On Thu, Sep 3, 2009 at 9:46 AM, Christophe de VIENNE<[email protected]> wrote:
> Hi Florent,
>
> 2009/9/3 Florent Aide <[email protected]>
>>
>> Hi Guys,
>>
>> I have worked on the 1.1 release and I have all tests working in TG
>> itself and in the QS template (the default one with SQLAlchemy, with
>> and without identity)!
>
>
> Well done !
>
>>
>> In the process I removed all deprecation warnings from SQLAlchemy
>> 0.5.5 about the session_mapper.
>>
>> At the moment I have just replaced the session mapper with the
>> sqlalchemy.orm.mapper. BUT this means that user code that relied on
>> the session_mapper behaviour (instanciating the object with arguments,
>> auto saving, query method on the class) will break.
>>
>> We could try to mimic the session_mapper behavior (it is documented in
>> the SQLAlchemy documentation) but I would rather avoid this cruft and
>> explain how to migrate from the old SQLAlchemy way to the new one...
>>
>> What do others out there think?
>
> Can we make the old behavior optional ? It would require an explicit
> activation, so that the developper knows that he is relying on a deprecated
> behavior.
>
I kind of agree with this. Do we really need to mimic it? It's
deprecated so an explanation that you should migrate seems best to me.

> My 2 cents,
>
> Christophe
>
> >
>

--~--~---------~--~----~------------~-------~--~----~
You received this message because you are subscribed to the Google Groups 
"TurboGears Trunk" group.
To post to this group, send email to [email protected]
To unsubscribe from this group, send email to 
[email protected]
For more options, visit this group at 
http://groups.google.com/group/turbogears-trunk?hl=en
-~----------~----~----~----~------~----~------~--~---

Reply via email to