in sanjay's example, its inserting an integer 1 or 0 value, not True.
I think that's the problem. He's saying:
What happens is, while fetching, boolean columns are fetched as integer
type (0, 1).
He's trying to set the show_ads column from the value of another column
(directly pulled from
My system was having PostgreSQL 8.1.4 and psycopg. After Alan's code
failing in my system, I doubted on psycopg. With some struggle, I
successfully installed psycopg2 and removed psycopg, and things worked!
Thanks a lot for the diagnosis, guys. Extremely sorry to have bothered
you.
My system was having PostgreSQL 8.1.4 and psycopg. After Alan's code
failing in my system, I doubted on psycopg. With some struggle, I
successfully installed psycopg2 and removed psycopg, and things worked!
You'll find really comfortable with psycopg2. I, too, used psycopg1 till
August 2006
In order to know whether a column is an autoincrement column, I am
checking the above flag. But it is returning True always, irrespective
of the column. I am using SQLAlchemy 0.3.3, PostgreSQL 8.1.4 and
psycopg2 on fedora 5.
Wondering whether it's the correct flag, anything is wrong in my
Sanjay wrote:
In order to know whether a column is an autoincrement column, I am
checking the above flag. But it is returning True always, irrespective
of the column. I am using SQLAlchemy 0.3.3, PostgreSQL 8.1.4 and
psycopg2 on fedora 5.
Here is a similar post that was replied:
Hello, I'm experiencing what I think is an unexpected behaviour using
SessionTransaction in SA 0.3.3. The following example is a slightly modified
one from SA's SessionTransaction docs.
#begin sa_example.py
sess = create_session()
trans = sess.create_transaction()
item1 =
I have the following three tables and trying to do a join. See below for
details:
g_main (
gene_id serial primary key,
.
);
g_refseq (
refseq varchar references refseq(refseq),
gene_id int references g_main(gene_id)
);
refseq (
refseq varchar primary key
);
from
Any suggestions on what I am missing?
I don't use sqlsoup and I don't know whether you not using it correctly;
btw, if you're in hurry, as a simple workaround, I suggest you add this line
to sqlsoup.py, line 369, just above klass_query =... :
klass._mapper.compile()
and see if it works.
we dont rollback the attributes of objects nor do we automatically
expunge() them. its impossible for SA to rollback in-memory objects
without great danger of stepping on other program state that the user
doesn't want affected.the rollback that is performed by
SessionTransaction is a
On Friday 22 December 2006 10:10, Alan Franzoni wrote:
Any suggestions on what I am missing?
I don't use sqlsoup and I don't know whether you not using it correctly;
btw, if you're in hurry, as a simple workaround, I suggest you add this
line to sqlsoup.py, line 369, just above klass_query
Michael Bayer [EMAIL PROTECTED] writes:
func.current_timestamp() (which evaulates without the parenthesis in
SQL)
Excellent, works like a charm. Thanks.
--~--~-~--~~~---~--~~
You received this message because you are subscribed to the Google Groups
FWIW, that works in MS-SQL and Postgres as well. One of the too-few
functions that works cross-DBengine.
On 12/22/06, Chris Shenton [EMAIL PROTECTED] wrote:
Michael Bayer [EMAIL PROTECTED] writes:
func.current_timestamp() (which evaulates without the parenthesis in
SQL)
Excellent, works
when i put PKs on the table, it worked fine. so that you get the error
is disturbing to me since I have tried to insure that there are no
paths into a Mapper that wont compile it first. It makes sense that
you might have problems I dont, since im testing on OSX whereas you are
on linux, and
I am not reflecting the tables. Example code:
from sqlalchemy import *
metadata = BoundMetaData('postgres://user:[EMAIL PROTECTED]/mydb',
echo=True)
company_tbl = Table('company', metadata,
Column('company_id', Integer, primary_key=True,
autoincrement=True),
Column('name',
14 matches
Mail list logo
