looking at the current `engine.dispose()` (which I should have done
sooner), that's pretty much what it's doing -- right?
https://bitbucket.org/zzzeek/sqlalchemy/src/55371f4cffa730f65f1b687e9f6287d2ac189227/lib/sqlalchemy/engine/base.py?at=master&fileviewer=file-view-default#base.py-1899:1923
On Mon, Apr 2, 2018 at 4:29 PM, Ruben Di Battista
wrote:
> Hello,
>
> I have some SQLAlchemy-persisted instances of objects that have some
> relationships coming from a parallel execution with multiprocessing. When I
> reduce the results coming from the several processes, I need to merge some
> re
you need a column-based expression there so you should use
literal_column:
http://docs.sqlalchemy.org/en/latest/core/sqlelement.html?highlight=literal_column#sqlalchemy.sql.expression.literal_column
but also that expression won't be usable in every case because you are
correlating it to the name "
we should fix that an add a new method like engine.detach() or
something like that which strictly de-associates the engine from the
pool of the parent process and makes a new one.
right now the way to do that is:
engine.pool = engine.pool.recreate()
gives it a brand new pool and just loses refer
Hello,
I have some SQLAlchemy-persisted instances of objects that have some
relationships coming from a parallel execution with multiprocessing. When I
reduce the results coming from the several processes, I need to merge some
relationships (`satellite` and `ground_station` objects in the code
I'm trying to write a column_property that queries a table that will be
defined after the current table is defined. I tried using raw SQL but I'm
getting an error.
On the "page" table I define a "count_posts" column_property.
count_posts = orm.column_property(text(
'SELECT count(post.id) F
On Monday, April 2, 2018 at 1:40:06 PM UTC-4, Mike Bayer wrote:
>
> now we can probably do better than that but the dispose() case is not
> supposed to be that common or critical, it's usually for when you just
> did a fork(), and you don't want to reference any of the filehandles
> in the par
On Mon, Apr 2, 2018 at 1:36 PM, Jonathan Vanasco wrote:
> clarifying: I had run tests, and the checked-out connections are still open
> after dispose(), and then garbage collected out.
>
now we can probably do better than that but the dispose() case is not
supposed to be that common or critical,
clarifying: I had run tests, and the checked-out connections are still open
after dispose(), and then garbage collected out.
--
SQLAlchemy -
The Python SQL Toolkit and Object Relational Mapper
http://www.sqlalchemy.org/
To post example code, please provide an MCVE: Minimal, Complete, and Veri
On Monday, April 2, 2018 at 1:21:26 PM UTC-4, Mike Bayer wrote:
>
> Connections that are checked out stay open, they
> still refer to that dereferenced pool, and I think at the moment they
> don't get closed until they are garbage collected along with that pool
> object.
>
Thanks. That's wha
On Mon, Apr 2, 2018 at 1:10 PM, Jonathan Vanasco wrote:
> a question came up in another list, and I just wanted to confirm a behavior
>
> it looks like `engine.dispose()` doesn't explicitly `close()` any open
> connections, but they inherently get closed by the method's behavior. is
> that right?
a question came up in another list, and I just wanted to confirm a behavior
it looks like `engine.dispose()` doesn't explicitly `close()` any open
connections, but they inherently get closed by the method's behavior. is
that right?
--
SQLAlchemy -
The Python SQL Toolkit and Object Relational
12 matches
Mail list logo