[sqlalchemy] Re: migrating from 0.4 to 0.5

2009-04-18 Thread dykang
Thanks for your response. Would it be unreasonable to request a global override for this? Perhaps a sessionmaker option? It feels like it is not totally unreasonable, and it helps me prevent a monkey patch or having to type "subtransactions=True" every time I call begin. Obviously not a huge prob

[sqlalchemy] Re: migrating from 0.4 to 0.5

2009-04-18 Thread Michael Bayer
On Apr 17, 2009, at 7:36 PM, dykang wrote: > > Hi, > > I was trying to migrate from 0.4 to 0.5, and I noticed a behavior in > 0.5 that wasn't listed in the migration document that I'd like to > disable. It appears that in 0.5, begin no longer allows for > subtransactions by default, and that the

[sqlalchemy] Re: moving an object

2009-04-18 Thread Michael Bayer
if we had some help. the next time I have time to work on SQLAlchemy the 0.6 release is my top priority. this particular issue is a behavioral quirk with multiple, straightforward workarounds, and it also not entirely clear if continuing event chains in the manner I'm proposing is even

[sqlalchemy] Eagerloading object/model values

2009-04-18 Thread hticker
I'm quite green with sqlalchemy so please excuse me if this question is very basic. I'm having a problem reading property values of an object after a query. First let me describe the layout of my application. I have a User class which is mapped into a database. The user class is defined as such

[sqlalchemy] Re: moving an object

2009-04-18 Thread jean-philippe dutreve
Hi, any chance to have a fix for this? On 6 avr, 17:16, "Michael Bayer" wrote: > OK in fact this can possibly be implemented if the "initiator" passed > during attribute mutation operations consisted of not just an > AttributeImpl but also a target instance, so that append/remove/set > operation