Ok. I'll do some testing against other engines when I get a chance.
Thanks for helping.
Cheers
Chris Miles
On Feb 6, 2:36 pm, Michael Bayer wrote:
> sqlite doesn't include CREATE TABLE statements within the scope of a
> transaction. I think that's a relatively rare behavior only seen in
>
sqlite doesn't include CREATE TABLE statements within the scope of a
transaction. I think that's a relatively rare behavior only seen in
Postgres, in fact - I dont think Oracle or MySQL have that behavior,
for example.
On Feb 5, 2009, at 10:01 PM, Chris Miles wrote:
>
> That did the trick
That did the trick, thanks.
Well, actually, it did the trick for PostgreSQL but sqlite isn't
rolling back. The SA logs show the same commands are being sent to
both. Here's an example:
$ rm test1.sqlite
$ python sa_create_table_transaction_test.py
2009-02-06 13:39:29,006 INFO sqlalchemy.engine
that's your flush() process flushing something pending in the
session. say session.flush() to see it happen. the error means
you've removed a child object from a parent, which would result in a
primary key that is also a foreign key being nulled out.
On Feb 5, 2009, at 8:51 PM, Gloria W
OK, a new problem on the same model:
I try this in my unit test:
memberProfile = self.session.query(MemberProfile).filter
(MemberProfile.memberID.in_(memberid)).order_by
(MemberProfile.memberID).filter(MemberProfile.city == 'Jamaica').all()
and I get this error:
Traceback (most recent call la
Wow, awesome, it works, thank you!
~G~
--~--~-~--~~~---~--~~
You received this message because you are subscribed to the Google Groups
"sqlalchemy" group.
To post to this group, send email to sqlalchemy@googlegroups.com
To unsubscribe from this group, send email to
> I was wondering if anyone was aware of a JDBC DBAPI module for
> cpython.
Interesting idea, and could be a killer feature for SA 0.6+ if it could be
made to work
Jpype could perhaps do the job:
http://jpype.sourceforge.net/
There's been at least some activity with accessing JDBC drivers from
you're ordering the Member and Gender relation()s by a column in the
parent table, which is producing the error.The order_by expression
should be local to the Member or Gender entity.
On Feb 5, 2009, at 11:54 AM, Gloria W wrote:
>
> Hi All,
> I have three classes, all using the same decl
Hi All,
I have three classes, all using the same declarative_base() instance,
as follows:
In a config file:
global Base
Base = None
def initBase():
global Base
if not Base:
Base = declarative_base()
return Base
~~~
Hi All,
I have three classes, all using the same declarative_base() instance,
as follows:
In a config file:
global Base
Base = None
def initBase():
global Base
if not Base:
Base = declarative_base()
return Base
~~~
After some experimenting I am able to generate the correct query
by two methods
(1) build query from ORM classes with ORM session.query()
(2) build query from underlying tables with sql expressions
I like the ORM based method better, because the code does not need
to know which columns ar
Hi guys, I managed to model a device scanner based on SA 0.5.2.
Everything works fine on the first run, but when i rerun a scan to
refresh data (specially Client.scantime) i do not see any UPDATE
statement called on clients table.
What's wrong with my design?
Thanks in advance
Massimo
Follow m
polymorphic_identity is intended to link to the "class" of an entity
in a one-to-one fashion, so using a date type for this column would
not be an appropriate usage.We have eventual plans to support
polymorphic_identity supplied by a function but that feature is
currently not implement
Hi,
So I (think I) understand that that polymorphic_on and
polymorphic_identity can be used to determine which class is
instantiated for a query result row. Is there any way I can use a range
of values for polymorphic_identity to map to the same class ? Of
specific interest to me right now ar
On Feb 5, 2009, at 9:04 AM, pwaern wrote:
> What I would like to be able to is to keep on working with detached
> objects like the user object in the code above , in a manner where the
> objects attribute values are the same as they were when the session
> was closed, i.e. without further databa
create() and create_all() take a "bind" argument which can be an
engine or connection. you want the connection in this case.
On Feb 5, 2009, at 5:27 AM, Chris Miles wrote:
>
> I notice that a table create (and drop/etc) is always followed by an
> implicit commit. Is it possible to suppress t
I have also experienced problems with this exception.Here is a simple
example:
import sqlalchemy
import sqlalchemy.orm as orm
engine = sqlalchemy.create_engine('sqlite:///:memory:', echo=True)
metadata = sqlalchemy.MetaData()
users_table = sqlalchemy.Table('users', metadata,
I notice that a table create (and drop/etc) is always followed by an
implicit commit. Is it possible to suppress the commit or force SA to
create multiple tables in one transaction so that if any fail they can
all be rolled back?
Here's some code to demonstrate what I want. In this example, the
18 matches
Mail list logo
