#61 check combining of diacriticals with long epsilon and omicro

[BODARD Gabriel]

In Unicode NFC, we could possibly optimize the combining of long diacriticals (circumflex/iota subscript) with omicron and epsilon. The combining diacritic for accent+breathing in a single codepoint will always look better than these two combining diacritics separatim. At the moment, what Transcoder seems to give you for these vowels is:

Beta Code:
O,O=,O|,O=|,O|=,O)=,O)|,O)=|,O)|=,O(=,O(|,O(=|,O(|=
E,E=,E|,E=|,E|=,E)=,E)|,E)=|,E)|=,E(=,E(|,E(=|,E(|=

Unicode NFC:
ο,ο̂,οͅ,ο̂ͅ,ο̂ͅ,ὀ̂,ὀͅ,ὀ̂ͅ,ὀ̂ͅ,ὁ̂,ὁͅ,ὁ̂ͅ,ὁ̂ͅ
ε,ε̂,εͅ,ε̂ͅ,ε̂ͅ,ἐ̂,ἐͅ,ἐ̂ͅ,ἐ̂ͅ,ἑ̂,ἑͅ,ἑ̂ͅ,ἑ̂ͅ

[Elli Mylonas]

Hugh will revisit later

[Hugh A. Cayless]

The thing is, precombined forms for omicron and epsilon with circumflex don't exist. They were left out of the Greek Extended block, and the thinking nowadays is that it's better to have rendering engines do the right thing with the combining diacritics than to use precombined forms, so there are never going to be UFC glyphs for them.