Invoking undo in REAPER after changes within ReaLearn can lead to unpleasant surprises because not each little change within ReaLearn is reported to REAPER (that would be resource-consuming but also difficult to do correctly because ReaLearn's state is naturally captured by REAPER's undo system due to its nature as VST plug-in).
ReaLearn and Playtime should be considered as its own "undo universe". Each Playtime Matrix already has its own undo history. And ReaLearn will probably also get its own one at some point.
That's why I want to ignore REAPER-undo-system-triggered state changes. With REAPER >= v7.09+dev0202 this is possible.
Invoking undo in REAPER after changes within ReaLearn can lead to unpleasant surprises because not each little change within ReaLearn is reported to REAPER (that would be resource-consuming but also difficult to do correctly because ReaLearn's state is naturally captured by REAPER's undo system due to its nature as VST plug-in).
ReaLearn and Playtime should be considered as its own "undo universe". Each Playtime Matrix already has its own undo history. And ReaLearn will probably also get its own one at some point.
That's why I want to ignore REAPER-undo-system-triggered state changes. With REAPER >= v7.09+dev0202 this is possible.