On Friday, Nov 22, 2002, at 17:30 Europe/London, Tod Harter wrote:

I never really honestly thought much of the SAX approach. In THEORY its nice,
but once I sat down and considered the logic required to do it, it got scary
fast! It would have to be a pretty elaborate stack-based state-machine.
XSP.pm is simpler than you think. The code's not that hard, people are just afraid of it because it has to maintain a bit of state in the stack, and has to special-case a few tags to do the right thing.

Turning XSP.pm into an XSLT engine would be a backward step here. It's much harder to do some of the heuristic stuff we do in XSP.pm in XSLT. Stick with the mantra "if it ain't broke...". Yes, there's a small bug in namespace output, but I'm willing to bet it's easy to fix.

Matt.


---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]



Reply via email to