We just noticed that some objects were not being cataloged correctly.
After a lot of debugging, we noticed the following:
The IntId utility wouldn't find the unique id for an object when a
modified event was sent. As a result, it wasn't being indexed.
We figured out that there was an odd difference between the object that
came into the IntId utility (through a modification event) and the
object that we could take from the folder: the one taken directly from
the folder has a __name__ and __parent__. Even though the memory address
looked the same when using repr(), they were not identical. As a result
the int id couldn't be found.
Finally with some help from Stephan Richter giving us the clue that this
__name__ and __parent__ information could only be lost if LocationProxy
is in play, we figured out what what we think is going on:
What is stored in the ZODB are LocationProxy wrapped objects. The
location proxy wrappers get a unique id reference in the IntId utility,
not the objects themselves. Now when you send an ObjectModificationEvent
from one the document's *own methods*, the event is sent with as payload
to catalog something that is not proxy wrapped. This means that the int
id can never be found for this object, and thus the reindexing cannot
take place properly.
This basically means that LocationProxy wrapped objects cannot reliably
respond correctly to ObjectModificationEvents, and thus won't be
cataloged correctly. If the ObjectModifiedEvent is sent by methods
inside the object itself, things will fail, if the ObjectModifiedEvent
is sent by code elsewhere, they'll likely work.
This is all fixed by subclassing Contained, but the catalog not working
reliably for LocationProxy wrapped objects sounds scary. You could do
something with the IntId utility automatically
un-location-proxy-wrapping the objects if necessary, but that would mean
that what is stored wouldn't know its location anymore, which would also
Zope3-dev mailing list