Hi,
I'd like to request a review for the addition of support for the passing of
modules to the DataObjectCache.register_class() method.
The explanation is a little long, but I promise the code changes aren't huge ;)
If a module is found in the parameters to register_class, then I use the
'inspect' module (thanks Drew) to fetch the classes in it, and then I do a
recursive call to the register_class() method with these classes.
I also discovered that in fetching of the classes, I was getting classes that
were imported (e.g. DataObject) which I shouldn't be registering, so I check
that any class to register is actually coming from the module I was asked to
check.
Matt also noticed an issue with the logging where the log messages weren't being
logged where expected, so I've provided a DataObjectBase.get_logger() class
method and DataObjectBase.logger property for easy access, so I'm also including
the fix for this. Keith also mentioned this in my original CR, but looks like I
was misunderstanding how the Python logging worked.
I had to implement the logging using class variables since it's desirable to log
from class methods (can_handle, from_xml, etc.) but also because pickle didn't
like if I tried to pickle anything using an instance variable to store the
reference too.
The webrev can be found at:
http://cr.opensolaris.org/~dkenny/fix.6996768.slim/
I ran the PyUnit tests for slim, to confirm that things are still working and
that I've not broken anything.
Thanks,
Darren.
_______________________________________________
caiman-discuss mailing list
[email protected]
http://mail.opensolaris.org/mailman/listinfo/caiman-discuss
