The hash above doesn't seem to be auto-linked. So here it is: rMITKa73e482c0387: moved labelset-related classes out from core
- Queries
- All Stories
- Search
- Advanced Search
- Transactions
- Transaction Logs
All Stories
Dec 13 2022
Dec 12 2022
A big leap into the Python direction would be to finally resolve T27923: Integrate possibility to execute Python Code in MITK , which is complete for a while now but we weren't able to figure out why it fails on Ubuntu build clients. Since the ITK v5 upgrade, we also need to adjust a little more I guess.
Task lists don't do anything special regarding saving/loading data. So in theory, this bug should be reproducible by regular loading and saving via the menu in the Workbench?
However, task lists are prone to expose existing bugs related to repeated saving/loading of data. We indeed noticed another bug in this direction with .nrrd files, which were erroneous when loaded a second time in certain situations. The reason was faulty handling of properties: T29274: Label Loss on repeatedly loaded and saved segmentations.
Dec 9 2022
Dec 8 2022
Dec 7 2022
Reading your other task regarding mitkscene files I wonder if it has something to to with attaching the level window property to the BaseData property list instead of the DataNode property list. In principle, the level window should be attached to the latter one as it is an "appearance" property and not something instrinsic to the data. A few years ago we changed the way properties are queried and when a property is not present in a DataNode property list, the Base Data property list is queried as fallback. Maybe it is not fully functional for level windows as it could be that the code somehow relies on / assumes that the properties are located in the DataNode property list.
Not sure if I fully understand what is going on here but the difference is that in your first example you have DataNode properties and in your second example you have BaseData properties, right? Let's talk about it in our group 5 meeting today. :)
Dec 6 2022
Dec 1 2022
@a178n I added cpp-httplib as dependency to MITK in the branch mentioned above so you can easily experiment with it. For details, see rMITKf88e3d05aeb0: Add cpp-httplib as optional dependency.
Pushed new branch to rMITK MITK: feature/T29421-httplib.
Nov 30 2022
After a meeting with Ralf and Stefan (30.11.2022), the header-only library cpp-httplib (GitHub: https://github.com/yhirose/cpp-httplib), was chosen as a promising alternative. The library seems the most popular in the GitHub community compared to Pistache, the second favourite library from the list.
Hands-on efforts to build server and client calls will be undertaken to test further suitability.
The literature survey searching for alternatives is summarized here: https://hub.dkfz.de/s/qsEgaWybwiWcfYN
Should be tackled as soon as T29206: [SEG] Implement a Qt model for LabelSets for MVC patterns is done.
Adding a data node reset-geometry-action was a first step and enough to tackle this issue. If there are more requirements for some "intelligence", new tasks need to be opened.
Render window specific visibility has to be covered, as mentioned in D756#26848. This will be done in T29403: [mxn multi widget] Concept for crosshair data type.
This task is invalid; related work will be continued in T29297: [mxn multi widget] Optimize data accessibility inside render windows.
This issue should also be regarded if T26959: [Segmentation] Newly created LabelSet segmentation is not displayed as "outline binary" is done.