On Friday, February 14, 2003, at 01:38 PM, Jakob Praher wrote: <snip>
i've taken a look again at the specs and had a think - and i now agree with you.as for the first, i think i need a little more time to think about this. my reading of the SAX specification indicates that the current behaviour is correct (since betwixt doesn't really support namespaces). on the other hand, i think that maybe applying this patch is the right thing to do since a lot of application rely on local name being set.I think it depends. I have looked again on the SAX website and I come to the following conclusion: [http://www.saxproject.org/?selected=namespaces] the "" for localName should only be used if the contentHandler has disabled namespace support. ie. "http://xml.org/sax/features/namespaces" I think this can't be said here, since it applies to the Parser configuration, which we have no control when given just a ContentHandler interface to write to. and as in the second patch, when you embed your fragment in an existing tree you can't tell whether the contentHandler requires namespace support. so I would heavily +1 for using the qName==localName, since you are always on the safe side: if the consumer is not capable of namespaces, it doesn't matter since the consumer will only use the qName. on the other hand if the consumer is capable of handling namespaces it works too.
but i don't want to apply your patch since it fixes the symptom rather than the cause. to solve the problem properly is going to require some refactoring which may take some time. i'd appreciate it if you could stick around and give it a test once it's finished.
- robert
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
