This is strange. I had problems reproducing the bug for a long time
and was now quite happy that I succeeded.
Yes, I'm using PostgreSQL 8.1.5 and psycopg2. And I still see the
behavior that I reported. That should narrow the problem down to
something in my local setup... I'm sorry for wasting
On Nov 7, 2007, at 11:05 AM, jason kirtland wrote:
Michael Bayer wrote:
On Nov 7, 2007, at 6:01 AM, klaus wrote:
This is strange. I had problems reproducing the bug for a long time
and was now quite happy that I succeeded.
Yes, I'm using PostgreSQL 8.1.5 and psycopg2. And I still see
By the way, on a larger database, drop_all() runs into an endless
loop. However, I have no small script to reproduce this yet.
Klaus
On 7 Nov., 16:39, klaus [EMAIL PROTECTED] wrote:
That's very kind.
I'm on the trunk (0.4.1dev_r3747) and on linux. And I think your
script (with a changed
That's very kind.
I'm on the trunk (0.4.1dev_r3747) and on linux. And I think your
script (with a changed connection string, nothing else) fails on my
machine. Here's the complete output:
2007-11-07 16:33:14,848 INFO sqlalchemy.engine.base.Engine.0x..b4
select relname from pg_class c join
On Nov 7, 2007, at 1:09 PM, jason kirtland wrote:
Michael Bayer wrote:
what happens if you change the create_engine to:
engine = create_engine('sqlite://', echo=True, paramstyle=named)
that doesnt reproduce for me on linux. i find it hard to believe its
a bug within the PG dialect
Michael Bayer wrote:
On Nov 7, 2007, at 6:01 AM, klaus wrote:
This is strange. I had problems reproducing the bug for a long time
and was now quite happy that I succeeded.
Yes, I'm using PostgreSQL 8.1.5 and psycopg2. And I still see the
behavior that I reported. That should narrow the
On Nov 7, 2007, at 6:01 AM, klaus wrote:
This is strange. I had problems reproducing the bug for a long time
and was now quite happy that I succeeded.
Yes, I'm using PostgreSQL 8.1.5 and psycopg2. And I still see the
behavior that I reported. That should narrow the problem down to
Michael Bayer wrote:
what happens if you change the create_engine to:
engine = create_engine('sqlite://', echo=True, paramstyle=named)
that doesnt reproduce for me on linux. i find it hard to believe its
a bug within the PG dialect itselfbuilding PG on my linux box now.
No
I cant reproduce this one. I see you have named bind params so I
tried with postgres. it also works with sqlite. works with release
0.4.0 as well as the trunk.output is (with echoing):
SELECT testView.id AS testView_id, testView.data AS
testView_data
FROM (SELECT test.id AS id,
