[ZODB-Dev] another reason to stop supporting versions
Dieter Maurer
dieter at handshake.de
Wed Apr 25 17:01:52 EDT 2007
Jim Fulton wrote at 2007-4-24 17:01 -0400:
>I'm 99.9% sure that version commit and abort are broken in ZODB.DB.
>The commit methods in CommitVersion, and AbortVersion (and
>TransactionalUndo) call invalidate on the databse too soon -- before
>the transaction has committed. This can have a number of bad
>effects, including causing inconsistent data in connections.
>
>An argument for keeping version in the past was that they "worked".
>Well, I think they don't work and I'm not interested in writing the
>test to fix them. Is anyone else?
The last time I used "Version"s was about 2 years ago to
change the indexes of a catalog without a downtime.
"Version"s was a great help then.
With "ManagableIndex", I could achieve this today without
"Version"s. But not all indexes have yet the necessary possibility
to determine the object values independent from the index id.
Thus, I might still miss "Version".
On this other hand, I currently do not have time to provide the
tests and fix the code...
--
Dieter
More information about the ZODB-Dev
mailing list