RE: conflict resolution for PersistentList

[Thomas Guettler]
>>> Try to use btrees. BTrees have conflict resolution
>>> and not the whole "list" needs to be stored if it gets
>>> changed.

[Tim Peters]
>> Sounds like good advice to me, but it was given before without
>> visible consequence so far.

[Diez Roggisch]
> It was?

Yes, Dieter suggested it last Thursday:

    http://mail.zope.org/pipermail/zodb-dev/2004-May/007372.html

> In my first post I asked:
>
> <cite>
> first of all I wonder that this is not detected and a ConflictError
> is raised.  And second, I'd like to know what the comonly used idiom
> for such a case is - are PersistenMappings with keys generated from a
> synchronized sequence better suited, or whatelse can I do?
> </cite>

I suspect people are stumbling over what "such a case" means, exactly.  I
really didn't understand what you were trying to accomplish, so I'm probably
projecting that on to everyone else too.

> The hidden conflict was my faulty exception handler - thats fixed
> already.

But you still shouldn't have seen *inconsistent* data.  That's another
mystery here.

> Maybe I missed it, but that question about a different idiom hasn't
> been answered so far by suggesting to use a *BTree - except from
> Thomas today, and as I said before, today I'm not at work.

Most people here get more done when they're not at work <0.5 wink>.

> But I'm happy to try a Tree of course, after all I'm interested in
> a solution - not in using persistent lists at all costs :)

I can't guess, because (as above) I'm not clear on what the problem/goal is.
If BTrees don't seem like the right answer to you when you get back to this,
perhaps you could explain more about what the problem is.


_______________________________________________
For more information about ZODB, see the ZODB Wiki:
http://www.zope.org/Wikis/ZODB/

ZODB-Dev mailing list  -  ZODB-Dev@xxxxxxxx
http://mail.zope.org/mailman/listinfo/zodb-dev