On Wed, 1 Aug 2012 18:09:27 -0700 Leo Broukhis <[email protected]> wrote:
> No, 20D7 is not a Diacritic, it is Other_Math, therefore the dot > should remain. In general, mathematical combining characters are not > diacritics. While not definitive, note that U+20D7 is in the 'Combining Diacritical Marks for Symbols' block. I would certainly expect soft-dotted letters to shed their dot for third or fourth derivatives written as fluxions, U+20DB COMBINING THREE DOTS ABOVE and U+20DC COMBINING FOUR DOTS ABOVE. (Mind you, I would arrange the dots in a triangle or square - am I using different characters?) > > Renderers that treat "combining" as a synonym for "diacritic" and > remove the dot are in error. > UAX 44 says, "Characters that linguistically modify the meaning of > another character to which they apply. Some diacritics are not > combining characters, and some combining characters are not > diacritics." The definitive text would appear to be Section 3.13 Definition D138 and Table 3-14, as confirmed by the usage in UCD file SpecialCasing.txt. The key property is canonical combining class 230, so 'i' would lose its dot for U+20D7, just as it should for a Hebrew accent or a Tai Tham or Tai Viet tone mark, but not a Thai or Lao tone mark. (The odds of getting reasonable, let alone compliant, rendering for any of these mixed script combinations is fairly low.) Richard.
