May be implement the new assertion feature for Xerces-J is another GSoC idea :-)
2010-03-23 ustbcoder 发件人： Michael Glavassevich 发送时间： 2010-03-23 20:51:18 收件人： xalan-j-users@xml.apache.org 抄送： 主题： Re: implement XPath 2.0 function for xalan PsychoPath is an XPath 2.0 processor. It doesn't implement XSLT. Xerces-J has been depending on and using it for quite some time now to support the new assertion [1] feature introduced by XML Schema 1.1. Thanks. [1] http://www.w3.org/TR/xmlschema11-1/#cAssertions Michael Glavassevich XML Parser Development IBM Toronto Lab E-mail: mrgla...@ca.ibm.com E-mail: mrgla...@apache.org xunlong gui <ustbco...@gmail.com> wrote on 03/23/2010 01:00:56 AM: > It seems that PsychoPath actually is another XSLT processor,it is > different with Xalan.But it?implement a subset XPath 2.0/XSLT 2.0 > specification. In fact,i think?the current condition about Xalan is > that many developers help it can implement XSLT2.0 but Xalan > community caught some difficults in implementing XSLT 2.0 for it,may > be because some developers lost reason. So,i think expectation can > do nothing, there must be some guy start this job and i want to do > it. If i have a draft version,i will post my codes to open source > community,and check if Xalan can accept my codes > 2010/3/22 doug <doug.mcn...@ascnet.com> > Yeah, > > I've been lurking for two years behind the scenes on the question of > XPATH/XSLT 2.0 in xalan. Still no one with cycles to get the ball > rolling! But perhaps on the XPATH 2.0 question alone looking at > xalan and PsychoPath (http://wiki.eclipse.org/ > PsychoPathXPathProcessor/UserManual) �might be a place to start > (check the archives)? Some "fail on schema use detected" > implementation might not be that difficult? Or is the "sequence" > question going to make it impossible? Not that I'm volunteering! > > Doug McNeil > > > On 3/12/2010 4:32 PM, Tatu Saloranta wrote: > --- On Fri, 3/12/10, kesh...@us.ibm.com<kesh...@us.ibm.com> �wrote: > > ? > I'd love to see Xalan tackle it -- but > Xalan lost its most active contributors a few years ago due > to time demands > from Real Jobs, and development slowed down as a result. > ?? > ... > ? > So... I'd say that if you like the idea, > you should start drafting versions of this and submit them to Xalan as > proposed changes when you've got a large enough bundle > to be an interesting > ?? > Wasn't there someone who quite recently wanted to fork off Xalan, to > work on XSLT 2.0? It might be good to check out archives, perhaps > group of like-minded pioneers could see if "Xalan next-gen" could > become reality... > (this just an outsider observation: I have not worked on Xalan > codebase at all, but can appreciate sizable effort that would entail...) > > But I would also suggest that "just start coding" is not a bad > approach. With git-based projects, for example, that's sort of > standard approach. > Trying to convince others of value is easier when there is some > action to back it up, beyond just hope& �plans. > > Just my 2c, > > -+ Tatu +- > > > > > ?
