Your string-to-codepoints example only works for ASCII upper/lower case
letters. It fails to recognize composed and decomposed diacritical characters
such as a combined uppercase A with a grave U+00C1, with an accute U+00C1,
with a circumflex U+00C2, etc. Yes you could detect these too with additional
logic, but matches() with a character class of \p{Ll}, \p{Lu}, \p{Lt} handles
all the messy details of Unicode.

Andy.

> -----Original Message-----
> From: Andrew Welch [mailto:andrew.j.welch@xxxxxxxxx]
> Sent: Thursday, November 03, 2011 12:16
> To: xsl-list@xxxxxxxxxxxxxxxxxxxxxx
> Subject: Re:  Testing for upper and lower case
>
> >  Is it possible to determine if the first
> > letter of a word is upper or lower case?
>
> string-to-codepoints($str)[1]
>
> will give you the codepoint of the first letter, you can then if its:
>
> 65 to 90 its upper case
>
> 97 to 122 its lower case
>
> is that what you meant?
>
>
> --
> Andrew Welch
> http://andrewjwelch.com