From: Yung-Fong Tang > Case mapping ? You have no way to generate mapping table for > case mapping with knowing the character unless you already > define those character have no case or only one case.
Um, Unicode defines a behavior and even properties for unassigned code points. If you choose not to implement this because you only handle "assigned code points" then that is actually a problem with your software. No one is arguing the point you make that until a code point is assigned, its exact *FINAL* behavior is not completely understood with regards to casing, collation, and everything else. So you do not need to continue arguing this point -- I am sure everyone agrees with it. But do you understand that there is certainly a defined behavior for it in the interim? In the time before it is assigned an actual character? That is I think the crux of the matter here. > Don't tell me there any people how implemented > HanCharacterStokeNumber(U+20000) in 1996, no body have a > implementation of HanCharacterStokeNumber(U+20000) until > U+20000 got defined. Actually, several companies had the mechanisms defined to convert that to a surrogate pair. Or to treat it as a single unassigned character for the purposes of collation. The difference between them and you would be that you do not recognize the existence of this state -- the time before direct assignment? MichKa Michael Kaplan Trigeminal Software, Inc. http://www.trigeminal.com/
