[756368]: Merge branch 'bug-18329-CrashInLiveWireTool'
Merged commits:
2014-10-22 17:50:19 Thomas Kilgus [803f64]
Removed wrong return.
[756368]: Merge branch 'bug-18329-CrashInLiveWireTool'
2014-10-22 17:50:19 Thomas Kilgus [803f64]
Removed wrong return.
[6be06a]: Merge branch 'bug-18307-DisableUnsupportedRecordings'
2014-10-21 14:46:20 Thomas Kilgus [0e9fd0]
Images which are not supported by a ToFDevice can no longer be recorded with the GUI.
[6143a9]: Merge branch 'bug-18303-RewriteOclTests'
2014-11-27 15:32:12 Thomas Kilgus [6eb0f3]
COMP: Added exception to the oclImage in case the format is not supported.
[415698]: Merge branch 'bug-18303-DeactivateTestsAgain'
2014-11-26 18:06:16 Thomas Kilgus [799e2e]
COMP: Reverting to old tests.
[6f05c5]: Merge branch 'bug-18303-RewriteOclTests'
2014-11-27 11:34:03 Thomas Kilgus [8e3c60]
COMP: Fixed SetInput call in BinaryThresholdFilter. Skipping test on hardware which does not support it.
I disabled the tests temporarily until we introduce a check if 3D images are supported on the respective machine. We also should double check if we can use the CPU option for this Mac.
Waiting till next release.
[d8716a]: Merge branch 'bug-18303-Ocl2DTestForMacOs'
2014-10-17 17:11:00 Thomas Kilgus [d35e7a]
COMP: Adapted ocl test to use a 2D image temporarily, because our mac continuous only supports 2D.
[70166f]: Merge branch 'bug-18303-DeactivateTestsAgain'
2014-11-26 17:50:01 Thomas Kilgus [33bd2d]
COMP: Added 2D kernel.
[e58ac3]: Merge branch 'bug-18303-DeactivateTestsAgain'
2014-11-26 17:03:13 Thomas Kilgus [9c79bc]
COMP: More debug output.
[fe0d33]: Merge branch 'bug-18303-Ocl2DTestForMacOs'
2014-10-17 17:28:17 Thomas Kilgus [cc47b2]
COMP: Tests disabled temporarily. See T18303.
[6f30ab]: Merge branch 'bug-18303-FixForUnsupportedDimension'
2014-11-26 16:28:13 Thomas Kilgus [f93eff]
Reactivated the OpenCL tests.
This will be fixed within T18735.
I found out that we save persistence data in:
[48ff76]: Merge branch 'bug-18268-AvoidReinitInUS'
2014-10-21 16:45:08 Thomas Kilgus [3889da]
Geometry3D is no longer suited for sliced images. Replacing with SlicedGeometry3D.
[c89d53]: Merge branch 'bug-18253-FixingOpenCLTest'
2014-10-14 16:06:09 Thomas Kilgus [bc470f]
New US_INITIALIZE_MODULE macro.
I found a VTK example:
http://www.vtk.org/Wiki/VTK/Examples/Cxx/IO/ReadPLY
Test data can be found in:
Loading of Ply and Obj files works.
Loading of textures and texture coordinates doesn't work. As far as I understand it, the texture coordinates are loaded correctly, but the rendering is somehow screwed.
VTK example to load a textured surface
Passing this to Capsar hoping that he might actually integrate it during bugsquashing some day in the future ;). Feel free to just close this.
VTK example to load a textured surface
This doesn't sound like a rendering bug to me. The warnings are clearly related to geometries ;).
In my opinion we have much more pressing issues than randomly failing rendering tests. Rendering tests are important to show us if a rendering feature is globally broken after a change. It is not worth the effort to make all these tests run all the time. If they randomly make the continuous clients red, we can decide to disable them.
What is the status here?
Passing this to Eric. He can decide to work on it during NaviBSQ or just close this.
What is the status here? Can we close it?
This should be fixed with with T18405.
[e470b6]: Merge branch 'bug-17878-DisablingShaderMsg'
2014-11-05 16:47:11 Thomas Kilgus [3ecf6a]
Removed disabling shader message.
May I change this for the release?
Fixed with the geometry project.
Postponed.
(In reply to Matt Clarkson from comment #4)
(In reply to Thomas Kilgus from comment #3)
One more try:
Matt, do you have a full patch or pull request by chance? :)No, i never made one. Im about to take a train. I may be able to do one
tonight. If you havent received one by tomorrow morning, then assume I have
forgotten or given up for whatever reason.Also, when is 2014.10 release being tagged?
M
One more try:
Matt, do you have a full patch or pull request by chance? :)
This sounds very reasonable to me. Matt, do you have a full patch or pull request by chance?
What is the status here?
It turned out that the PMD/lib path was missing in mitkFunctionGetLibrarySearchPaths. Sascha and I also figured out that the OpenCV path OpenCV-build\lib is different between linux and windows (and fixed that also).
[386148]: Merge branch 'bug-17740-PathVariablesForInstaller'
2014-05-28 14:37:37 Thomas Kilgus [8e2c1c]
Added correct PMD lib and OpenCV lib to search library path.
Illustration of the bug
This bug is very important, as many nightlies are affected.
[7fe635]: Merge branch 'bug-17547-LookupTableAndColorMaps'
2014-04-09 15:10:44 Thomas Kilgus [6bf5b4]
Updated documentation.
I removed the LEVLEWINDOW_COLOR enum value of the rendering mode property and adapted tests accordingly.
I wrote a core change request.
Seems to be generic. Most likely related to the changes for the color-maps.
Reverted enum values. This fixes the test and the scene file compatibility.
I will have a look.
Passing to Ingmar, as he was working on this.
Hi Ingmar,
Hi Ingmar,
Hi Ingmar,
What is the status here?
[397534]: Merge branch 'bug-17410-ReactivateCVGrabberTest'
2014-06-18 16:52:06 Thomas Kilgus [110753]
Changing property on connection of device.
Clearing property list works if the data nodes are removed upon disconnect. Now the images are not correctly initialized for the MITK player device. Only the distance image works correctly. Kinect and CamCube should be working now (tested only Kinect).
I fixed the playing of data while recording.
I could fix the dimension issues. Now the distance image is set correctly. I assume that the properties of the player are incorrect.
[ed4e28]: Merge branch 'bug-17384-NamingConventionKinect'
2014-03-03 17:19:57 Thomas Kilgus [608f29]
COMP: Fixed export names of Kinect 1 and 2 modules.
[691e48]: Merge branch 'bug-17384-OpenCLNamingConvention'
2014-03-07 18:04:16 Thomas Kilgus [2f2580]
COMP: Fixed OpenCL exports according to new naming convention.
[43273a]: Merge branch 'bug-17349-ImageDenoisingModule'
2014-02-26 15:42:13 Thomas Kilgus [a54910]
Warnigns as errors and dependency.
I am sorry, but this description is not enough for me to reproduce the potential leak. Can you give us information how to reproduce this in the workbench? Is this still relevant for you?
Bug is fixed. Waiting for green dashboard to merge ...
[fdefe7]: Merge branch 'bug-17318-OpenCLIncludeWin'
2014-02-26 14:14:52 Thomas Kilgus [5af97b]
Fixed bug where windows would not receive additional include directory.
This was fixed within another bug.
[5d35e1]: Merge branch 'bug-17124-IntegrationTexutreMapping'
2014-02-14 14:49:42 Thomas Kilgus [40cb76]
Adapted ToFUtil. Should work for all cameras.
[6b955c]: Merge branch 'bug-17173-NoNULLinMitkEqual'
2014-02-19 17:04:53 Thomas Kilgus [2b174e]
COMP: Fixed macros in Diffusion.
[96a7d4]: Merge branch 'bug-17173-NoNULLinMitkEqual'
2014-02-19 16:20:43 Thomas Kilgus [66ea78]
Changed interface to const reference. Older method is deprecated. Fixed some warnings.
[6eb926]: Merge branch 'bug-17124-GuiChangesForKinect2'
2014-02-24 10:05:22 Thomas Kilgus [67a0f0]
Finished work at surface generation widget. Fixed colortransferfunction of the vis. widget.
[ccfb09]: Merge branch 'bug-17124-IntFinal'
2014-04-02 18:53:53 Thomas Kilgus [dad3e2]
Merge branch 'bug-17124-IntegrationTexutreMapping' into bug-17124-IntFinal
Can be merged after release (now).
[ee5125]: Merge branch 'bug-16805-FixToFDevicesAndOpenCL'
2014-01-15 13:19:47 Thomas Kilgus [8323a7]
Added package depends config for OpenCL and target_link_libraries for additional libs.
Yes, that's the main reason that this is not merged yet.
Stefan will do the final review for this bug and it will be merged after the upcoming release.
I asked for different reviews, but nobody feels responsible for this old code ;). If anyone gives the release flag, we can merge the changes. If not, we have to wait until the release is over.
Great! I will have a look.
Ok, so the parent node is actually required? That sounds not so easy to fix to me then...
I have no clue whatsoever why there is a parent node in that mapper. Could be historically related to segmentations, which could need a parent. Should be very easy to fix.
[84aff2]: Merge branch 'bug-16731-ImageGrabberTests'
2013-12-27 12:08:08 Thomas Kilgus [94640e]
COMP: Converted ToFImageGrabberTest to new test suite system. Fixed some minor issues found by the test.