On Dec 18, 2006, at 6:21 PM, Marek Went wrote:
> I suppose national characters should be supported (utf-8), but I have
> problem with this. Is it possible to work with utf-8 wide character
> set?
Yes, it certainly should be, but we keep finding places where the
code falls short of full unicode compliance. Guess that's what
happens when you get two Americans writing it! ;-)
This is one of those cases where the exception handling we built in
works against us. It's needed because it has to rollback any failed
transactions, but it also masks the actual location of the code that
is causing the problem.
If I were to give you modified copies of dBizobj.py and
dCursorMixin.py for testing, would you have a test database that you
could work with - one that you don't care if transactions get messed
up, and can delete afterwards? I don't want you to test on important
data, since removing these exception handlers can possibly leave the
database in a corrupt state.
-- Ed Leafe
-- http://leafe.com
-- http://dabodev.com
_______________________________________________
Post Messages to: [email protected]
Subscription Maintenance: http://leafe.com/mailman/listinfo/dabo-users
