[Home] [By Thread] [By Date] [Recent Entries]
Florent Georges wrote:
There, it is not clearly stated that applies to AVT too. But a little bit further in the same section, there is: Michael Kay just confirmed this to be a bug indeed. I guess you mean: Oops! You are right! If I'm right, this is not permitted by XML. So it is the XML parser that will generate the error, before the XSLT processor can see the input. You are right on XML 1.0. 1.1 however, allows for almost everything, except \x00. But XML is not the problem. I'd like to use unparsed-text(), which by definition, bypasses the XML processor. But in a way I can understand the conflict rising when allowing data to enter the processor with illegal characters, because there seems no way to represent them in XML, and so in XSLT. However, this shouldn't be too hard, should it: unparsed-text( [uri], [codepage], [illegal-char-treatment] ) where "illegal-char-treatment" would be something like "remove", "spaces", "error" etc. But here I am inventing my own extension function, like Michael Kay suggested. Thanks & Cheers, Abel
|

Cart



