Align function enhancement; add new 'relative=true/false' argument

[luciansmith]

If I set the layout size to 1000,1000, call autolayout, then align some element (i.e. 'alignTop'), sometimes the position of those elements is not in the appropriate swath of the layout: when I call 'alignTop', for example, nine times out of 10, the Y value is 30, and often it's in the 30's to 200's. But sometimes it's close to the middle, like '432'. I think that's too far off to really say that's at the 'top' of the layout; it's more like the hCenter.

One time, I got 'nan' values, which is also not ideal, but I don't know how to reliably get that to repeat.

[adelhpour]

'alginTop' function calculates the min of the 'y' values of the selected items and set the 'y' values of all the items to that one. So, I wouldn't be surprised if you occasionally get something near to the center due to the randomness of the autolayout. In my test, I compare all the 'y' values to the min value after calling 'alignTop'.

[luciansmith]

I think this behavior is not what people would expect. I think think if you're aligning elements to 'the top', people think of 'the top' as 'the top of the page' not 'the top of the cluster of elements that I don't know where they are'.

That said, it's possible that people might want your behavior in some conditions. Maybe a 'relative=true/false' argument to the functions?

(for version 2; I think the current state of things is probably fine for now.)

[adelhpour]

Great. We will implement it in the version 0.2.