overwriting silence in patterns

[larkob]

I'm still trying to understand how silence works and how it can be "undone" (no use case, just out of curiousity). This time, I've tried to fill out a part in a pattern that initially was empty, and even though at first the result seemed to be promising using legato, once I added another struct function to split the cycle up again, I didn't get the expected result. See here an example where I compare a pattern using silence + legato with a pattern that has legato applied to the third item via mini-notation syntax. Again, is this working as designed? And why? Is there maybe some better way for me to inspect the results of patterns rather than the map/showWhole approach?

"c2 d2 e2 ~".legato("1 1 2 0").struct("x x x x").firstCycle().map(h=>h.showWhole())`
// ['0/1 -> 1/4: c2', '1/4 -> 1/2: d2', '1/2 -> 3/4: e2']

"c2 d2 e2@2".struct("x x x x").firstCycle().map(h=>h.showWhole())
// ['0/1 -> 1/4: c2', '1/4 -> 1/2: d2', '1/2 -> 3/4: e2', '3/4 -> 1/1: e2']

[felixroos]

you can visualize using .pianoroll: https://strudel.tidalcycles.org?0KNPD8AmV-ms

[larkob]

That's good to get the timing, but it also doesn't give me the full information. The strange thing is that "c2 d2 e2@2" and "c2 d2 e2 ~".legato("1 1 2 0") seem to produce exactly the same pattern for all that I can see with show, drawLine or pianoroll. So, how come the struct function produces a different result with them, as if there was some hidden information that Strudel knows about but doesn't reveal?