Closed alexandermorgan closed 9 years ago
Wow, thanks for the awesome screenshot! You can fix this :D
Is this solved now?
Yes but I didn't close the issue because my fix doesn't pass the tests and so isn't integrated into master.
Oh right---this caused a big mess in the WorkflowManager's unit tests, and I said I would fix it. Assigning to myself.
I'm going to have to rewrite these unit tests in the WorkflowManager anyway, for the 2.x series. So I'm going to move this issue there...
Closed with commit 05266a25b4fedbe1d346bae13fec214cb581a748.
Horizontal intervals labelled in n-grams analyze the last event rather than selecting notes according to the offset interval. In the example below (note the C clefs) produced using an offset interval of 2 (the half note), the first 2-gram should read 6 3 3 rather than 6 2 3. The 2 in the 6 2 3 is presumably arrived at by calculating the horizontal interval from D to E instead of from C to E as was expected. Note that in the fourth 2-gram shown it is evident that even eighth notes will engender the same bug. This may be a bug in the FilterByOffset class. (Note that the piece shown is Lassus's first of twelve duets).