lmms
lmms copied to clipboard
LMMS
Bump veal submodule
- Adds MSVC support to "Calf" plugins.
- Adds several upstream patches
See also https://github.com/LMMS/veal/pull/9, https://github.com/LMMS/veal/pull/5, https://github.com/LMMS/veal/pull/6, (etc)
~Have found some issues. Old project files where i've used the calf plugins seem to crash. Maybe cause of missing upgrade mappings.~ I happened to mess up on the building, no such issues when i rebuilt clean. Also, old files load fine, not much issues. Haven't tested indepth though. If someone could test it better, would be appreciated.
old files load fine, not much issues. Haven't tested indepth though. If someone could test it better, would be appreciated.
Thanks for testing. I realize now that quite a bit has happened on our fork, this adds more than just MSVC support. We can merge it and cross our fingers or wait on some detailed testing.
feel free to merge this before #6758. That will make it easier to debug calf related issues in the future and keep that pr for enabling support alone, bumping can be done by this pr instead.
If it were just a matter of MSVC changes, I'd merge this without question, but since there were several fast-forwards from upstream code changes, it increases the chance of a regression. I can merge this, but I will not have time to jump into tracking down regressions, so this is a risk. I just want another developer to OK this.
Happened to take a look at the code again and saw that the majority changes are, in fact, msvc changes. Also, most of the other changes look harmless too. The only suspect possible of causing any regression is in the orfanidis_eq.h file. I wanted to test deeper but college is interfering pretty hard so can't do anything other than looking at the diff on GitHub via mobile.
Edit : took another look and found that the remainder of the changes were either version bumps, xml changes or gui related (which we don't currently use as we have our own). Also about the orfanidis_eq.h file. There was a refactor in that file. Don't think that might cause any regressions or issues.
Inviting @zonkmachine for testing.
From a code perspective, no complaints against this (but testing would not harm).
Note that in the recent days, I merged a few commits to calf's master, and there may follow some in the next week, as I am setting up a CMake base CI for it, including MSVC support. Maybe it is better to still wait? But I am OK with merging, too.
The veal submodule has 3 new commits since this PR was opened, so they could be included too.
Though if #7015 is merged first, that won't be needed.
