Handling 'nuktas'

[mhosken]

With the advent of Unicode 11, we now have U+1133C as the double dotted 'nukta' and U+1133B as the single dotted. The font should be modified to take this into account. The styling for ring versus filled dot can remain as it is with appropriate defaults.

I suggest u1133C->u1133B, u1133C.double -> u1133C, u1133C.ring -> u1133B.ring

This is a breaking change for existing data. The good news is that the change will be obvious and not too functionally bad (although the styling will be messed up).

[devosb]

The single ring style is now the default. That does not contradict your suggestion, but I wanted to mention this so everyone is clear what the changes will mean.

The two codepoints mentioned have different properties. Both have ScriptExtensions of {Gran Taml}. Ken Whistler in https://www.unicode.org/L2/L2016/16084-badaga-single-dot.pdf writes "in implementations that support Script Extensions". So, are there implementations that do not support Script Extensions?

If so, then using U+1133B both both single and double forms might work better, as it has a script of Inherited, while U+1133C has a script of Grantha. However, Ken (and he write that Michael Everson as well) think that the single and double forms should have different codepoints.