drop accented chars (unicode) to ascii equivalents

unicodedata_test.py

import unicodedata

accented= {
	'czech' : u'ťúůýžšřóňíěéďčá',
	'french' : u"ùûüÿàâæçéèêëïîôœ",
	'finnish' : u'äåö',
	'danish' : u'åæéø',
	'german' : u'äöüß',
	'hungarian': u'áéíöóőüúű',
	'icelandic': u'áæðéíóöþúý',
	'italian': u'àèéìòóù',
	'norwegian': u'åæâéèêøóòô',
	'polish': u'ąćęłńóśźż',
	'portuguese': u'úüãáâàçéêíõóô',
	'romanian': u'ăâîşșţț',
	'spanish': u'áéíñóúü',
	'swedish': u'äåéö',
	'welsh': u'ûüúùŵẃẅẁŷÿýỳäáàêëéèîïíôöóò'
}

for ( lang , chars ) in accented.iteritems() :
    print "-----"
    print lang
    print "  %s" % chars
    print "  %s" % unicodedata.normalize('NFKD',chars).encode('ascii','ignore')