Hi Willie, Art, Richard:
Currently, the most valuable thing we can probably accomplish in the
short term would be to document suggested changes and append them to Art
Hundiak's proposal addressing other areas of improvement to Zend_Log.
If we can at least agree upon these changes, get the proposal approved,
and push through these smaller API improvements in time for the 0.2
release, then we prevent solidification around the current Zend_Log API
(and its associated weaknesses that some members have previously described).
The current Zend_Log offers a small subset of the capabilities found in
log4j:
http://logging.apache.org/log4j/docs/documentation.html
http://framework.zend.com/manual/en/zend.log.html
However, given timelines, availability, etc., I'm inclined to think that
Art's proposal represents our most likely path to successfully improving
ZF's logging facilities in time for ZF 1.0. If we all focus on making
specific changes and improvements to the existing flyweight Zend_Log by
helping with Art's proposal, we can bend Zend_Log into something more
flexible and closer to "standards" (e.g. log4j) without actually doing
any "wave-making" changes. As adoption of ZF increases, some will want
a full implementation of log4j, but given Andi's direction on getting
closure with ZF 1.0, perhaps we can think about a "full" implementation
after shipping 1.0.
Cheers,
Gavin
