After updating
to latest beta (2017.03.21 14:21), I’m afraid I still get
wrong
alphabetic sorting for both Spanish and ancient Greek texts.
See
http://www.ousia.tk/grc-order.pdf (and its source file
http://www.ousia.tk/grc-order.tex).
In
Spanish, acute accented vowels are sorted after their unaccented
versions.
In
ancient Greek, some words are simply misplaced.
I don’t know
what I am missing. In the suite, Linux 32bit contains
luateX-1.0.3. I
don’t know whether this might affect to word sorting.
I must
confess that I don’t understand why replacements require Greek
letters
with diacritics to be converted to the Greek letter without
diacritic
and a Latin character. Or why replacements are needed at all.
I
mean, I understand that "ä", "ö", "ü" are sorted in German in three
different
ways. But these are different letters, after all.
If diacritics
don’t affect to sort order in Greek, which is the use of
the
replacements?
{ "α", "αa" }, { "ά", "αb" }, { "ὰ", "αc" }, {
"ὰ", "αd" }
In the line above, isn’t "ὰ" replaced by two
different values? Is this ok?