So, I guess, 1) to match Unicode classes in J means I have to get Unicode-compliant jpcre.dll and 2) to match them is hard because of various ways to represent Unicode. And I agree that C# isn't frequently used with Unicode identifiers - but the specification says so, and I don't want to really bend that.
Sigh. Looks like I need a practical workaround. And I wonder how to make Yahoo to display Unicode properly. ----- Date: Mon, 16 Mar 2009 23:28:28 -0700 (PDT) From: Oleg Kobchenko <[email protected]> Subject: Re: [Jprogramming] regex matching Unicode classes? To: Programming forum <[email protected]> Message-ID: <[email protected]> Content-Type: text/plain; charset=us-ascii Yeah, you need to fix your email client. So you are parsing C#. I have yet to see a C# program with Unicode identifiers. Obviously Unicode character classes are not supported by PCRE shipped with J. Also it may need to operate on two0byte Unicode rather than UTF-8. ---------------------------------------------------------------------- For information about J forums see http://www.jsoftware.com/forums.htm
