Michael Gregorius

I always assumed that the root `include` directory was meant as the public interface to LMMS. Which is also the reason why I had put the include of my conversion...

Thanks, @Veratil! Good to know. I guess it will stay like that because if they were ever moved next to their implementation files like in other projects then this would...

I like the simplicity of the `DataFileUpgrade` class. It's straight forward and provides useful helper methods. It should even be possible to implement the `UpgradeContainer` schema with it if an...

> The problem is that it's too much boilerplate to have to define a whole new class that only contains a simple `upgrade` function that runs only a few lines...

The advantage of a base class is that the code already contains the full context of the upgrade, i.e. the XML data as well as all available helper methods. If...

@JohannesLorenz, a while ago I always made sure that my commit messages all had line breaks at around 74 characters. However, this makes it rather tedious to write commit messages...

I have now merged this. It should be rather safe as it "only" upgrades files that are loaded but should have no effect on files that are written. So, if...

Might be. What are the affected instruments that do not null in the two exports? Can you do isolated tests with them, i.e. not in such a complex project?

> _Produce only similar output_: Kicker, Triple Oscilator, LB302, Organic, Vibed, ZynAddSubFx _Some "stars":_ OpulenZ, VeSTige :: DSK Harmonica _Little noise shadow_: Nescaline > > _Tested also (produce the same...

The files had the BPMs added and thus have been renamed with pull request #6747 by @mirk0dex. Might be that the upgrade routine that was also provided by the PR...