CompoundTuplet incorrect initialization

[massimobio]

let triplet = try? Tuplet(3, .eighth, notes: [eighthNote, eighthNote, eighthNote])
let compoundTuplet = try? Tuplet(5, .eighth, notes: [eighthNote, eighthNote, triplet!, eighthNote])
print(compoundTuplet!)

prints this: 6[1/8a1, 1/8a1, 3[1/8a1, 1/8a1, 1/8a1], 1/8a1] Which is incorrect.

There is a related test related to this in TupletTests.testInitSuccessForComplexCompound()

[sprzenus]

@massimobio What should be the result of this initializer in this case then?

[drumnkyle]

If I’m looking at it right, I believe it is just that the print out is wrong. I think the initialization seems correct but the print out should probably be a 5 instead of a 6. Because it is a tuplet of 5

[sprzenus]

If print is wrong, flatIndexes has wrong value.

[drumnkyle]

Hmm. I’d be surprised if that’s the case. I thought the tests should cover this and I don’t think the tests for this functionality are broken. Can you verify that this is still happening or do you need me to try it out?

[sprzenus]

I don't see any tests on noteCount or flatIndexes in TupletTests on master branch. Are you sure it is tested? If no, maybe it would be good to test it.

The function mentioned here is added to the travis branch. I would prefer adding the missing tests to another branch.

[drumnkyle]

Hmm. Yeah another PR would be fine. Let me try to look at it and see what’s going on first. I’ll comment back.

[sprzenus]

I'm not sure if it isn't just an understanding problem. You create tuplet with a duration of 5 eighths which is ok for 3 eighths and the triplet. And you get 6 notes. It seems to be correct

[drumnkyle]

Thanks @sprzenus. You're right. This is a misunderstanding. We are printing the note count. It would probably be nice to print the ratio, but we can make a separate issue to do that later if we want. I will close this issue then.