For a simple character set named MYSET
, sorting rules are specified in the
configuration file using MYSET
.xml<map>
array elements within <collation>
elements. If the sorting rules for your language are too complex to be handled with simple arrays, you must define string collating functions in the ctype-
source file in the MYSET
.cstrings
directory.
The existing character sets provide the best documentation and examples to show how these functions are implemented. Look at the ctype-*.c
files in the strings
directory, such as the files for the big5
, czech
, gbk
, sjis
, and tis160
character sets. Take a look at the MY_COLLATION_HANDLER
structures to see how they are used. See also the CHARSET_INFO.txt
file in the strings
directory for additional information.