Generate a regex to detect unicode blocks in mysql.

mysql_unicode_regex.py

#!/usr/bin/python
# -*- coding: utf-8 -*-

start, end = u'가힝' # Korean
#start, end = u'あゟ' # Japanese

def esc(c):
    if c in '[]\\\'\"':
        return '\\' + c
    return c

def inc(c):
    return chr(ord(c) + 1)

def dec(c):
    return chr(ord(c) - 1)

def rr(u, v):

    l = len(u)
    m = l - 1

    if u == '':
        return ''

    if u[0] == v[0]:
        return esc(u[0]) + rr(u[1:], v[1:])

    if l == 1:
        if u == '\x00':
            if v == '\xFF':
                return '.'
            else:
                return '[^%s-\xFF]' % (esc(inc(v[0])),)
        else:
            return '[%s-%s]' % (esc(u), esc(v))

    u_complete = u == '\x00' * l
    v_complete = v == '\xFF' * l

    if u_complete and v_complete:
        return '.' * l
    elif u_complete:
        return '(' + '|'.join([
            rr('\x00', dec(v[0])) + rr(u[1:], '\xFF' * m),
            esc(v[0]) + rr('\x00' * m, v[1:]),
        ]) + ')'
    elif v_complete:
        return '(' + '|'.join([
            esc(u[0]) + rr(u[1:], '\xFF' * m),
            rr(inc(u[0]), '\xFF') + rr('\x00' * m, v[1:]),
        ]) + ')'

    parts = [esc(u[0]) + rr(u[1:], '\xFF' * m)]
    if ord(v[0]) - ord(u[0]) > 1:
        parts += [rr(inc(u[0]), dec(v[0])) + '.' * m]
    parts += [esc(v[0]) + rr('\x00' * m, v[1:])]
    return '(' + '|'.join(parts) + ')'

def to_mysql_x(regex):
    return ''.join(
        '%02X' % (ord(c),)
        for c in regex[1:-1]
    )

start_bytes = start.encode('utf-8')
end_bytes = end.encode('utf-8')

regex = rr(start_bytes, end_bytes)

print to_mysql_x(regex)