bug#34524: wc: word count incorrect when words separated only by no-break space

[Paul Eggert]

Bruno Haible wrote:
> I would find it best to introduce an option '--unicode'
> to 'wc', that would produce Unicode compliant results, at the cost of
>    - not following POSIX to the letter,

It'd make sense to have an option. How about a more-general option --words, that 
would let the user define what a word is? This option's operand could use ERE 
syntax, or a shorthand beginning with '+' for common combinations. For example, 
the command:

wc --words='[[:alnum:]]+'

would say that a word consists of the longest contiguous sequence of 
alphanumeric characters. And

wc --words='+unicode'

would use the Unicode definition of word, whatever it is.