From: Eric B. <er...@go...> - 2007-02-04 17:12:47
|
Colin Paul Adams wrote: > Eric> I think that would have to write a preparser that would > Eric> translate XPath regular expressions into PCRE regular > Eric> expressions. > > I don't think it can work, because the trick of using the UTF-8 byte > representation will fail in these kind of tests. I must confess that I don't know what you are talking about with "these kind of tests". Are you talking about \i and \c? What does "all initial name characters and all name characters" mean? To what extent wouldn't that work with the UTF-8 byte representation? Even if it's a multibyte, you can replace [\i:] by (multibyte1|...|multibyten|:) and likewise for similar regexp constructs. -- Eric Bezault mailto:er...@go... http://www.gobosoft.com |