hi
all my tests run ok on this (as well 0.4), except the concrete-inh
case reported in
http://groups.google.com/group/sqlalchemy/browse_thread/thread/92417a6f215fa8d8/c72eb562a3070bd9
- attached again.
it's about concrete inh B of A, where polymunion contains only B, and
there are instances of
one thing that might go in a wishlist - query.filter_or()
http://groups.google.com/group/sqlalchemy/browse_thread/thread/f6798eb5ef2c0bfe
should i make it into a ticket?
as you might have noticed we've merged 0.5 into the trunk.
--~--~-~--~~~---~--~~
You
what does q.filter(x==5).filter_or(x==9).filter(y=17).filter_or(x==27)
do ? (x=5 or x=9) and (y=17 or x=27) ? ((x=5 or x=9) and y=17) or
x=27 ? etc ..
On May 12, 2008, at 8:06 AM, [EMAIL PROTECTED] wrote:
one thing that might go in a wishlist - query.filter_or()
Hi, I am using SQLALchemy 0.4.5 and PostgreSQL 8.1 db with psycopg2.
If I restart the db I need to also restart my python server that is
using the SQLAlchemy for interfacing the PostgreSQL db. Is there any
means in the SQLAlchemy to automatically reconnect to the db without
needing to restart the
On May 12, 2008, at 7:02 AM, nymaol wrote:
Hi, I am using SQLALchemy 0.4.5 and PostgreSQL 8.1 db with psycopg2.
If I restart the db I need to also restart my python server that is
using the SQLAlchemy for interfacing the PostgreSQL db. Is there any
means in the SQLAlchemy to automatically
On Monday 12 May 2008 17:01:23 Michael Bayer wrote:
what does
q.filter(x==5).filter_or(x==9).filter(y=17).filter_or(x==27) do ?
(x=5 or x=9) and (y=17 or x=27) ? ((x=5 or x=9) and y=17) or x=27
? etc ..
what pythons/C x==5 or x==9 and y==17 or x==27 does?
i know... the parenthesises.
Hello All,
I am trying to understand how to use SA and need some help.
I have several tables with 2 many-to-many relations with extra columns
and 1 only with foreign keys.
See below for the definitions of tables and mappers. I also created
classes for all tables (entities and associations).
1)
-1.
It's confusing, and there's already an extant or_ function that's documented
and not confusing. The proposal is no more cooked than it was five months
ago.
On Mon, May 12, 2008 at 11:58 AM, [EMAIL PROTECTED] wrote:
On Monday 12 May 2008 17:01:23 Michael Bayer wrote:
what does
I dont see how this:
cls.query.left_bracket().filter_or(cls.y ==
17).filter_or(cls.x==27).right_bracket()
is clearer than this:
cls.query.filter(or_(cls.y == 17, cls.x==27))
Also, another vote for cutting off python 2.3. Seriously, it's 2008.
On Mon, May 12, 2008 at 11:58 AM, [EMAIL PROTECTED]
On May 12, 2008, at 11:58 AM, [EMAIL PROTECTED] wrote:
On Monday 12 May 2008 17:01:23 Michael Bayer wrote:
what does
q.filter(x==5).filter_or(x==9).filter(y=17).filter_or(x==27) do ?
(x=5 or x=9) and (y=17 or x=27) ? ((x=5 or x=9) and y=17) or x=27
? etc ..
what pythons/C x==5 or x==9
On Monday 12 May 2008 18:12:39 Bobby Impollonia wrote:
I dont see how this:
cls.query.left_bracket().filter_or(cls.y ==
17).filter_or(cls.x==27).right_bracket()
is clearer than this:
cls.query.filter(or_(cls.y == 17, cls.x==27))
it's not. it's not about replacing the or_. noone should use it
cls.query.left_bracket().filter_or(cls.y ==
17).filter_or(cls.x==27).right_bracket()
is clearer than this:
cls.query.filter(or_(cls.y == 17, cls.x==27))
it's not. it's not about replacing the or_. noone should use it
that way...
u have horses and owners with houses and hats.
hi
i have some pre_save automatic operations and decided MVC-like to
split them into 2 phases - a per-object validation-only that goes at
save(), and a per-flush set-up 2nd phase for putting timestamps etc.
The 2nd one is via SessionExtension.before_flush().
i'm looking at the session state
On May 12, 2008, at 1:48 PM, [EMAIL PROTECTED] wrote:
cls.query.left_bracket().filter_or(cls.y ==
17).filter_or(cls.x==27).right_bracket()
is clearer than this:
cls.query.filter(or_(cls.y == 17, cls.x==27))
it's not. it's not about replacing the or_. noone should use it
that way...
u
On May 12, 2008, at 2:08 PM, [EMAIL PROTECTED] wrote:
hi
i have some pre_save automatic operations and decided MVC-like to
split them into 2 phases - a per-object validation-only that goes at
save(), and a per-flush set-up 2nd phase for putting timestamps etc.
The 2nd one is via
i have some pre_save automatic operations and decided MVC-like to
split them into 2 phases - a per-object validation-only that goes
at save(), and a per-flush set-up 2nd phase for putting
timestamps etc. The 2nd one is via
SessionExtension.before_flush().
i'm looking at the session
On May 12, 2008, at 2:37 PM, [EMAIL PROTECTED] wrote:
orphans i dont care, but the rest...
So at what time these are available - after_flush? or before_commit?
for foreign key attributes written by a dependency, before_insert().
i need to iterate over *all* would-be written-to-DB objects,
lets all repeat the mantraassociation tables with any columns
beyond the two foreign keys use the association object pattern. I
guess its a little too wordy to be catchy. Documented at
http://www.sqlalchemy.org/docs/04/mappers.html#advdatamapping_relation_patterns_association
.
orphans i dont care, but the rest...
So at what time these are available - after_flush? or
before_commit?
for foreign key attributes written by a dependency,
before_insert().
it might be useful if there is a simple sequence diagram - textual
is ok - for which kind of extension and
On May 12, 2008, at 3:16 PM, [EMAIL PROTECTED] wrote:
orphans i dont care, but the rest...
So at what time these are available - after_flush? or
before_commit?
for foreign key attributes written by a dependency,
before_insert().
it might be useful if there is a simple sequence diagram -
I saw the posts about SQLAlchemy updating an id when this is generated
by SQLAlchemy by setting Sequence() on the table's primary key Column,
but being an old Oracle hacker, I am generating the primary keys via
pre-insert triggers on the tables.
As SQLAlchemy is not selecting seq_name.nextval
hi.
i have a sort-of multicolumn m2m association table, where the primary
key is composed of all the links. At least 1 link (actualy, 2) is
always present, but never all. so i am defining all of those columns
with primary_key=True, nullable=True.
which is fine in sqlite, but doesnot work in
On May 12, 2008, at 2:54 PM, Dr.T wrote:
I saw the posts about SQLAlchemy updating an id when this is generated
by SQLAlchemy by setting Sequence() on the table's primary key Column,
but being an old Oracle hacker, I am generating the primary keys via
pre-insert triggers on the tables.
Hey -
I wrote you a reasonable test application using your tables. So I
apologize, you werent mapping secondary to an association table, I
thought I saw that but it was an email formatting issue. You were
creating overlapping names to relations though, so the attached script
resolves
24 matches
Mail list logo
