Vyhledávání
Word boundaries: [[:<:]] [[:>:]]
Problems with UTF-8
REGEXPs doesn't support utf-8 chars!
For example "SELECT 'koš' REGEXP 'ko[[:alnum:]]';" returns 1 in command-line, but when matching against some utf-8 col stored in the DB it returns 0.
http://stackoverflow.com/questions/19774618/mysql-regex-utf-8-characters
Workaround with HEXa conversion:
# SELECT HEX("fotografický") => 666F746F6772616669636BC3BD
# Hexa of space = 20, tab = 09, newline = 0a
SELECT HEX("a fotografický seminář") REGEXP '(20|09|0a|^)666F746F6772616669636BC3BD(20|09|0a|$)';