Romain Slootmaekers wrote:
Shane Hathaway wrote:
If we have veto(), it should probably expect a string argument that explains the reason for the veto. Then if something tries to commit, we can raise VetoedError(explanation). Otherwise, it seems like failed transactions would be opaque and hard to decipher.
hm this is starting to look an awful lot like Persistent Java Beans.
maybe you should take a look there and borrow some architecture.
No, it's not really related. JBs let you veto property changes, which is not like vetoing transactions. Besides, we decided we don't need any kind of transaction veto today.


Zope-Dev maillist - [EMAIL PROTECTED]
** No cross posts or HTML encoding! **
(Related lists - )

Reply via email to