I'm just reporting, no pressure from my side ;)
I guess this is not specific to Fedora (not even to Linux) as we do not yet package python into the installer as far as I know. As this task is a bit moe vorbose than T26258: MITK and python only works with the specific python version the installer was created with, I will close the other one as duplicate.
Oct 16 2020
- I renamed the context menu entries to Load/Save LabelSet Preset
- I made these actions also available as tool buttons in the Multilabel Segmentation View
- I fixed quite a few bugs in the loading mechanism
- Existing label properties are now overridden by a preset
- Layers/labels in other layers work now
- The state of the Multilabel Segmentation View is now updated after a preset was loaded
Pushed new branch bugfix/T27851-FixLabelSetPresetsAndMakeThemMoreAccessible.
Oct 15 2020
Deleted branch hotfix/T27887-fix-packaging-with-system-qt.
Pushed new branch hotfix/T27887-fix-packaging-with-system-qt.
stabilityByAlgorithmStacked is now replaced by stabilityByAlgorithm(....., stacked=TRUE)
The functionality is already in a separate class LabelSetIOHelper so it is not restricted to DataManager actions per se. The code in these actions is of pretty bad quality, though. The first step will be to clean up these actions. I will also rename the actions in the context menu to Load LabelSet Preset and Save LabelSet Preset to make them more distinguishable. Then I will look into the view to see if the functionality is already there somewhere or if we must add it. I also already noticed that while the preset files contain information about layers as well, only the labels of the first layer are loaded.
I'm all in for that. We already discussed that we will focus again on this task {T26754} and its subtasks. This also includes
Automatic GUI testing
and
move as much tests as possible to Unit-Tests
Thanks for reporting. The whole method seems to be dubious. The lines of code you mentioned are plain wrong, the loop should start with k=1 and the data type of numPixelsInDataset should be more appropriate.
Oct 14 2020
OK. But I tend/think more and more that we should strive for a unit test testbed that tests all that on the tool class level (either the respective base classes or the tools itself where it is easily possible (e.g. because the have no interaction (but even thous shouldn't be a show stopper).
The checklist don't say anything about multilabel segmentation. This should be addressed with a special focus on creating and combining multiple labels / layers.
I couldn't find any other checklist with a focus on the multilabel functionality.
I updated the document and added content to explicitly tests
I can verify this behavior. Also, if you close the segmentation view while the position nodes are shown, they stay visible.
If you open the segmentation view again and select the 3-Dimensional interpolation, the position nods are still shown, although the checkbox is not checked.
Checking it and unchecking it will hide the position nodes again.
Alright -> removed the task from release-workboard
In T27706#212501, @schererj wrote:@floca do you know something about the status of the "new" radiomics container?
No.
@floca do you know something about the status of the "new" radiomics container?
Since we don't have the new container yet, we will keep the old radiomics pipepline for this release if that's ok.
Deleted branch feature/T27879-MITK_BUILD_NAME_SUFFIX.
Pushed new branch feature/T27879-MITK_BUILD_NAME_SUFFIX.
No, not yet. I've sent an email and asked for a more detailed description of the problems and will post it in this task when I hear back from Jerome.