On Feb 25, 1:10 pm, Don Dwiggins <d....@dondwiggins.net> wrote:
Rick Morrison wrote:
Ugg, I am not a big table reflection fan....:-(
I am in the middle of readying a presentation, so unless someone else
wants to jump on this, I'll take a look at this later on this week,
along with integrating a reflection speed-up patch I remember from a
while back.
Thanks. I'm willing to use something other than reflection, if there's
a good way I can avoid having to duplicate my schema (or significant
chunks of it) in SA declarations. I'd like to stick to the DRY
principle if at all possible. (I'm dealing with a legacy DB, and can't
really take SA declarations as "ground truth".)
The approach I have taken is to to dump the legacy database
definitions into python module using autocode.
http://www.sqlalchemy.org/trac/wiki/UsageRecipes/AutoCode.
(I had to tweak the recipe slightly to work with Oracle)
The resulting code needed clean up but still it was a win - no more
waiting for the reflection to finish. Now if DB schema changes I have
to update my module but I do not mind, the changes are rare.
It is a legacy database after all :-).
Waldemar
10 messages in com.googlegroups.sqlalchemy[sqlalchemy] Re: Avoiding reflection ...
.patch