Florent Guillaume wrote:
Ok after a week the consensus is pretty much in favor of
2. no traceback

1. There's enough votes for BLATHER that I'll add zope.conf option to change the level, but it will default to INFO.

As an idea, make your zope.conf key be of the same type as the "level" parameter in the eventlog section, that way, you should be able to change the code to be:

logger.log(whatever_your_zconfig_option_is,...whatever I used...)

...making it even more flexible at no extra cost. Some people (maybe me ;-) might even like this stuff logged at logging.error so I get a MailingLogger entry _whenever_ a ConflictError is logged, not just when a user sees it. (this is because I have a good idea that ConflictErrors due to silly use of the temp_folder in the largest of the projects I'm currently working on is a major source of poor performance)

Also, in case you don't manage the above, please use logging.debug, not BLATHER, since I'll be eradicating that silly name as part of my work to deprecate zLOG...

2. No traceback will be logged anymore. Note that you'll still have the traceback from point 3.

Indeed, and it's still just as useless there ;-)

3. Here maybe things were unclear. Everybody agrees that the error should happen at ERROR level, but I must point out again that no explicit logging is needed because it is already done if you configure error_log to copy exceptions to the event.log.

Yup :-)

So I propose another little change: have the error_log copy to event.log be the default behaviour. Today the default is off.

+lotz :-)


Simplistix - Content Management, Zope & Python Consulting
           - http://www.simplistix.co.uk
Zope-Dev maillist  -  Zope-Dev@zope.org
**  No cross posts or HTML encoding!  **
(Related lists - http://mail.zope.org/mailman/listinfo/zope-announce
http://mail.zope.org/mailman/listinfo/zope )

Reply via email to