[Zope-Coders] Re: [ZC] 182/ 2 Comment "Zope/CMF inconsistencies: secondary exceptions"

[Dieter Maurer]

Collector: Zope Bugs, Features, and Patches ... writes:
 > Issue #182 Update (Comment) "Zope/CMF inconsistencies: secondary exceptions"
 >  Status Pending, Zope/bug medium
 > To followup, visit:
 >   http://collector.zope.org/Zope/182
 > 
 > ==============================================================
 > = Comment - Entry #2 by tseaver on Dec 16, 2002 1:51 pm
 > 
 > Note that, on the CVS head, this code has now moved to
 > Zope.App.startup.py, line 128.
 > 
 > BTW, Dieter, can you give an example of what kind of thing 't'
 > actually is when you hit this problem?  Strings have already
 > been excluded, so I'm guessing that it might be an ExtensionClass
 > instance?
 > ________________________________________
 > = Request - Entry #1 by d.maurer on Jan 28, 2002 5:03 am
 > 
 > 
 > Uploaded:  "exc_subclass.pat"
 >  - http://collector.zope.org/Zope/182/exc_subclass.pat/view
 > Any exception in the CMF action box seems to
 > trigger a secondary exception in "Zope.__init__"
 > "TypeError: issubclass first argument must be a class".
The problem is no longer reproducible -- at least not with
exceptions raised by action box processing.

I have never tried to understand what type "t" really had.
Python's error message has been sufficiently explicit to
provide an immediate and safe workaround (something that
is not a class cannot be a subclass of "ConflictError").


I would not mind when you rejected the report, as, apparently, the problem
no longer occurs.


Dieter