- User Since
- Aug 1 2016, 12:10 PM (383 w, 6 d)
Aug 28 2017
Aug 25 2017
Integrating this one would be great there is a lot of enhancement in OpenCV 3.1.0. Our MITK-based application use it to have the Eigen Face module available in the contribs.
We fixed using a new "-D" parameter for superbuild, and #ifdef in the code to allow both 2.4.x and 3.x series of OpenCV
Aug 10 2017
May 4 2017
Mar 7 2017
Mar 6 2017
Here is a simple patch wich fixes the command format. The PYTHONPATH has to be set before CMake is called.
basically, it has to be set to [compilation folder]ep/Lib/site-packages/
Mar 3 2017
We fixed that in the previous version patching CMakeExternals/SimpleITK.cmake :
Mar 2 2017
Nov 18 2016
Sep 20 2016
Aug 9 2016
Thanks i better see. The implementation is good and follows the Eclipse 3 Job API. The job listeners should work, and the "job view" can be plugged using a 'ProgressProvider' concept. I think we will work on that part soon.
We also have the plugin tests quite working (the tested plugin is still not loaded, but whole code now compiles). It should be done within next month -i hope- because we really need testing :)
Existing test code for job api would be then plugged back.
no problem. Thanks for the link i'm not yet aware of a lot of Phabricator features
There is still problems with scheduling, but I have to go on other tasks for the moment. The API is useable even there are still problems with scheduling. I should come back on this ticket once we get the plugin test infrastructure back to work.
After some more learning in Berry and smart pointer, I'm now quite done. No more freeze and clean Job subclassing.
Aug 8 2016
In the other hand, if i create a job using a berry::SmartPointer, I no longer have the early deletion problem thanks to internal ref counter. In that case, I can launch the job only one time : the second time the job is fired, the first one has not been destroyed and i face an UI freeze (and have to kill the application)
In some functions, 'this' (as instance of InternalJob) is wrapped into berry::SmartPointer.
When the function ends, then the pointer counter falls to 0 and the object is destroyed.
From my point of view we can use pointer everywhere, and the Job pointer can be deleted when finished,
if no rescheduled, when all listeners has been called.
Aug 5 2016
There are some other errors in the API with the smart pointers. I often have my Job deleted before execution.
I'm investigating on it.
The fix is very simple but without it, the Job API is quite unusable.
Aug 2 2016
Opened this bug because the splash screen wasn't visible.
After deeper analysis, we discover that the splash screen may appear and disappear quickly, make it "unvisible".
It is configured to be closed just after BlueBerry platform is up (plugins started), and some DLL may be loaded afterward. Hence there should be a delay between splash close and actual workbench display
CTK contains code for splashscreen management. I will go ahead this way
Work ongoing on :
Work is done and available in pull request : https://github.com/MITK/MITK/pull/131
The splashscreen is closed by CTK, and thus the code on their side has been updated too. I can provide a updated patch for CTK-in-superbuild if needed
Thanks for information. I'll stop creating bug for this topic, and wait for next release to try the upgraded build system. We were also waiting for Qt 5.6, but Digia is a bit late and we had to go ahead.
Patch which add /wd4180 for MSVC build in CppMicroServices
Patch for the VTK 6.2.0 patch, allowing MSVC 14 build
Extremely small patch that fixes the bug
I am a bit novice in medical image format, i'm working with other industry devices which produce 2D images (TEM, AFM, SEM).
So, I need the 2D tiff support and will do my best to provide a fix here.
I also have some tiff files which are not loading and even throw a segmentation fault in file vtkMitkLevelWindowFilter.cpp at line 209. We are working on this since we are using a lot of tiff images. I created a simple image with Gimp which triggers the problem.
Forget my last comment, i just found http://bugs.mitk.org/show_bug.cgi?id=12811
A TIFF file which trigger a segmentation fault in WorkBench