Hello.
I am just starting with SQLAlchemy and have a small problem: my "get" query
is compiled every time I use it.
According to cProfile:
ncalls tottime percall cumtime percall filename:lineno(function)
100.0000.0000.0020.000
/usr/local/lib/python2.7/dist-packages/SQLAlchem
the Customer.comments attribute proxies through "comment_association" which is
configured as a scalar.this should probably be improved in this particular
recipe, but for starting assignment you need to assign a collection first:
c1 = Customer()
c1.comments = []
I don't ever use "discrimina
if you're doing a query that is causing the database connection to simply fail,
then you'd need to address that issue primarily. you'd probably want to
perform multiple insert statements, chunking about 5-10K records at at time.
On Apr 2, 2013, at 1:26 PM, algotr8...@gmail.com wrote:
> To cl
I was attempting to implement the example shown in
https://github.com/ContextLogic/sqlalchemy/blob/master/examples/generic_associations/discriminator_on_association.py
trace: https://gist.github.com/brukhabtu/2294f0873830243845db
my code: https://gist.github.com/brukhabtu/1423c9825252ddcf85a5
Cu
To clarify my environment. I have a VM (Linux Redhat) which has MySQL
server running on it. My script runs locally on the same machine. It is a
simple script that is doing a insert many after processing/parsing a csv
file. I don't have any web apps or anything of that nature.
On Monday, April
I tried to include pool_recycle = 10 (seconds) in my create_engine call but
that doesn't fix the problem. I still get the same error. Hmm
On Monday, April 1, 2013 7:04:48 PM UTC-7, algot...@gmail.com wrote:
>
> I am using the sqlalchemy package in python. I have an operation that
> takes som
Below is the traceback. When it attempts to perform the conn.execute(),
which is the db insert is where it seems to realize the connection is
stale.
Traceback (most recent call last):
File "myscript.py", line 126, in
main()
File "myscript.py", line 33, in main
insert_data(final_dat
Mike, thanks for the clarification. I never noticed that particular
implication of the default cascade!
On Mon, Apr 1, 2013 at 1:23 PM, Michael Bayer wrote:
>
> On Apr 1, 2013, at 2:10 PM, Michael Merickel wrote:
>
> > I ran into a situation the other day where I would create a new object
> but
Wonderful! Thanks a lot.
2013/4/2 Michael Bayer
> The insert() construct supports a call called returning() to emit whatever
> RETURNING you want, but if you're using the ORM, then the insert()
> construct is generated by your mappings. In this case, the ORM right now
> favors being able to ba
