Re: Latin vs non-Latin words in text search parsing

Michael Glaesemann <grzm@seespotcode.net> writes:
>> Tom Lane wrote:
>>> asciiword
>>> word
>>> numword

> No huge preference, but I see benefit in what Gregory was saying re:  
> asciiword, alphaword, alnumword. word itself is pretty general, while  
> alphaword ties it much closer to its intended meaning. They've got  
> pretty consistent lengths as well. Maybe it leans too Hungarian.

I stuck with the previous proposal, mainly because I was already pretty
well into making the edits by the time I saw your message.  But I think
that with this definition "word" matches pretty well with everyone's
understanding of that, and the other two are supersets and subsets that
might have specific uses.
        regards, tom lane