There are generic sort descriptions for various code pages.

You could write one for a particular language.


An ordering of characters for a given code page.
Characters are represented either as themselves (in unicode) or
as two or more hex digits of the unicode representation.

There are three ordering strengths represented in this file.

These are Primary (different letters), secondary (different
accents), tertiary (different case).
See the java documentation for the Collator class for some more
discussion of the strength concept and examples.

Note that primary differences always determine the order even if
they are later in the word than secondary differences.
ie A B comes after A-acute A, even though A-acute sorts after A.

The word 'code' starts the ordering section.

Primary differences are represented by the '<' separator.
Characters with secondary differences are separated by semicolons
and characters with tertiary differences are separated by commas.

The code section ends if the word 'expansion' is seen.
This introduces a character that should sort as though it is
two (or more) separate characters.


ID values
---------

I believe that these are arbitary identifiers.  Here is a registry of
values we are using.  If you make a variation on a code-page
sort-order then give it a different id2 value.

code-page  id1  id2

1250       12   1
1251        8   1
1252        7   2
1253       13   1
1254       14   1
1255       15   1
1256       16   1
1257       17   1
1258       18   1
874        11   1
932         9   1
936         5   1
949        10   1

65001      19   4
0          0    0