At 21.01 11/02/2005 +0000, Gareth Reakes wrote:
On 11 Feb 2005, at 20:11, James Berry wrote:
Anything else bold that needs to be done? Integral xpath support anyone?
I would like to see this eventually. Maybe we can bring Pathan in. With a full XPath engine we can borrow some ideas from the .net people. Bringing Pathan in will be a lot of work, so I suggest we go with this plan first and then see whats happening.
We need to keep in mind that Pathan is heavily based on STL; even if we tried to use the common subset of Visual Studio 6 and gcc, there are still platforms were this subset is not supported.
When Neil, Alby, Lucien and I gave a talk at XML 2004 we had some interesting feedback about where xerces should go. When we have the basics done we should revisit this and see what makes most sense for users.
Right. We should not plan to have 3.0 as the next immediate release, or we will risk to have to ship a 4.0 afterwards, if we forgot about something "big".
So branching 3.0 while keeping doing bug fixes on 2.x until 3.0 is feature ready seems reasonable.
Things that in the past someone asked for:
- XInclude support;
- a build system that could exclude parts of the code to make smaller builds, or at least a "plain parser" version of Xerces (no SAX/schema/PSVI/DTD validation, just WFXMLScanner and DOM)
- support for the ACE "platform" (that person also stated that ACE had the interest in adopting Xerces as their parser instead of maintaining the current, home-grown one)
- a more complete test suite
Alberto
--------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
