Some changes were made to ResourceManagerImpl, so it could be a true failure.
On 1/7/14 11:10 AM, "Erik de Bruin" <[email protected]> wrote: >Mustella just crashed again, on: > >TypeError: Error #1009: Cannot access a property or method of a null >object reference. >at >mx.resources::ResourceManagerImpl/findResourceBundleWithResource()[C:\jenk >ins_slave\workspace\flex-sdk_mustella\frameworks\projects\framework\src\mx >\resources\ResourceManagerImpl.as:882] >at >mx.resources::ResourceManagerImpl/findBundle()[C:\jenkins_slave\workspace\ >flex-sdk_mustella\frameworks\projects\framework\src\mx\resources\ResourceM >anagerImpl.as:1094] >at >mx.resources::ResourceManagerImpl/getStringArray()[C:\jenkins_slave\worksp >ace\flex-sdk_mustella\frameworks\projects\framework\src\mx\resources\Resou >rceManagerImpl.as:952] >at mx.controls::DateField/set dayNames() >at mx.controls::DateField/resourcesChanged() >at >mx.core::UIComponent()[C:\jenkins_slave\workspace\flex-sdk_mustella\framew >orks\projects\framework\src\mx\core\UIComponent.as:1722] >at mx.controls::ComboBase() >at mx.controls::DateField() >at mx.core::Container/createComponentFromDescriptor() >at mx.core::Container/createComponentsFromDescriptors() > >I'm currently not able to make time to fix whatever keeps causing >these RTEs. I rebooted the machine today, but that didn't fix it >either. There have never been three consecutive runs that all crashed >the player in such a way that Mustella was unable to recover. > >EdB > > > >On Tue, Jan 7, 2014 at 3:29 PM, Alex Harui <[email protected]> wrote: >> Did you ever try the ContinueAfterRuntimeException applet? It is a >> Windows C++ program whose source is in >> mustella/as3/src/mustella/ContinueAfterRuntimeException >> >> It tries to look for top-level windows with the exception dialog and >>kill >> the process. >> >> -Alex >> >> On 1/7/14 12:26 AM, "Erik de Bruin" <[email protected]> wrote: >> >>>Yes, #639. But the log won't be very helpful as I had to restart to >>>get Jenkins to reset the builds properly. I haven't figured out how to >>>stop a running suite once the player crashes, as it is not started by >>>Jenkins so much as it is by the script. >>> >>>EdB >>> >>> >>> >>>On Tue, Jan 7, 2014 at 8:40 AM, Alex Harui <[email protected]> wrote: >>>> Maybe. Is this from run #639? I'm not sure if jenkins causes this or >>>> vice versa. >>>> >>>> -Alex >>>> >>>> On 1/6/14 10:22 PM, "Erik de Bruin" <[email protected]> wrote: >>>> >>>>>Instant karma? The Mustella VM just crashed the Flash Player with: >>>>> >>>>>TypeError: Error #1009: Cannot access a property or method of a null >>>>>object reference. >>>>>at >>>>>mx.resources::ResourceManagerImpl/findResourceBundleWithResource()[C:\ >>>>>je >>>>>nk >>>>>ins_slave\workspace\flex-sdk_mustella\frameworks\projects\framework\sr >>>>>c\ >>>>>mx >>>>>\resources\ResourceManagerImpl.as:882] >>>>>at >>>>>mx.resources::ResourceManagerImpl/findBundle()[C:\jenkins_slave\worksp >>>>>ac >>>>>e\ >>>>>flex-sdk_mustella\frameworks\projects\framework\src\mx\resources\Resou >>>>>rc >>>>>eM >>>>>anagerImpl.as:1094] >>>>>at >>>>>mx.resources::ResourceManagerImpl/getString()[C:\jenkins_slave\workspa >>>>>ce >>>>>\f >>>>>lex-sdk_mustella\frameworks\projects\framework\src\mx\resources\Resour >>>>>ce >>>>>Ma >>>>>nagerImpl.as:925] >>>>>at spark.validators.supportClasses::GlobalizationValidatorBase/set >>>>>requiredFieldError() >>>>>at >>>>>spark.validators.supportClasses::GlobalizationValidatorBase/resourcesC >>>>>ha >>>>>ng >>>>>ed() >>>>>at spark.validators::CurrencyValidator/resourcesChanged() >>>>>at spark.validators.supportClasses::GlobalizationValidatorBase() >>>>>at spark.validators.supportClasses::NumberValidatorBase() >>>>>at spark.validators::CurrencyValidator() >>>>>at assets::CV_FlexModule() >>>>> >>>>>Might this be related? >>>>> >>>>>EdB >>>>> >>>>> >>>>> >>>>>On Mon, Jan 6, 2014 at 10:03 PM, OmPrakash Muppirala >>>>><[email protected]> wrote: >>>>>> On Jan 6, 2014 1:00 PM, "Alex Harui" <[email protected]> wrote: >>>>>>> >>>>>>> How do you know you don't have that locale? It may not be the >>>>>>>locale >>>>>>> itself, but maybe some ISO table that recognizes fr-CH as a locale >>>>>>>id. >>>>>>> The error in the failing test seems to be associated with the >>>>>>> currencyStringError. >>>>>>> >>>>>>> I think the character display is not related. Your editor either >>>>>>>is >>>>>>>set >>>>>>> up to use a font with those chars or not. But the bytes in the >>>>>>>file >>>>>>>are >>>>>>> probably the right ones. In the FDB debugger on the mac I get >>>>>>>????????!!! >>>>>>> but I'll bet the char codes are correct. >>>>>>> >>>>>> >>>>>> Makes sense. I will try to get this locale on the VM and see if the >>>>>>test >>>>>> passes. >>>>>> >>>>>> Thanks for taking a look. >>>>>> >>>>>> Thanks, >>>>>> Om >>>>>> >>>>>>> -Alex >>>>>>> >>>>>>> On 1/6/14 12:47 PM, "OmPrakash Muppirala" <[email protected]> >>>>>>>wrote: >>>>>>> >>>>>>> >On Jan 6, 2014 12:25 PM, "Alex Harui" <[email protected]> wrote: >>>>>>> >> >>>>>>> >> I looked at it a bit more and have another guess. This test >>>>>>>file >>>>>>>tries >>>>>>> >>to >>>>>>> >> assign all kinds of weird locales to the validators. In test >>>>>>>29, >>>>>>>it is >>>>>>> >> assigning fr-CH. >>>>>>> >> >>>>>>> >> My guess is that the fr-CH locale is not installed on that >>>>>>>machine. >>>>>>> > >>>>>>> >My question is why is it passing in my machine? I obviously don't >>>>>>>have >>>>>>> >that locale in my machine. Whereas, in the VM that it fails, it >>>>>>>does >>>>>> seem >>>>>>> >to at least display the characters correctly. >>>>>>> > >>>>>>> >I will see if I can install that locale on that VM. >>>>>>> > >>>>>>> > >>>>>>> >> >>>>>>> >> -Alex >>>>>>> >> >>>>>>> >> On 1/6/14 12:10 PM, "Alex Harui" <[email protected]> wrote: >>>>>>> >> >>>>>>> >> >Is it only one test in this file? It looks like other tests >>>>>>> >>successfully >>>>>>> >> >test for the chinese text? >>>>>>> >> > >>>>>>> >> >-Alex >>>>>>> >> > >>>>>>> >> >On 1/6/14 11:51 AM, "OmPrakash Muppirala" >>>>>>><[email protected]> >>>>>> wrote: >>>>>>> >> > >>>>>>> >> >>The test in question is: >>>>>>> >> >> >>>>>>> >> >> >>>>>>> >> >>>>>>> >>>>>>>>>spark/validators/CurrencyValidator/Methods/CV_validate_error_messa >>>>>>>>>ge >>>>>>>>>_l >>>>>>>>>oc >>>>>>> >>>a >>>>>>> >> >>l >>>>>>> >> >>ized_tester >>>>>>> >> >> >>>>>>> >> >>On my local machine, this test passes, but on my new VM, this >>>>>>>fails. >>>>>>> >> >> >>>>>>> >> >>The problem is that on my local machine, it appears that the >>>>>>>required >>>>>>> >> >>(Chinese?) fonts are missing. Yet, somehow the test passes. >>>>>>>Look >>>>>> at >>>>>>> >> >>screenshot [1] >>>>>>> >> >> >>>>>>> >> >>On the VM, the test fails, but it appears that the Chinese >>>>>>>text >>>>>>>seems >>>>>>> >>to >>>>>>> >> >>be >>>>>>> >> >>there. Screenshot [2] >>>>>>> >> >> >>>>>>> >> >>Is this a case of tests passing because they render the fonts >>>>>>> >incorrectly >>>>>>> >> >>in the same way? And it fails when the font is actually >>>>>>>available? >>>>>>> >> >> >>>>>>> >> >>Any ideas? >>>>>>> >> >> >>>>>>> >> >>Thanks, >>>>>>> >> >>Om >>>>>>> >> >> >>>>>>> >> >>[1] http://snag.gy/61l8J.jpg (invalid characters, test passes) >>>>>>> >> >>[2] http://snag.gy/t5PKu.jpg (valid characters, test fails) >>>>>>> >> > >>>>>>> >> >>>>>>> >>>>> >>>>> >>>>> >>>>>-- >>>>>Ix Multimedia Software >>>>> >>>>>Jan Luykenstraat 27 >>>>>3521 VB Utrecht >>>>> >>>>>T. 06-51952295 >>>>>I. www.ixsoftware.nl >>>> >>> >>> >>> >>>-- >>>Ix Multimedia Software >>> >>>Jan Luykenstraat 27 >>>3521 VB Utrecht >>> >>>T. 06-51952295 >>>I. www.ixsoftware.nl >> > > > >-- >Ix Multimedia Software > >Jan Luykenstraat 27 >3521 VB Utrecht > >T. 06-51952295 >I. www.ixsoftware.nl
