Hi all. Possibly a real basic question but one I seem to have gone
round in circles with and failed to find an answer for so far.
I have an (Oracle) db with some tables similar to this. I'll create a
simple in-memory sqlite db just to illustrate.
from sqlalchemy import *
db =
Hello,
Michael Bayer said the following on 09.08.2007 22:53:
OK just to double check, the syntax looks like:
SELECT * FROM sometable ORDER BY foo NULLS FIRST
SELECT * FROM sometable ORDER BY foo DESC NULLS LAST
Yes.
? i.e. is DESC/ASC before the NULLS part ? or doesn't matter ?
It
OK ive added ticket #723 for this, it will go into 0.4xx.
On Aug 10, 2007, at 5:13 AM, Oleg Deribas wrote:
Hello,
Michael Bayer said the following on 09.08.2007 22:53:
OK just to double check, the syntax looks like:
SELECT * FROM sometable ORDER BY foo NULLS FIRST
SELECT * FROM
On Aug 10, 2007, at 5:18 AM, Alexandre CONRAD wrote:
It sounds to me that the session is global and needs to be cleared
everytime. Which I think is wrong (or SessionContext works
differently).
I think a *new* session should be created and attached to every new
request. The session is
On Aug 10, 2007, at 3:18 AM, Andy Hird wrote:
because it's trying to execute the sql: UPDATE account_stuff SET
credit=?
because I assume account_stuff has no primary key (updates to
account_ids specify a where clause).
How can I get the above to work? I assume the join in the mapper
Michael Bayer wrote:
SQLAlchemy has always supported this feature using the
version_id_col keyword argument to mapper(), which is described at:
http://www.sqlalchemy.org/docs/04/
sqlalchemy_orm.html#docstrings_sqlalchemy.orm_modfunc_mapper
Argh, just a little bit too far in the
Er, where is it you're not supposed to use .c? The code in MikeB's
example seems to have .c in every possible location. How do you
access a column without .c?
On 8/10/07, Jeronimo [EMAIL PROTECTED] wrote:
Excelent ! It works perfectly !!
Thank you very much Michael. I was going crazy trying
On Aug 10, 2007, at 8:32 AM, Grzegorz Adam Hankiewicz wrote:
Is there any documentation about this or do sqlalchemy applications
deal
with concurrent writes in another way? If the unit of work tracked
those
internal version numbers the flushing would fail and an exception
raised, but
On Aug 10, 1:04 pm, Grzegorz Adam Hankiewicz
[EMAIL PROTECTED] wrote:
Cool. In the unit test you provide the version_id_col parameter is
specified manually. Does SQLAlchemy do automatic reflection on the table
and use a version_id column if it already exists, without having to
specify it on
On 8/10/07, Alexandre CONRAD [EMAIL PROTECTED] wrote:
Okay Mike,
about scoped_session(), I think I got it right this time. I was trying
to get inspired from SAContext, how it's beeing wrapped around with
Pylons. Since SAContext has not yet upgraded to SA 0.4, maybe I was just
getting
On Fri, Aug 10, 2007 at 02:51:51PM +0200, Hermann Himmelbauer wrote:
My web-application, that is in front of SQLAlchemy has various input-fields,
which should act as a where-clause (with like()), a query looks e.g. like
this:
Excelent ! It works perfectly !!
Thank you very much Michael. I was going crazy trying to figure how to
move
subquery to the from clause.
On Aug 9, 8:06 pm, Michael Bayer [EMAIL PROTECTED] wrote:
OK sorry, i didn't look carefully enough. when you use a scalar
subquery, you shouldn't access
On Thursday 09 August 2007 13:04:44 Paul Johnston wrote:
Hi,
A little update;
Also, in the same direction, complete copy of some database seems to
consist of (at least) 3 stages:
1 recreate/remove the old one if it exists
2 copy structure
3 copy data
3 is your copy loop, which
Hi,
wouldn't this be working equivalent? it also copes with empty tables..
Yep, I like yours better. Thanks!
i'm Wondering if all the unicode strings (at least table/column names)
should be converted back into plain strings as they have been before
autoload reflecting them from database.
Okay Mike,
about scoped_session(), I think I got it right this time. I was trying
to get inspired from SAContext, how it's beeing wrapped around with
Pylons. Since SAContext has not yet upgraded to SA 0.4, maybe I was just
getting inspired from some different mechanism. Or SAContext it doing
On Thursday 09 August 2007 13:04:44 Paul Johnston wrote:
Hi,
A little update; this code handles the case where columns have a
key attribute:
model = __import__(sys.argv[1])
if sys.argv[2] == 'copy':
seng = create_engine(sys.argv[3])
deng = create_engine(sys.argv[4])
for tbl
Hi,
1 recreate/remove the old one if it exists
how about 1? it also does depend badly on db-dialect.
Personally, I'd do this step manually. Not sure I quite trust a script
that has the potential to drop database
btw: why is the 'text_as_varchar=1' considered only if it is in url
(see
Hi.
I've been reading the documentation on the unit of work at
http://www.sqlalchemy.org/docs/04/sqlalchemy_orm_session.html and I
wonder how this can be merged with what I've seen called as optimistic
locking.
By optimistic locking I understand the possibility of tracking if two
users
Hi,
My web-application, that is in front of SQLAlchemy has various input-fields,
which should act as a where-clause (with like()), a query looks e.g. like
this:
session.query(MyObj).filter(table1.c.input1.like(inputfield_1+'%'),
I haven't tried it with your specific example, but I'm pretty sure the
theory should work just the same.
The first thing you need to do is create a list of the clause
objects. The other thing you'll need is a way to associate the input
names with the column names (either make them the same or
On Aug 10, 1:56 pm, Mike Orr [EMAIL PROTECTED] wrote:
Er, where is it you're not supposed to use .c? The code in MikeB's
example seems to have .c in every possible location. How do you
access a column without .c?
ive clarified this in 0.4's behavior.
lets take a query like this:
btw: why is the 'text_as_varchar=1' considered only if it is in
url (see mssql.py create_connect_args()) and not if it is in the
connect_args argument to create_engine()?
Fair question, and the short answer is because that's all I needed.
We did have a discussion about unifying
i'm Wondering if all the unicode strings (at least table/column
names) should be converted back into plain strings as they have
been before autoload reflecting them from database.
Well, some databases do support unicode identifier names, some
don't. I'd say don't do any conversion for
23 matches
Mail list logo
