Doesn't handle UTF-8 properly

The _autotag_search_field function is using preg_split to split the fields into words, which it then checks to see if they match terms or synonyms in the selected vocabularies. However, preg_split doesn't handle utf-8 correctly. Any accented characters are being filtered out. A word like "cumpleaños" is being split into "cumple" and "os". The strtolower function being used to set all characters to lowercase, apparently also does not handle utf-8.

I have found one possible solution, which is as follows:

Replace this line:
$words_including_small = preg_split('/[\ `!"£$%^&*()_\-+={\[}\]:;@\'~#<,>.?\/|\\\]/', strtolower($field), -1, PREG_SPLIT_NO_EMPTY);

with these 2 lines:

mb_regex_encoding( "utf-8" );
$words_including_small = mb_split('\W', mb_strtolower($field,'UTF-8'));

There is also one other call to strtolower lower down in this function which should also be replaced with mb_strtolower.