On May 29, 2006, at 11:33 AM, Brad Clements wrote:
Or in other words, why can't I use the same code under both plain and
threadlocal?
you can. __del__ and close() are always available in the exact same
way regardless of threadlocal status, both return the connection to
the pool (in the
hi list -
Pursuant to this discussion, I propose we change the close() method,
illustrated below:
result = sometable.select().execute()
result.close()
to this:
result = sometable.select().execute()
result.close_connection()
to eliminate confusion this may
On 29 May 2006 at 13:16, Michael Bayer wrote:
Pursuant to this discussion, I propose we change the close() method,
illustrated below:
result = sometable.select().execute()
result.close()
to this:
result = sometable.select().execute()
result.close_connection()
to eliminate
On 29 May 2006 at 13:09, Michael Bayer wrote:
I want to use resultset.close() in ALL of my code, so that it will
work correctly under either strategy.
go nuts. should all work fine. all the work ive done in this area
was designed specifically towards your request for this
only because close does not do the traditional thing that close
() does on a cursor object, which is simply close the cursor, but
not affecting the connection which the cursor is assocaited with.
this particular close() method insetad returns a connection to a
connection pool.
but it
On Sat, 27 May 2006 20:36:56 -0400, Michael Bayer [EMAIL PROTECTED] wrote:
sure, right here:
http://www.sqlalchemy.org/docs/dbengine.myt#dbengine_connections_context
just switch on 'create_engine('foo', strategy=threadlocal)' and youve got
identical 0.1 behavior.
Not really since I tried
creating the postgres engine with the separate QueuePool, like youre
doing, is a good idea to get a better result here. the
threadlocal strategy uses this silly TLEngine subclass which is
only trying to do that old engine.begin()/engine.commit() thing,
which id rather get rid of, as
I have committed a new version of TLEngine which handles transaction
context in the same way as that of SQLEngine in version 0.1, and
added new unit tests that test the threadlocal and nesting behavior
of this engine. the commit()/rollback() methods directly off the
engine were not
On Sun, 28 May 2006 12:19:04 -0400, Michael Bayer [EMAIL PROTECTED] wrote:
creating the postgres engine with the separate QueuePool, like youre doing,
is a good idea to get a better result here. the threadlocal strategy
uses this silly TLEngine subclass which is only trying to do that
On May 28, 2006, at 2:36 PM, Valentino Volonghi aka Dialtone wrote:
The new way of specifiying the dabatase parameters that uses an url
is worse than the previous one IMHO since it doesn't allow me to
avoid specifying an host and a port and I have to rely on passing
my own QueuePool()
On Sun, 28 May 2006 14:51:55 -0400, Michael Bayer [EMAIL PROTECTED] wrote:
On May 28, 2006, at 2:36 PM, Valentino Volonghi aka Dialtone wrote:
The new way of specifiying the dabatase parameters that uses an url is
worse than the previous one IMHO since it doesn't allow me to avoid
this is unrelated to the URL scheme. i committed a one liner in 1539
that does not include the blank parameters in the keyword argument
list if not found in the URL.
to see why PG was doing this, try this:
psycopg2.connect(database='blabla2', user='blalbla1',
password='blabla', host='')
On Sun, 28 May 2006 15:10:37 -0400, Michael Bayer [EMAIL PROTECTED] wrote:
if you are having deadlocks within one thread, that is almost certainly
because you are using two simultaneous connections from the database, one
of which has a lock on a table (due to a transaction) which the other
heres the bug fix:
--- lib/sqlalchemy/pool.py (revision 1540)
+++ lib/sqlalchemy/pool.py (revision 1541)
@@ -71,7 +71,7 @@
class Pool(object):
def __init__(self, echo = False, use_threadlocal = True,
logger=None, **kwargs):
-self._threadconns = {}
Michael Bayer wrote:
why are you using twisted *with* threads ? didnt we all agree that was
sort of unnecessary ?
Twisted operates without needing to run *in* a thread by doing
everything asynchronously. Every call made via the Twisted event loop
must either return (synchronous) results very
15 matches
Mail list logo
