[erlang-questions] [announce] Unicode Collator

Michael Uvarov freeakk@REDACTED
Tue Feb 28 19:15:09 CET 2012


It handles strings as utf-8 binaries.
It is slower than ICU in difficult cases, but it has optimizations for
ASCII subset chars. It is fast when comparing string with a common
prefix and for cases with differences in the beginning of the strings.
Difficult cases are when there are ignorables in the beginning of the strings:
" test" and "test".
And it definitely faster than ux.



More information about the erlang-questions mailing list