New Dispatcher Issues
-
@ulrik I compiled again and now it worked...
-
@ulrik @Christoph-Hart but if I compile one of my projects, that easily compiled with the "develop code", it fails with "No global header file was found" using the NewDispatcher code, how come?
-
- Try deleting the entire Builds folder
- Try a new empty project
-
@Christoph-Hart Ok, I found the issue, the project I was compiling had some files in the "AdditionalSoureCode" folder which doesn't belong to the project.
I deleted it and now it's compiling!
Thank you Christoph! -
Building HISE on Linux is currently segfaulting. (Using latest commits)
[Thread debugging using libthread_db enabled]Using host libthread_db library "/lib/x86_64-linux-gnu/libthread_db.so.1".Programreceived signal SIGSEGV, Segmentation fault.juce::CharPointer_UTF8::isEmpty (this=0x2f7e258 <juce::juce_xmltextContentAttributeName>) at ../../../../JUCE/modules/juce_core/text/juce_CharPointer_UTF8.h:7373 bool isEmpty() const noexcept { return *data == 0; }
-
@d-healey I just reenabled the Linux Font Handler (I usually have it disabled because it slows down HISE startup). And I get a bit further before the segfault
-
jupp, that's on me. Weird that it crashes though. Have you changed the
HISE_USE_NEW_CODE_EDITOR
preprocessor? That would explain it, otherwise... -
@Christoph-Hart It seems to be the fix autocomplete commit where the issue started, the commit before that works.
-
Develop branch still segfaulting on Linux. Anything I can do to help narrow it down?
-
@Christoph-Hart said in New Dispatcher Issues:
Have you changed the HISE_USE_NEW_CODE_EDITOR preprocessor?
Oh somehow I totally missed your question. I haven't touched that. I'm just building the unmodified develop branch straight from your repo.
-
@Christoph-Hart Still segfaulting on Linux with the latest develop branch
-
Maybe this is useful...
It's returning a nullptr from
juce::Component::findParentComponentOfClass<hise::ComponentWithBackendConnection>()
So is there some component not being created when it should be?
Segfault occurs on line 95
-
I've narrowed it down to this commit - https://github.com/christophhart/HISE/commit/b2471b5bbe54fe177ee43eb57a4a7839bcd0aa48
Commenting out line 118 in
ScriptingPanelTypes.cpp
gets rid of the segfault, not sure why though... -
@Christoph-Hart Any suggestions??
-
Triggered by line 78 too. I think I'm getting close.
Shouldn't
any
here be a component not a mouse listener? -
@d-healey I ran into this now after resetting the layout. It should be fixed now but let me know if it still crashes on Linux.
-
@Christoph-Hart WooHoo! Thank you, it's working here.
-
I'm seeing these
Skipped changed() callback
messages each time I hit compile in the latest version. Are they necessary and any way to turn them off? -
@d-healey I noticed this as well. Is it actually a warning we need to be aware of?
-
I vaguely remember that there was an issue with the changed function not doing anything in the onInit callback.
However I can‘t change this behaviour for backwards compatibility reasons so I‘ve added a warning there to communicate this. So nothing has changed, but now you know what‘s up.