[Schematron] Help sought: implementation of Character Repertoire in XSLT2 for embedding in schematron

Dave Pawson dave.pawson at gmail.com
Sat Sep 20 10:22:30 EDT 2008


2008/9/20 David Carlisle <d.p.carlisle at googlemail.com>:
>>
>> That bothers me; processing a single character at a time is grossly
>> inefficient.
>
>
> tough:-)
>
> Since the whole point of crepdl is to check that every character is in a
> specified set of characters then something is going to have to process every
> character and compare it against the specified set. If you have a system
> that has a regex engine embedded in an xpath engine embedded in an xslt
> engine, hidden as a schematron engine, then there is some flexibility as to
> which bit checks what, but _something_ has got to walk over  every character
> checking it against the specified character set.


So it would make sense to reduce the number of calls from one per character
at the top of this heap? Processing a 'string' (however obtained) would
reduce that overhead by len(string) calls?


regards

-- 
Dave Pawson
XSLT XSL-FO FAQ.
Docbook FAQ.
http://www.dpawson.co.uk



More information about the Schematron mailing list