Yes. Of course it's unclear whether people still want type-aware processing, now that so many have gone to typeless JSON, but...
The second biggest thing 3.0 changed, for me, was dropping the distinction between node sets and temporary trees. That's always been a nuisance. As long as we claim 2.0 compatibility we can't really fix that; moving to 3.0 justifies the breaking change. You've been working on the stuff that can be addressed as extensions to 2.0 without breaking it, which is a good stopgap approach. But at some point we're going to have to go beyond that if we want to be more that 2.0++. -- /_ Joe Kesselman (he/him/his) -/ _) My Alexa skill for New Music/New Sounds fans: / https://www.amazon.com/dp/B09WJ3H657/ () Plaintext Ribbon Campaign /\ Stamp out HTML mail! ________________________________ From: Mukul Gandhi <[email protected]> Sent: Wednesday, October 11, 2023 9:26:47 PM To: [email protected] <[email protected]> Subject: Re: Xalan retirement Hi Joseph, Thanks for the thoughts. > On Wed, Oct 11, 2023 at 6:46 AM Joseph Kesselman <[email protected]> wrote: > And possibly recreating the type-aware version of DTM. I'm quite keen to have (see) this support within XalanJ's XSLT 3.0 and XPath 3.1 processors (that shall allow us to have 'schema aware' XSLT and XPath processors within XalanJ, which is nice). Currently, the XalanJ development we're doing on dev repos branch xalan-j_xslt3.0, is by design building non 'schema aware' XSLT 3.0 and XPath 3.1 processors. With non 'schema aware' XSLT and XPath processors, to get general support of XML Schema data types within XPath 3.1 expressions (and to lesser extent within XSLT 3.0 stylesheets), we need to mostly do explicit casts within XPath expressions (for e.g, xs:date(xml_elem) gt current-date(). With 'schema aware' XPath processor, it should be possible to write XPath expressions like following : xml_elem gt current-date() where an XML element xml_elem has a run-time type annotation of xs:date by virtue of an XML element xml_elem been validated by an XML schema). -- Regards, Mukul Gandhi --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
