I have a number of database that have the same tables in them.
I create a session for 1 data and do some work. This work causes
a number of table table object to be reflected from the database
and then mapped. This works fine.
Now I Switch to another database which has the same tables. If I
acce
Reflection of temporary tables under MySQL works around revision 4000
and probably as recently as May but doesn't seem to work under
revision 5000 or later; I get a NoSuchTableError. Is this change
intentional?
If the only change I make to my code is to create and reflect a normal
table versus a
Michael,
After some tweaking around, it works! Thanks for all your help, it was
invaluable! I'm sure I'll be back for more though... J
Doug
From: sqlalchemy@googlegroups.com [mailto:[EMAIL PROTECTED]
On Behalf Of Michael Bayer
Sent: Friday, September 26, 2008 5:50 PM
To: sqlalchemy@googl
On Sep 26, 2008, at 5:42 PM, Doug Farrell wrote:
> I’m have the ‘type’ column in the press_routing_press linking table;
> something like this:
>
> Sess.query(Press,
> PressRoutingPress
> .type).join(Press.routes).filter(PressRouting.code==’A’).all()
>
> I’ve tried some variations of this, bu
I'm have the 'type' column in the press_routing_press linking table;
something like this:
Sess.query(Press,
PressRoutingPress.type).join(Press.routes).filter(PressRouting.code=='A'
).all()
I've tried some variations of this, but no luck. I've also tried
applying what the documentation says
sess.query(Press,
PressRouting
.code).join(Press.routes).filter(PressRouting.code=='A').all()
will return tuples in the form:
(some Press object, "somecode")
On Sep 26, 2008, at 3:25 PM, Doug Farrell wrote:
> Michael,
>
> You’re right of course, your version works fine. I cut and pasted
Sorry, I've re-checked, it was my typo, all properties co-exist ok,
Thanks again,
Alex
On 26 сент, 21:37, Michael Bayer <[EMAIL PROTECTED]> wrote:
> On Sep 26, 2008, at 1:09 PM, Alex K wrote:
>
>
>
> > Thanks, Michael.
>
> > I've just tried your example with column_property, everything works
> >
Michael,
You're right of course, your version works fine. I cut and pasted what
you have into my code framework, cleaned up some stuff and it works
there as well. One thing I cleaned up was this:
# initialize the mysql engine and SqlAlchemy base objects
engine = create_engine(__config__.da
Doug -what happened with the example I pasted ? It generates the exact SQL you describe. A full script is attached, using the mappings you've sent. It uses sqlite, so you can just run it.from sqlalchemy import *
from sqlalchemy.orm import *
engine = create_engine('sqlite://', echo=True)
metada
Michael,
Thanks perfect response. Got me to where I need to be.
Still getting used to when things are just queries vs actually being
executed i tacked on an ".all()" there at the end... and it works
prefect.
Steven
On Sep 26, 12:00 pm, Michael Bayer <[EMAIL PROTECTED]> wrote:
> On Se
On Sep 26, 2008, at 2:42 PM, Steven wrote:
>
> Sorry for being the newbie.. i just tried the "many-to-many" example
> that you provided:
>
> records =
> meta.Session.query(func.count("*").label('cnt'),
> model
> .Tag
> ).filter
> (model.Tag.id==model.pagetag_table.c.tagid).group_by(model.Tag.nam
Michael,
Thanks for your help once again. Unfortunately I'm just not 'getting
it'; I can't make the query work the way you describe. So I'm bailing
out and going back to plan SQL statements till I have more time to work
on this.
Again, thanks for your help and guidance, I'm determined to m
Sorry for being the newbie.. i just tried the "many-to-many" example
that you provided:
records =
meta.Session.query(func.count("*").label('cnt'),
model.Tag).filter(model.Tag.id==model.pagetag_table.c.tagid).group_by(model.Tag.name,
model.Tag.id).order_by(model.Tag.name.asc())
==
On Sep 26, 2008, at 12:40 PM, Steven wrote:
>
> I'm new to SQLAlchemy and still getting used to it. I'm trying to
> learn how to use the higer level apis.. the mapped objects. I still
> think in SQL and then have to work "backwards" to figure out how to
> best get SQLAlchemy to do something si
On Sep 26, 2008, at 2:56 PM, [EMAIL PROTECTED] wrote:
>
> hmm. the joins are ... hard to generate, especialy in proper/sensible
> combinations. (i've nearly given up on pushing further my
> python-expression-to-sa converter, the logic fails on 1/3 of the
> a.b.c join-cases.)
sorry you've given
Thanks.. make more sense now ( I suppose it generally does once
someone else has solved it for you ).
Hmm... it almost looks like a little tool for beginners could be
created where you enter a sql statement and it generates the
corresponding SQLAlchemy call.
Thanks again.
On Sep 26, 10:50 am, M
> > hi
> > say i have Person, Address, Street:
> > person has (many) addresess, address has (many) streets,
> > ... could be more levels down; and the relations are m2m for
> > better taste (:
> > how should the query 'who lives at street X', or
> > "who lives at street with .name.startswith('A')"
Thanks, that's what I needed to know. The use for limiting the scope
of drop_all is to avoid removing data from the base package when
uninstalling one of the dependent packages. In the case of
create_all, it's unlimited scope is probably ok, since it will only
create the missing tables.
PS: Sor
On Sep 26, 2008, at 1:09 PM, Alex K wrote:
>
> Thanks, Michael.
>
> I've just tried your example with column_property, everything works
> fine, except one thing:
>
> Does column_property hides the properties it relies on?
>
> So, once I added full_name, last_name and first_name disappeared, is
>
Using comparable property as a selected element in SQL is a use case
which hasn't been addressed as of yet - the main usage was so far just
for producing WHERE/ORDER BY/etc. expressions. In particular, the
comparable property doesn't have any query setup capabilities (i.e.
where it puts
Thanks, Michael.
I've just tried your example with column_property, everything works
fine, except one thing:
Does column_property hides the properties it relies on?
So, once I added full_name, last_name and first_name disappeared, is
it possible to have them existing along with full_name?
Alex
I'm new to SQLAlchemy and still getting used to it. I'm trying to
learn how to use the higer level apis.. the mapped objects. I still
think in SQL and then have to work "backwards" to figure out how to
best get SQLAlchemy to do something similar.
===
Michael,
You are right, I have been using "String" with no length when I want a
CLOB column and "Unicode" for varchar2 datatypes in my Entity
definitions (using Elixir).
I'll change my model to explicitly use UnicodeText.
N.B.
This change in behaviour is noted in the 05Migration notes as you sa
mraer schrieb:
> This simple test fails with error:
> DataError: (DataError) unable to parse time None None
Which version are you using?
Your test works for me (except I needed to change "Mapper" to "mapper")
with SA 0.5rc1 and psycopg 2.0.8, Py 2.5, Postgres 8.3.1 on WinXP.
As an aside, you s
My configuration is:
Python 2.5
SA 0.5rc1
psycopg 2.0.8
Postgres 8.3.4
Windows XP
The problem was in locale. I have russian locale and Moscow time zone.
DB had configuration which outputed this date (in 1919 year) as
string: "1919-01-01 15:04:48+03:30:48". Look at time zone offset.
Psycopg2 fails
On Sep 26, 2008, at 11:23 AM, picoplex wrote:
>
> The code that chokes with SA 0.5.0rc1 is as follows:
>
> filteredRecords =
> theChildClass
> .query().filter(filterString).join(relationshipToParentName,
> aliased
> =requiresAliased).filter(parentFilterString).order_by(orderByString)
> [start:(s
On Sep 26, 2008, at 11:40 AM, Jon wrote:
>
> I have a request for the 0.5 documentation.
> I spent /way/ too much time looking for the functionality provided by
> the version_id_col Mapper __init__ param. For those of us using
> database structures formerly (or still) in use by hibernate this is
Hi All,
I have created comparable property "JoinedValues" that represents any
joined columns, e.g. full name for first_name and last_name.
Everything works, fine, except the moment when I try to select this
property explicitly, e.g. like:
session.query(User.full_name)... i get error " 'ClauseLis
I have a request for the 0.5 documentation.
I spent /way/ too much time looking for the functionality provided by
the version_id_col Mapper __init__ param. For those of us using
database structures formerly (or still) in use by hibernate this is a
big deal. Therefore, I request a slight bump in it
The code that chokes with SA 0.5.0rc1 is as follows:
filteredRecords =
theChildClass.query().filter(filterString).join(relationshipToParentName,
aliased=requiresAliased).filter(parentFilterString).order_by(orderByString)
[start:(start+limit)]
where:
theChildClass: blnver
filterString: '1=1'
Information on "primaryjoin" and "secondaryjoin" is available at:
http://www.sqlalchemy.org/docs/05/mappers.html#advdatamapping_relation_customjoin
On Sep 25, 2008, at 8:18 PM, Jorge Vargas wrote:
>
> hello I'm having a bit of troubles with the following case.
>
> http://paste.turbogears.org/p
On Sep 26, 2008, at 6:25 AM, Petter Urkedal wrote:
>
> I have a package which implements users, groups, and access control
> using
> SQLAlchemy. On top of that I'm writing applications as separate
> packages.
> These will make foreign keys into the tables of the user management
> package.
> At
On Sep 26, 2008, at 6:59 AM, [EMAIL PROTECTED] wrote:
>
> hi
> say i have Person, Address, Street:
> person has (many) addresess, address has (many) streets,
> ... could be more levels down; and the relations are m2m for better
> taste (:
> how should the query 'who lives at street X', or
> "who
On Sep 26, 2008, at 9:23 AM, Doug Farrell wrote:
> I'm trying to represent a many-to-many relationship between the
> press_table and the press_routing table using the linking table,
> press_routing_press. I think I've got the table structure and
> mapping set up, but now I need some help to
SQLAlchemy's PGDateTime object does nothing, so its likely the issue
you're having is a psycopg2 issue. You should ask on their list
(using psycopg2-only code).
On Sep 26, 2008, at 5:28 AM, mraer wrote:
>
> This simple test fails with error:
> DataError: (DataError) unable to parse time No
On Sep 26, 2008, at 9:39 AM, picoplex wrote:
>
> I have been porting an app from Windows to linux.
>
> Whereas the windows app using SA 0.4.5 worked fine, I had a SA erro
> with the installed version on Linux - SA 0.4.0 (AttributeError: 'list'
> object has no attribute 'values' in get_children i
I have been porting an app from Windows to linux.
Whereas the windows app using SA 0.4.5 worked fine, I had a SA erro
with the installed version on Linux - SA 0.4.0 (AttributeError: 'list'
object has no attribute 'values' in get_children in expression.py).
I therefore upgraded SA on linux to th
Hi all,
Well, I'm still stumped by SqlAlchemy syntax, but it's getting better.
I've got some tables built this way using SqlAlchemy 0.5:
press_table = Table('press', metadata, autoload=True)
press_routing_table = Table('press_routing', metadata, autoload=True)
press_routing_press_table
I have a package which implements users, groups, and access control
using
SQLAlchemy. On top of that I'm writing applications as separate
packages.
These will make foreign keys into the tables of the user management
package.
At the moment I use separate metadata for each package, and the
foreign
Hi,
I was only able to achieve this via joins and eagerloads, so I would
be also interested in the relations based example
Alex
On 26 сент, 14:59, [EMAIL PROTECTED] wrote:
> hi
> say i have Person, Address, Street:
> person has (many) addresess, address has (many) streets,
> ... could be more l
Thank you again for a great answer.
It seems like I lack full knowledge about how the basic things works
with SqlAlchemy and Elixir. I'll try to start from scratch, read the
documentation better and follow your suggestions when I rebuild.
Thank you for your help.
--~--~-~--~~-
hi
say i have Person, Address, Street:
person has (many) addresess, address has (many) streets,
... could be more levels down; and the relations are m2m for better
taste (:
how should the query 'who lives at street X', or
"who lives at street with .name.startswith('A')" look like?
is it expres
This simple test fails with error:
DataError: (DataError) unable to parse time None None
def testDate1919(self):
class DateHolder(object):
def __init__(self, date):
self.date = date
metadata = MetaData()
dateTable = Table(
"date",
43 matches
Mail list logo
