Newest jar @ www.ambrosoft.com This time it is XSLTMark's "trend" that is 'fixed'.
I quoted 'fixed' because the problems discovered are very serious and the real fix will require more reprogramming. "trend" works today and brings back the same results as Saxon and Xalan, but because of a conceptual bug the computation is O(n^2). Also, "trend" only helped to point towards a much more fundamental issue with "preceeding-siblings" -- this error should have been caught by much more fundamental quality tests (just like "alphabetize" before). In other words some fundamental XPath computations are broken in XSLTC :-( Next, I am going to work on some Cocoon showstoppers and other hi priority bugs. If I could sustain current rate of bug per day XSLTC should work soon :-) And it would be a good thing because I now need it for my commercial work, i.e. I need XSLTC as a USER! --Jacek _________________________________________________________ Do You Yahoo!? Get your free @yahoo.com address at http://mail.yahoo.com --------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, email: [EMAIL PROTECTED]
