has something changed in Scripnode compiled nodes?

[Dan Korneff]

After the latest update, the "Frozen" icon is no longer visible when I compile DSP networks as dll.
I'm also noticing some inconsistency with compiling scripts as well as compiling all scripts. Sometimes my changes take effect, and sometimes they don't

[Christoph Hart]

After the latest update, the "Frozen" icon is no longer visible when I compile DSP networks as dll.

Actually yes. I've removed the ability to toggle between the compiled node and the DSP network (which is what the button was for). Now it just uses the compiled node without creating a (hidden) DSP network for each compiled node. This was causing multiple issues & glitches (especially when using external filter display coefficients) and I thought the functionality was only useful at the beginning for me to check that the behaviour is consistent.

I'm also noticing some inconsistency with compiling scripts as well as compiling all scripts.

Can you elaborate on that? I've done quite a bit of refactoring of the entire DLL node system (mainly to allow it connect to global cables etc), but I don't see any regression on my projects, but if some of your nodes don't work anymore, let me know then I'll investigate.

[Dan Korneff]

@Christoph-Hart said in has something changed in Scripnode compiled nodes?:

Can you elaborate on that?

This part is not related to Scriptnode (as far as I can tell).
If I create a variable in my script and hit Compile, sometimes it doesn't show up in the ScriptWatchTable. I have to hit Shift+F5. And even then, sometimes that doesn't fix it. I have to restart. Even then the variable doesn't show up until I hit compile.

I'm noticing a lot of overall crashes with HISE standalone as well.

Any my exported plugins are now crashing with:

Exception thrown at 0x00007FF971D0567C in PluginDoctor64.exe: Microsoft C++ exception: juce::JSONParser::ErrorException at memory location 0x00000035EA8FF4E8.
The thread 0x46b8 has exited with code 0 (0x0).
The thread 0x2abc has exited with code 0 (0x0).
Debug Error!

HEAP CORRUPTION DETECTED: after Normal block (#1714932) at 0x000001CCB2C5BF70.
CRT detected that the application wrote to memory after end of heap buffer.

Looks like I have a nice long day of debugging ahead of me :detective: