Client application doesn't mean web application and SqlAlchemy would
be a too big hammer for SQL console (in my opinion). Well it will be an
information system written with wxPython running on multiple computers.
I don't want users to see information for direct access to the database.
I was
Hi,
I am trying to translate to sqlalchemy my queries having on object
level clauses like the a.b.c.d == some_value where a is instance of
class A, b - instance of class B and...
How such queries can be expressed in sqlalchemy code (supposing that
every class is mapped to its own table)? Are
Michael Bayer ha scritto:
use ForeignKeyConstraint.
Yes, I have used it.
But why ForeignKey does not allow this?
Regards Manlio Perillo
--~--~-~--~~~---~--~~
You received this message because you are subscribed to the Google Groups
sqlalchemy group.
To
As attachement, here's the patch (against rev 2132). It's local to
sqlalchemy/orm/, so:
cd /sqlalchemy/orm
patch -p0 attribute_cache.patch
About the patch itself:
1. the cache should be a WeakKeyDictionary
OK, done. Performances are still OK, differences between built-in dict are
I still get the error:
Traceback (most recent call last):
File U:/ADA_tools/pickle_tables.py, line 27, in ?
pickle.dump(metadata_to_pickle,pfile)
File C:\Python24\lib\pickle.py, line 1382, in dump
Pickler(file, protocol, bin).dump(obj)
File C:\Python24\lib\pickle.py, line 231, in
SA can form joins between tables automatically if the tables express
the proper foreign key relationship between each other, and if there is
no ambiguity in that relationship; i.e. table A and table B have only
one ForeignKeyConstraint (or single ForeignKey) between each other.
if you have table
it was modeled after SQL CREATE TABLE syntax itself, where the
ForeignKey keyword looks like column REFERENCES
sometable(othercolumn) and does not allow any ON UPDATE or ON DELETE
clauses at that level; they would be declared at the bottom of the
table def inside the CONSTRAINT foo FOREIGN KEY...
latest trunk, this test works for me:
from sqlalchemy import *
meta = BoundMetaData('mysql://user:[EMAIL PROTECTED]/dbname')
table = Table('foo', meta, autoload=True)
meta2 = MetaData()
t2 = table.tometadata(meta2)
import pickle
pickle.dumps(meta2)
post your test case (with some sample
the test object you are loading in the update function is local to
the update function itself. when the function completes, test
falls out of scope and is removed from the session (since it is weakly
referenced in the session).
a change needs to be made to session whereby dirty objects dont get
Hello,
I would like to query a big table and display the information. But at
the top of the list I'm going to display, I need a little counter that
tell how many items where returned.
What would be the most efficient way to count the results ?
client_list = session.query(Client).select()
I'm trying to use assign_mapper with polymorphic multiple table
inheritance, and running into problems with the primary key sequencing.
Is this a supported use of sqlalchemy? I'm running version 0.3.1.
Sample code below. If use_assign_mapper is false, then the script
works with both postgres
Michael Bayer ha scritto:
the test object you are loading in the update function is local to
the update function itself. when the function completes, test
falls out of scope and is removed from the session (since it is weakly
referenced in the session).
a change needs to be made to
if you are going to fetch all results in all cases, then len(result) is
fine. if the result list is enormous and youre only going to fetch a
small portion of it within a request, the count() query as a separate
operation is better.
--~--~-~--~~~---~--~~
You
whoops...
I've been using IDLE to test this and I guess it has to be reopened in
order for imports to be redone (I was just closing the shell window).
pickle.dump(metadata) works fine now.
But now I get a (recursive?) error when I try to unpickle (and I have
to kill the shell):
import pickle
On Dec 8, 2006, at 1:30 PM, Michael Bayer wrote:
the reason that get_history calls attribute_manager.get_history is
because all awareness of attribute history is handled by the
attribute module. the reason attribute_manager then calls the
InstrumentedAttribute off of the class is because
while youre free to make your own InstrumentedAttribute, i think that
is a long uphill battle which also might be incompatible with new
releases.
I think you might want to go with sticking some hidden attributes on
your pickleable object so that your custom PickleType can just use
those to
On 12/4/06, [EMAIL PROTECTED] [EMAIL PROTECTED] wrote:
Can you use real numbers and date into SQL Alchemy database?
In a word: yes. For real numbers, you'd use the Float() column type,
and for dates, you'd use the Date() column type. (Or the DateTime()
column type if you wanted both a date and
Jonathan Ellis píše v Pá 08. 12. 2006 v 11:05 -0700:
On 12/7/06, ml [EMAIL PROTECTED] wrote:
I want to have a client application accessing a remote Postgres database
but I don't want to distribute the user/password to the DB. I want
clients to authenticate against some other table of
On 12/8/06, ml [EMAIL PROTECTED] wrote:
Jonathan Ellis píše v Pá 08. 12. 2006 v 11:05 -0700:
On 12/7/06, ml [EMAIL PROTECTED] wrote:
I want to have a client application accessing a remote Postgres database
but I don't want to distribute the user/password to the DB. I want
clients to
this is a simple example of what youre trying to do. its not even as
fancy as what i described above and just uses a threadlocal to keep
track of the current Session:
from sqlalchemy import *
from sqlalchemy.orm.session import object_session
from cStringIO import StringIO
from pickle import
Jonathan Ellis píše v Pá 08. 12. 2006 v 15:40 -0700:
On 12/8/06, ml [EMAIL PROTECTED] wrote:
Jonathan Ellis píše v Pá 08. 12. 2006 v 11:05 -0700:
On 12/7/06, ml [EMAIL PROTECTED] wrote:
I want to have a client application accessing a remote Postgres database
but I don't want to
On 12/8/06, Michael Bayer [EMAIL PROTECTED] wrote:
when SA was first released, someone immediately suggested that the
identity map of Session be weak referencing, which appeared to be an
obvious improvement, so that you could load as many objects as you want
from the session and whatever you
22 matches
Mail list logo
