Hi all, i've added a new table recentli in main DB, but at first
insertion have this problem:
any time i try an insertion the sqlalchemy engine search nextval of a
partial primary key.
SQLError: (ProgrammingError) relation scadenziario_id_piano_seq does
not exist
'select
there was some recent thread on this 2-3 weeks ago, lookup..
On Friday 17 August 2007 11:28:34 Glauco wrote:
What's is the best solution for a web procedure , in TurboGear,
that produce a large amount of insert into ? (from 2000 to 2
insertions on a submit)
i've done some try with
Hi all,
This is sort of a general db question I was wondering about. So, I'm
trying to write a small blog system (not using the mapper, only sql
queries written on my own until now) and I am not sure:
Say, I want to show one blog post and all the comments for that blog
entry (one-to-many). I
Hi,
Any chance Ticket #731 get into 0.4 ? I'd prefer not relying on a
patch if possible.
Thanks a lot,
Christophe
--~--~-~--~~~---~--~~
You received this message because you are subscribed to the Google Groups
sqlalchemy group.
To post to this group, send
Martin wrote:
a) 2 separate select-queries? One query going like:
SELECT * from blog_posts WHERE post_id = my_post_id
and the other sth. like
SELECT * from comments WHERE parent_id = my_post_id.
b) one join-query? Joining the two tables over my_post_id, so I only
need 1 sql-query?
Two
On Aug 17, 2007, at 4:21 AM, Glauco wrote:
SQLError: (ProgrammingError) relation scadenziario_id_piano_seq does
not exist
'select nextval(\'scadenziario_id_piano_seq\')' {}
but
the table is:
CREATE TABLE scadenziario (
id_pianoINTEGER NOT NULL
In Postgres at least, serial columns are typically defined as:
CREATE TABLE FOO (
id_piano serial primary key,
...
)
This automatically creates a sequence foo_id_piano_seq. Of course,
you can also create a sequence separate from a table with CREATE
SEQUENCE, but this is (IMHO) wordy.
justanyone ha scritto:
In Postgres at least, serial columns are typically defined as:
CREATE TABLE FOO (
id_piano serial primary key,
...
)
This automatically creates a sequence foo_id_piano_seq. Of course,
you can also create a sequence separate from a table with CREATE
SEQUENCE,
Could someone please explain the following behaviour:
from django.http import HttpResponseRedirect, Http404
from django.shortcuts import render_to_response
from df.tst import Path as ThePath
from sqlalchemy import *
from sqlalchemy.orm import *
metadata = MetaData()
u_table = Table('auth_user',
Hi,
Any chance Ticket #731 get into 0.4 ? I'd prefer not relying on a
patch if possible.
I'd applied this to trunk and rel_0_3. I don't have alert mails setup
from trac (maybe I should) so if you drop me a mail when you want an
MSSQL patch looking at, I'll do my best.
Paul
the traditional wisdom is that the join is faster, because you have
fewer round trips to the database. particularly if you are loading
many parent items each with many child items. if you are loading just
a single parent item, the second SELECT to get its child items is not
a very large hit at
its a pickling error. Pickle is not managing to map your pickled data
back to the correct module name from which to import the Path
object.If this module (i.e., where the Path class originates) is
run by django in some kind of dynamic importing scheme, that could
definitely break pickle's
On 8/17/07, Michael Bayer [EMAIL PROTECTED] wrote:
its a pickling error. Pickle is not managing to map your pickled data
back to the correct module name from which to import the Path
object.If this module (i.e., where the Path class originates) is
run by django in some kind of dynamic
2007/8/17, Paul Johnston [EMAIL PROTECTED]:
Hi,
Any chance Ticket #731 get into 0.4 ? I'd prefer not relying on a
patch if possible.
I'd applied this to trunk and rel_0_3. I don't have alert mails setup
from trac (maybe I should) so if you drop me a mail when you want an
MSSQL patch
well at the very least you want to ensure that executemany() is being
used for the inserts (i.e. with sqlalchemy, dont use ORM, use
connection.execute(statement, [param, param, param...]).
If you use SQLAlchemy's executemany() facilities, we've just
increased their efficiency by about 60% in the
Glauco ha scritto:
Thank you kevin , i've worked a lot over PG and this tecnique is
consolidated in our model.. but here the problem is that the column
is a simplycolumn_name INTEGER NOT NULL references
other_table(id) and i cannot understand why sqlalchemy whant to use it
as a
Lukasz Szybalski ha scritto:
bind_meta_data()
users_table = Table('users', metadata, autoload=True)
class Users(object):
pass
usersmapper=mapper(Users,users_table)
assign_mapper() in place of mapper()
mysession=session.query(Users)
1. What would be the code from now on to
hello -
you generally use mapper() and relation() to set up how you'd like
your classes to correspond to your table relationships. as far as
compound keys, if they are defined with a primary key constraint you
shouldn't have to worry about them.
---
Ok. So we are using mapper()
ok...correct me if I'm wrong.
#we start by importing
from turbogears.database import metadata, session,bind_meta_data
from sqlalchemy.ext.assignmapper import assign_mapper
from turbogears import widgets, validators
import sqlalchemy
#Then we bound to database
bind_meta_data()
#create a table
On 8/17/07, Michael Bayer [EMAIL PROTECTED] wrote:
hi Vetle -
the Table object supports a join() method, which can be called in a
chain, i.e.
table1.join(table2).join(table3)...
if you need to specify the ON part of the join, its the second
argument to join() (below illustrated with
On Aug 17, 11:54 am, Vitaliyi [EMAIL PROTECTED] wrote:
I've tried this :
import psycopg2
...
s = curs.fetchone()[0]
ls = cPickle.loads(str(s))
and it returned :
AttributeError: 'module' object has no attribute 'Path'
just as django did. it pretends that 'Path' is not defined. Seems
21 matches
Mail list logo
