thank you! now it works! On 17/12/2013 06:25 a.m., Sascha Zelzer wrote: > Hi, > > further investigations revealed that using the "lib" prefix in the > US_INITIALIZE_MODULE macro and the usFunctionEmbedResources CMake > function solves the problem (works both on Windows and Linux). > > We should have followed the documentation of the macro and CMake > function closely, which states that you have to use the library name. > On Linux, omitting the "lib" prefix still works (MITK modules do not > have the "lib" prefix on Windows, but they do on Linux whereas MITK > plug-ins always have a "lib" prefix) but on Windows the library is not > found if you are using for example "org_mitk_gui_qt_imagenavigator" > instead of "liborg_mitk_gui_qt_imagenavigator". > > I attached again a test-patch for the imagenavigator which did the > trick for me on both Linux and Windows. > > Best, > Sascha > > On 12/13/2013 11:07 AM, Miguel Nunes wrote: >> Hi Sascha, >> >> Those are good news! I hope you can find the bug really quick, and >> provide us with a solution :) >> >> Best regards and have a nice weekend! >> Miguel >> On Friday, December 13, 2013 11:01 CET, Sascha Zelzer >> <[email protected]> wrote: >>> Hi, >>> >>> just a short update. While it did work for me on Linux, it didn't work >>> on Windows. It is related to some library naming issues (using the >>> "lib" >>> prefix or not). I am still investigating it and will let you know >>> when a >>> fix is available. >>> >>> Best, >>> Sascha >>> >>> On 12/09/2013 12:14 PM, Miguel Nunes wrote: >>>> Hi Sascha, >>>> >>>> when I apply the patch there are no xml files... I guess I have to >>>> add them manually. >>>> >>>> I am not connected to the SVN. I just got the zip with the source. >>>> >>>> So, I rebuild MITK with the new patch with success... but it still >>>> doesn't load the xml files. >>>> >>>> 0.50 blueberry.ui.wrkbncPlg: LOG: Unable to create view ID >>>> org.mitk.views.imagen >>>> avigator: Part initialization error: >>>> D:\src\MITK2013.09.0\Core\Code\Interactions >>>> \mitkStateMachineContainer.cpp:73: >>>> Resource not valid. State machine pattern not found:MyInteractor.xml >>>> 0.64 blueberry.ui.wrkbnc: BlueBerry Workbench ready >>>> >>>> >>>> As you can see I am using 2013.09. >>>> >>>> Best, >>>> Miguel >>>> On Monday, December 9, 2013 11:40 CET, Sascha Zelzer >>>> <[email protected]> wrote: >>>>> Hi, >>>>> >>>>> could you try applying the patch as it is and check if the patched >>>>> imagenavigator plug-in "works"? >>>>> >>>>> Also, what MITK version are you using again? >>>>> >>>>> - Sascha >>>>> >>>>> On 12/08/2013 06:15 PM, Miguel Nunes wrote: >>>>>> Hello Sascha, >>>>>> >>>>>> I have created a new plugin and tried to do the basics of adding >>>>>> the xml resources, but without success. >>>>>> >>>>>> Maybe it only works in plugins directly connected to MITK. My >>>>>> plugin is in new MITK project that uses my base MITK build as an >>>>>> external MITK. >>>>>> >>>>>> Is it possible that MITK gets lots somehow by having this structure? >>>>>> Best, >>>>>> Miguel >>>>>> On Friday, December 6, 2013 16:15 CET, Sascha Zelzer >>>>>> <[email protected]> wrote: >>>>>>> On 12/05/2013 05:17 PM, Miguel Nunes wrote: >>>>>>>> Thanks for going through all that work Sascha. >>>>>>>> >>>>>>>> My resources vector has size zero, but a name is printed which >>>>>>>> is the same I pass on the line: US_INITIALIZE_MODULE("My >>>>>>>> Module", "org_proj_plugin") >>>>>>> Maybe something is not right in your CMake files then. Looks >>>>>>> like the >>>>>>> generated resource .cpp file was not compiled into the plug-in's >>>>>>> shared >>>>>>> library. >>>>>>> >>>>>>>> I wonder if you could send us your own plugin, sascha, in zip >>>>>>>> file, so we can compare, build and run locally. >>>>>>> I attached a patch file where I just added some code to the image >>>>>>> navigator plug-in for testing purposes. >>>>>>> >>>>>>> Best, >>>>>>> Sascha >>>>>>> >>>>>>>> Kind regards, >>>>>>>> Miguel >>>>>>>> On Thursday, December 5, 2013 16:23 CET, Sascha >>>>>>>> Zelzer <[email protected]> wrote: >>>>>>>>> Guys, I just tried it myself and added some interaction xml >>>>>>>>> files to a >>>>>>>>> plug-in and instantiate a mitk::DataInteractor and loaded the >>>>>>>>> plug-ins >>>>>>>>> xml files. Everything worked okay so far. >>>>>>>>> >>>>>>>>> @Sebastian I noticed the your config xml file was written >>>>>>>>> differently in >>>>>>>>> the files.cmake file and your C++ code. >>>>>>>>> >>>>>>>>> >>>>>>>>> You can list all embedded resources of your plug-in by doing >>>>>>>>> something >>>>>>>>> like this: >>>>>>>>> >>>>>>>>> us::Module* module = us::GetModuleContext()->GetModule(); >>>>>>>>> std::vector<us::ModuleResource> resources = >>>>>>>>> module->FindResources("/", "*", true); >>>>>>>>> for (std::vector<us::ModuleResource>::iterator iter = >>>>>>>>> resources.begin(); iter != resources.end(); ++iter) >>>>>>>>> { >>>>>>>>> MITK_INFO << "[" << module->GetName() << "] resource: " >>>>>>>>> << iter->GetResourcePath(); >>>>>>>>> } >>>>>>>>> >>>>>>>>> >>>>>>>>> - Sascha >>>>>>>>> >>>>>>>>> On 12/05/2013 03:21 PM, Miguel Nunes wrote: >>>>>>>>>> yes, I confirm. >>>>>>>>>> >>>>>>>>>> Also, on the plugin's ..._resources.cpp fie I can see the xml >>>>>>>>>> were found and generated: >>>>>>>>>> >>>>>>>>>> static const unsigned char us_resource_data[] = { >>>>>>>>>> // D:\src\* PATH * >>>>>>>>>> \resources\Interactions\StatemachineVrvis.xml >>>>>>>>>> 0x0,0x0,0x0,0x90, >>>>>>>>>> 0x0,0x0,0x0,0xdd,0x55,0x8f,0x4d,0xa,0x2,0x31,0xc,0x85,0xf7,0x9e,0x22,0xe4, >>>>>>>>>> 0x2,0x73,0x81,..., >>>>>>>>>> // D:\src\ >>>>>>>>>> *PATH*\resources\Interactions\StatemachineVrvisConfig.xml >>>>>>>>>> 0x0,0x0,0x0,0xbf, >>>>>>>>>> 0x0,0x0,0x1,0x9c,0x95,0x91,0x41,0xe,0x82,0x40,0xc,0x45,0xf7,0x9e,0xa2,0xb2, >>>>>>>>>> 0x27,0x5c,... >>>>>>>>>> }; >>>>>>>>>> On Thursday, December 5, 2013 15:12 CET, >>>>>>>>>> Sebastian Ordas <[email protected]> wrote: >>>>>>>>>>> yes, just like this: >>>>>>>>>>> >>>>>>>>>>> void QmitkMultiLabelSegmentationView::InitializeListeners() >>>>>>>>>>> { >>>>>>>>>>> if (m_Interactor.IsNull()) >>>>>>>>>>> { >>>>>>>>>>> us::ModuleContext* moduleContext = >>>>>>>>>>> us::GetModuleContext(); >>>>>>>>>>> m_Interactor = mitk::SegmentationInteractor::New(); >>>>>>>>>>> m_Interactor->LoadStateMachine( >>>>>>>>>>> "SegmentationInteraction.xml", >>>>>>>>>>> moduleContext->GetModule()); >>>>>>>>>>> m_Interactor->SetEventConfig ( >>>>>>>>>>> "SegmentationConfig.xml", >>>>>>>>>>> moduleContext->GetModule()); >>>>>>>>>>> us::GetModuleContext()->RegisterService<mitk::InteractionEventObserver>( >>>>>>>>>>> >>>>>>>>>>> >>>>>>>>>>> m_Interactor.GetPointer(), us::ServiceProperties() ); >>>>>>>>>>> } >>>>>>>>>>> } >>>>>>>>>>> >>>>>>>>>>> On 05/12/2013 11:10 a.m., Sascha Zelzer wrote: >>>>>>>>>>>> Just to make sure: Did you call LoadStateMachine("...xml", >>>>>>>>>>>> GetModuleContext()->GetModule()) using the second argument >>>>>>>>>>>> to provide >>>>>>>>>>>> the module instance which contains the embedded xml file? >>>>>>>>>>>> >>>>>>>>>>>> On 12/05/2013 02:38 PM, Miguel Nunes wrote: >>>>>>>>>>>>> Sebastian, >>>>>>>>>>>>> >>>>>>>>>>>>> I just tested my plugin and I continue without success. >>>>>>>>>>>>> The same >>>>>>>>>>>>> Error remains in the MITK plugin tab says: >>>>>>>>>>>>> " Part initialization error: >>>>>>>>>>>>> D:\src\MITK2013.09.0\Core\Code\Interactions\mitkStateMachineContainer.cpp:73: >>>>>>>>>>>>> >>>>>>>>>>>>> >>>>>>>>>>>>> Resource not valid. State machine pattern not >>>>>>>>>>>>> found:StatemachineVrvis.xml " >>>>>>>>>>>>> Do you have any new stuff on this issue? >>>>>>>>>>>>> >>>>>>>>>>>>> On Thursday, December 5, 2013 13:37 CET, "Miguel Nunes" >>>>>>>>>>>>> <[email protected]> wrote: >>>>>>>>>>>>>> Hi Sebastian, >>>>>>>>>>>>>> >>>>>>>>>>>>>> Sorry for taking so long to reply but out of nowhere Poco >>>>>>>>>>>>>> stopped >>>>>>>>>>>>>> working and I am re-buidling the whole MITK framework >>>>>>>>>>>>>> from scratch, >>>>>>>>>>>>>> which takes some hours as u know. >>>>>>>>>>>>>> >>>>>>>>>>>>>> Comparing my code to yours, I can only say we have >>>>>>>>>>>>>> exactly the same >>>>>>>>>>>>>> thing from all points from 1 to 4, except in dont have >>>>>>>>>>>>>> the line >>>>>>>>>>>>>> Q_EXPORT_PLUGIN2(..., mitk::PluginActivator) >>>>>>>>>>>>>> >>>>>>>>>>>>>> And since MITK is not ready yet in my machine I cannot >>>>>>>>>>>>>> test with >>>>>>>>>>>>>> this extra line. >>>>>>>>>>>>>> >>>>>>>>>>>>>> As soon as I am ready, I will email the list. >>>>>>>>>>>>>> Best, >>>>>>>>>>>>>> Miguel >>>>>>>>>>>>>> On Thursday, December 5, 2013 13:21 CET, Sebastian >>>>>>>>>>>>>> Ordas >>>>>>>>>>>>>> <[email protected]> wrote: >>>>>>>>>>>>>>> Hi Miguel, >>>>>>>>>>>>>>> >>>>>>>>>>>>>>> Are you still stuck on loading the xml files from your >>>>>>>>>>>>>>> plugin? >>>>>>>>>>>>>>> Let me know if you have tried the suggestion below. It >>>>>>>>>>>>>>> did not work >>>>>>>>>>>>>>> for me >>>>>>>>>>>>>>> >>>>>>>>>>>>>>> best regards, >>>>>>>>>>>>>>> sebastian >>>>>>>>>>>>>>> >>>>>>>>>>>>>>> On 04/12/2013 02:29 p.m., Sascha Zelzer wrote: >>>>>>>>>>>>>>>> Hi, >>>>>>>>>>>>>>>> >>>>>>>>>>>>>>>> in >>>>>>>>>>>>>>>> >>>>>>>>>>>>>>>> m_Interactor->LoadStateMachine( >>>>>>>>>>>>>>>> "SegmentationInteraction.xml"); >>>>>>>>>>>>>>>> >>>>>>>>>>>>>>>> >>>>>>>>>>>>>>>> you are not specifying a module as the second argument. >>>>>>>>>>>>>>>> It will >>>>>>>>>>>>>>>> default to the "Mitk" module for searching the XML >>>>>>>>>>>>>>>> file, which of >>>>>>>>>>>>>>>> course doesn't contain it. Use >>>>>>>>>>>>>>>> >>>>>>>>>>>>>>>> m_Interactor->LoadStateMachine( >>>>>>>>>>>>>>>> "SegmentationInteraction.xml", >>>>>>>>>>>>>>>> GetModuleContext()->GetModule()); >>>>>>>>>>>>>>>> >>>>>>>>>>>>>>>> >>>>>>>>>>>>>>>> instead. The same goes for the SetEventConfig() method >>>>>>>>>>>>>>>> call. I agree >>>>>>>>>>>>>>>> that the documentation is lacking and the API could be >>>>>>>>>>>>>>>> improved... >>>>>>>>>>>>>>>> >>>>>>>>>>>>>>>> Best, >>>>>>>>>>>>>>>> >>>>>>>>>>>>>>>> Sascha >>>>>>>>>>>>>>> ------------------------------------------------------------------------------ >>>>>>>>>>>>>>> >>>>>>>>>>>>>>> >>>>>>>>>>>>>>> >>>>>>>>>>>>>>> Sponsored by Intel(R) XDK >>>>>>>>>>>>>>> Develop, test and display web and hybrid apps with a >>>>>>>>>>>>>>> single code base. >>>>>>>>>>>>>>> Download it for free now! >>>>>>>>>>>>>>> http://pubads.g.doubleclick.net/gampad/clk?id=111408631&iu=/4140/ostg.clktrk >>>>>>>>>>>>>>> >>>>>>>>>>>>>>> >>>>>>>>>>>>>>> >>>>>>>>>>>>>>> _______________________________________________ >>>>>>>>>>>>>>> mitk-users mailing list >>>>>>>>>>>>>>> [email protected] >>>>>>>>>>>>>>> https://lists.sourceforge.net/lists/listinfo/mitk-users >>>>>>>>>>>>>> -- >>>>>>>>>>>>>> VRVis Zentrum fuer Virtual Reality und Visualisierung >>>>>>>>>>>>>> Forschungs-GmbH FN: 195369h, HG Wien >>>>>>>>>>>>>> Donau-City-Strasse 1 +43(0)1 20501 30100 >>>>>>>>>>>>>> A-1220 Wien, Austria web:<http://www.vrvis.at> >>>>>>>>>>>>>> >>>>>>>>>>>>>> >>>>>>>>>>>>>> ------------------------------------------------------------------------------ >>>>>>>>>>>>>> >>>>>>>>>>>>>> >>>>>>>>>>>>>> >>>>>>>>>>>>>> Sponsored by Intel(R) XDK >>>>>>>>>>>>>> Develop, test and display web and hybrid apps with a >>>>>>>>>>>>>> single code base. >>>>>>>>>>>>>> Download it for free now! >>>>>>>>>>>>>> http://pubads.g.doubleclick.net/gampad/clk?id=111408631&iu=/4140/ostg.clktrk >>>>>>>>>>>>>> >>>>>>>>>>>>>> >>>>>>>>>>>>>> >>>>>>>>>>>>>> _______________________________________________ >>>>>>>>>>>>>> mitk-users mailing list >>>>>>>>>>>>>> [email protected] >>>>>>>>>>>>>> https://lists.sourceforge.net/lists/listinfo/mitk-users >
------------------------------------------------------------------------------ Rapidly troubleshoot problems before they affect your business. Most IT organizations don't have a clear picture of how application performance affects their revenue. With AppDynamics, you get 100% visibility into your Java,.NET, & PHP application. Start your 15-day FREE TRIAL of AppDynamics Pro! http://pubads.g.doubleclick.net/gampad/clk?id=84349831&iu=/4140/ostg.clktrk _______________________________________________ mitk-users mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/mitk-users
