From: Kenneth Whistler (kenw@sybase.com)
Date: Thu Apr 10 2003 - 17:19:43 EDT
Chris Hopkins asked:
> Many thanks for the explanation. So by mapping the archaic forms of a Greek
> letter to the basic Unicode Greek codepoint, the sorting will be correct. I
> think I see how this works for Variation Selectors (which must be officially
> approved), but I don't see how it could work if using things I control, say,
> OpenType unmapped glyphs with <aalt> or <salt> entries. Can these be sorted?
They can, but you'd have to do some preprocessing and possibly some
tailoring.
Effectively, you preprocess your OpenType unmapped glyphs through
a filter to some character value. For example, if you were
representing a square-theta-with-a-dot as an OpenType unmapped
glyph, you could either filter that to convert it to a theta
for sorting (in which case you wouldn't need tailoring for
the collation table), or filter it to a private use code
point, or to a theta plus a private use code point representing
your own scheme for tracking variants. In the latter two cases,
you'd need access to an implementation of Unicode collation
which would allow you to tailor your table to incorporate
your use of private use code points, appropriately. The
ICU library implementation, for example, gives you a Unicode collation
algorithm implementation and also has a public mechanism for
tailoring, so with that you could accomplish what you need
to do.
--Ken
This archive was generated by hypermail 2.1.5 : Thu Apr 10 2003 - 18:12:51 EDT