[sqlalchemy] Re: Inconsistent results in session.flush()
Michael Bayer wrote: On Mar 17, 2007, at 11:39 AM, Simon King wrote: I had assumed that post_update was only necessary when you are inserting two rows that are mutually dependent, but in this case I wasn't inserting either a ReleaseLine or a Label. I suppose post_update can actually have a knock-on affect across the whole dependency graph. Yes, you are correct. I did the requisite three hours of staring and uncovered the actual condition that led to this problem, which is definitely a bug. Because the second flush() did not require any update operation to either the ReleaseLine or the Label object, the circular dependency graph that would be generated for them just gets thrown away, but that also threw away a processor thats supposed to be setting the release_line_id column on your Branch table. so the fix is, if the circular dependency graph doesnt get generated for a particular cycle between mappers, or if a particular mapper within the cycle isnt included in the circular dependency graph, make sure the flush task for that mapper still gets tacked on to the returned structure so that it can operate on other things external to the cycle, in this case your Branch. the Branch is related to the ReleaseLine/Label via a many-to-one, which was not as stressed in the unit tests so its harder for this issue to pop up (also requires the two separate flushes...so very hard to anticipate this problem). So you can take the post_update out and update to rev 2424. Yep. I can definitely say that I wouldn't have figured that one out. Thanks again for all your help, Cheers, Simon --~--~-~--~~~---~--~~ You received this message because you are subscribed to the Google Groups sqlalchemy group. To post to this group, send email to sqlalchemy@googlegroups.com To unsubscribe from this group, send email to [EMAIL PROTECTED] For more options, visit this group at http://groups.google.com/group/sqlalchemy?hl=en -~--~~~~--~~--~--~---
[sqlalchemy] Re: Inconsistent results in session.flush()
Michael Bayer wrote: So you can take the post_update out and update to rev 2424. i have a problem with this 2424, before that was ok. As i can't separate a short case now, here the conditions: multi-table-inheritance, polymorphic. Nothing else too fancy. table_Entity has primary db_id, hand-made/sequence obj_id and other data; While populating the database: ... * SA: INFO INSERT INTO Entity (disabled, obj_id, atype) VALUES (?, ?, ?) * SA: INFO [0, 1, 'Person'] * SA: INFO INSERT INTO Person (name_id, db_id) VALUES (?, ?, ) * SA: INFO [2, 1] and now 2423 does: * SA: INFO COMMIT while 2424 does: * SA: INFO INSERT INTO Entity (disabled, obj_id, db_id, atype) VALUES (?, ?, ?, ?) * SA: INFO [0, 1, 1, 'Person'] * SA: INFO ROLLBACK for whatever reason it decides that the same object is dirty AND new, and attempts to reinsert it once more. Any idea? do u want logs? --~--~-~--~~~---~--~~ You received this message because you are subscribed to the Google Groups sqlalchemy group. To post to this group, send email to sqlalchemy@googlegroups.com To unsubscribe from this group, send email to [EMAIL PROTECTED] For more options, visit this group at http://groups.google.com/group/sqlalchemy?hl=en -~--~~~~--~~--~--~---
[sqlalchemy] Re: Conflicting columns with secondaryjoin
Apparently my reply a couple of days ago didn't appear, hmm. No matter, it was a load of old tosh anyway. The issue was resolved by my updating to 0.3.5. I was still using 0.3.3 at the time - should've checked that. However I then started experiencing strangeness in that SA was attempting to insert two of the same row into the secondaryjoin table (causing an IntegrityError). Having just noticed a mistake in my code, it seems I was trying to put an instance of the secondaryjoin table into the 'tags' list attribute, where an instance of the primaryjoin should've gone... Is there some way to protect against/catch such things? WrongObjectTypeYouEejitError? ;) Also, the tags attribute doesn't seem to have any problem with flushing AFAICT, it works exactly as expected. :) Takk, - Mel C --~--~-~--~~~---~--~~ You received this message because you are subscribed to the Google Groups sqlalchemy group. To post to this group, send email to sqlalchemy@googlegroups.com To unsubscribe from this group, send email to [EMAIL PROTECTED] For more options, visit this group at http://groups.google.com/group/sqlalchemy?hl=en -~--~~~~--~~--~--~---
[sqlalchemy] eager load with polymorphic
Hi,
I tried to eager load some data from class that have reference to
polymorhic class. The object model is: class Base inherited by both
Manager and Address, Managers has reference to Address - when I stop
here SA can eager loading the addresses of managers. The problem arise
when I add another class MailAddress that inherits Address - in this
case address of Managers are simply lazy loaded, despite that it is
explicitly set to be eager. Maybe the problem is that somehow
polymorhic and eager cannot be combined?
please advice
TIA
Stefan
Below is example of code demonstrating the behavior:
from sqlalchemy import *
db_sqlite = create_engine( 'sqlite:///:memory:')
MODEL_WITH_MAIL_ADDR = 1 # 1 - WITH MAILADDRESS, 0 - WITHOUT
MAILADDRESS
def checkWith( db):
meta = BoundMetaData( db)
meta.engine.echo = 0
table_Base = Table( 'base', meta,
Column( 'discriminator', type= String, ),
Column( 'id', Integer, primary_key= True, ),
)
table_Manager = Table( 'manager', meta,
Column( 'address_id', Integer, ForeignKey( 'address.id', name=
'address_id_fk',), ),
Column( 'name', type= String, ),
Column( 'id', Integer, ForeignKey('base.id', name=
'manager_inh_fk'), primary_key= True, ),
)
table_Address = Table( 'address', meta,
Column( 'country', type= String, ), #for case without
mailaddress
Column( 'city', type= String, ),
Column( 'id', Integer, ForeignKey('base.id', name=
'address_inh_fk'), primary_key= True, ),
)
if MODEL_WITH_MAIL_ADDR:
table_Mail = Table( 'mailaddress', meta,
Column( 'country', type= String, ),
Column( 'id', Integer, ForeignKey('base.id', name=
'address_inh_fk'), primary_key= True,
)
class Base( object):
def set( me, **kargs):
for k,v in kargs.iteritems(): setattr( me, k, v)
return me
def __str__(me): return str(me.__class__.__name__)
+':'+str(me.name)
__repr__ = __str__
class Manager( Base):
def __str__(me):
res = Base.__str__(me)+':'+str(me.address.city)
if MODEL_WITH_MAIL_ADDR:
res += ':'+str(me.address.country)
return res
__repr__ = __str__
class Address( Base): pass
class MailAddress( Address): pass
meta.create_all()
def make_mappers():
propdic = {
'manager': table_Base.join( table_Manager, onclause=
(table_Manager.c.id==table_Bas
'address': table_Base.join( table_Address, onclause=
(table_Address.c.id==table_Bas
'base':
table_Base.select( table_Base.c.discriminator=='base'),
}
if MODEL_WITH_MAIL_ADDR:
propdic.update( { 'mailaddress':
table_Base.join( table_Address.join( table_Mail
, onclause=
(table_Mail.c.id==table_Address.c.id))
, onclause=
(table_Address.c.id==table_Base.c.id)
), })
join_Address = polymorphic_union( {
'address': table_Base.join( table_Address, onclause=
(table_Address.c.id==table_Bas
'mailaddress': table_Base.join( table_Address,
onclause=
(table_Address.c.id==table_Base.c.id)).join( table_Mail,
onclause=
(table_Mail.c.id==table_Address.c.id)
),
}, None)
join_Base = polymorphic_union( propdic, None)
mapper_Base = mapper( Base, table_Base, select_table=
join_Base
, polymorphic_on= join_Base.c.discriminator,
polymorphic_identity= 'base')
mapper_Manager = mapper( Manager, table_Manager,
properties= { 'address' : relation( Address, primaryjoin=
(table_Manager.c.address_id==t
, inherits= mapper_Base
, inherit_condition= (table_Manager.c.id==table_Base.c.id)
, polymorphic_identity='man'
)mapper_Address = mapper( Address, table_Address
, inherits= mapper_Base
, inherit_condition= (table_Address.c.id==table_Base.c.id)
, polymorphic_identity='adr'
)
if MODEL_WITH_MAIL_ADDR:
mapper_Address.select_table = join_Address
mapper_MailAddress = mapper( MailAddress, table_Mail
, inherits= mapper_Address
, inherit_condition=
(table_Mail.c.id==table_Address.c.id)
, polymorphic_identity='mai'
)
make_mappers()
if not MODEL_WITH_MAIL_ADDR:
MailAddress = Address
c = Manager().set( name= 'pencho')
d = Manager().set( name= 'torencho')
e = Manager().set( name= 'mnogoVojdMalkoIndianec')
f = MailAddress().set( city= 'varna', country= 'BG')
g = MailAddress().set( city= 'sofia', country= 'RU')
h = MailAddress().set( city= 'burga', country= 'US')
c.address, d.address, e.address = f, g, h
session = create_session()
session.save(c)
session.save(d)
[sqlalchemy] SELECT 'a fixed string', ...
hi, I am currently using sqlalchemy to build SQL queries and it's a fantastic tool! By now, I am looking for a way to build: SELECT 'a_fixed_string', atable.col1, atable.col2 FROM atable using the syntax: select([XXX, atable.c.col1, atable.c.col2]) but I don't know what to put at XXX. Does anyone have a clue ? -- Bertrand Croq, Ingénieur Développement ___ Net-ng Tel : +33 (0)223 21 21 52 Immeuble Germanium Fax : +33 (0)223 21 21 60 80 av. des Buttes de CoesmesWeb : http://www.net-ng.com 35000 RENNESe-mail: [EMAIL PROTECTED] FRANCE Ce message et tout document joint sont confidentiels. Toute diffusion ou publication en est interdite. Si vous recevez ce message par erreur, merci d'en avertir immédiatement l'expéditeur par e-mail et de supprimer ce message et tout document joint. This message and any attachment are confidential. Any use is prohibited except formal approval. If you receive this message in error, please notify the sender by return e-mail and delete this message and any attachment from your system. --~--~-~--~~~---~--~~ You received this message because you are subscribed to the Google Groups sqlalchemy group. To post to this group, send email to sqlalchemy@googlegroups.com To unsubscribe from this group, send email to [EMAIL PROTECTED] For more options, visit this group at http://groups.google.com/group/sqlalchemy?hl=en -~--~~~~--~~--~--~---
[sqlalchemy] SQLite and ON CONFLICT clause?
Hi all, browsing the SQLAlchemy docs I coudn't find any references to a particular SQLite feature, ON CONFLICT clause (http://www.sqlite.org/ lang_conflict.html), is there any support for this in SA? What I want to do is this, create my database as such: CREATE TABLE songs( id INTEGER PRIMARY KEY ON CONFLICT REPLACE, name TEXT ); Now, when I do a bunch of INSERTs without specifying 'id' field, everything works as you would expect. But, when I issue a INSERT with a 'id' that allready exists, SQLite automatically and automagicaly does an UPDATE instead. Now, this is very handy because I don't have to do a SELECT prior to modifying a table. Is there a way to use this SQLite feature from SA? Or somehow simulate it without too much gory details? Thanks guys, klm. --~--~-~--~~~---~--~~ You received this message because you are subscribed to the Google Groups sqlalchemy group. To post to this group, send email to sqlalchemy@googlegroups.com To unsubscribe from this group, send email to [EMAIL PROTECTED] For more options, visit this group at http://groups.google.com/group/sqlalchemy?hl=en -~--~~~~--~~--~--~---
[sqlalchemy] Firebird backend
Hi all, I attached a snapshot of my current source of the Firebird backend for SA to ticket #410. This version passes several more unittests, in particular wrt the supported types: BLOBs and Unicode should work now. Since I'd be very sad knowing I'm breaking someone else project, I'd like to hear a voice from some other FB user, before asking Michael to accept this work. There are mainly two areas that still need some work: a) better handling of dialect 1 (a.k.a. Interbase 5.x) datatypes b) correct the way the backend deals with case-sensitive a) requires some way to reach the actual connection from the dialect.type_descriptor(), that needs to know which dialect the connection is talking to give the right answer. b) is now alleviated, in dialect.has_table(), by using the LIKE operator instead of a straight = on the table name; more, the dialect.reflecttable() now initialize the case_sensitive flag on each column, but this clearly needs more work, mainly to better understand the whole logic behind case_sensitive on my part. Please see: http://www.sqlalchemy.org/trac/attachment/ticket/410/firebird.py Thanks in advance for any feedback, ciao, lele. --~--~-~--~~~---~--~~ You received this message because you are subscribed to the Google Groups sqlalchemy group. To post to this group, send email to sqlalchemy@googlegroups.com To unsubscribe from this group, send email to [EMAIL PROTECTED] For more options, visit this group at http://groups.google.com/group/sqlalchemy?hl=en -~--~~~~--~~--~--~---
[sqlalchemy] Re: Quoting column names on table create?
Karlo Lozovina wrote: I was just wondering, why does SA quote column names that have mixed case in them, and leaves them unquoted for lowercase column names? Because in general SQL engines tend to be case-insensitive, that is, under most of them the query SELECT column_a, column_b FROM table is perfectly equivalent to SELECT COLUMN_A, Column_B FROM Table To preserve the case of the entities you have to quote them (for example because you have two distinct tables, table and Table): this forces the engine to take the name literally, so that the first query above is *not* equivalent to the following SELECT column_a, column_b FROM table at least not when the default case used by the engine is uppercase (just imagine that the database server change every non-quoted entity name to either upper- or lower-case, and you are close to the fact :) So when SA thinks that the entity name it is handling is case sensitive (that always is when the name is CamelCased) it put automatically put quotes around it. Hope this clarifies a little, ciao, lele. --~--~-~--~~~---~--~~ You received this message because you are subscribed to the Google Groups sqlalchemy group. To post to this group, send email to sqlalchemy@googlegroups.com To unsubscribe from this group, send email to [EMAIL PROTECTED] For more options, visit this group at http://groups.google.com/group/sqlalchemy?hl=en -~--~~~~--~~--~--~---
[sqlalchemy] Re: eager load with polymorphic
On Mar 19, 2007, at 8:29 AM, che wrote: Hi, I tried to eager load some data from class that have reference to polymorhic class. The object model is: class Base inherited by both Manager and Address, Managers has reference to Address - when I stop here SA can eager loading the addresses of managers. The problem arise when I add another class MailAddress that inherits Address - in this case address of Managers are simply lazy loaded, despite that it is explicitly set to be eager. Maybe the problem is that somehow polymorhic and eager cannot be combined? please advice thats correct, in many cases eager loading will degrade to lazy loading, particularly when it detects a self-referential table relationship (which happens often with polymorphic relationships). if you turn on sqlalchemy.orm logging you should see this in the logs to confirm. --~--~-~--~~~---~--~~ You received this message because you are subscribed to the Google Groups sqlalchemy group. To post to this group, send email to sqlalchemy@googlegroups.com To unsubscribe from this group, send email to [EMAIL PROTECTED] For more options, visit this group at http://groups.google.com/group/sqlalchemy?hl=en -~--~~~~--~~--~--~---
[sqlalchemy] Re: SELECT 'a fixed string', ...
use literal_column('fixedstring') in the column clause
On Mar 19, 2007, at 9:38 AM, Bertrand Croq wrote:
hi,
I am currently using sqlalchemy to build SQL queries and it's a
fantastic
tool! By now, I am looking for a way to build:
SELECT 'a_fixed_string', atable.col1, atable.col2
FROM atable
using the syntax:
select([XXX, atable.c.col1, atable.c.col2])
but I don't know what to put at XXX.
Does anyone have a clue ?
--
Bertrand Croq,
Ingénieur Développement
___
Net-ng Tel : +33 (0)223 21 21 52
Immeuble Germanium Fax : +33 (0)223 21 21 60
80 av. des Buttes de CoesmesWeb : http://www.net-ng.com
35000 RENNESe-mail: [EMAIL PROTECTED]
FRANCE
Ce message et tout document joint sont confidentiels. Toute
diffusion ou
publication en est interdite. Si vous recevez ce message par erreur,
merci d'en avertir immédiatement l'expéditeur par e-mail et de
supprimer ce message et tout document joint.
This message and any attachment are confidential. Any use is
prohibited
except formal approval. If you receive this message in error, please
notify the sender by return e-mail and delete this message and any
attachment from your system.
--~--~-~--~~~---~--~~
You received this message because you are subscribed to the Google Groups
sqlalchemy group.
To post to this group, send email to sqlalchemy@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at
http://groups.google.com/group/sqlalchemy?hl=en
-~--~~~~--~~--~--~---
[sqlalchemy] Re: SQLite and ON CONFLICT clause?
since its SQLite specific, you should just use literal DDL statements to create the table with this feature. On Mar 19, 2007, at 10:57 AM, Karlo Lozovina wrote: Hi all, browsing the SQLAlchemy docs I coudn't find any references to a particular SQLite feature, ON CONFLICT clause (http://www.sqlite.org/ lang_conflict.html), is there any support for this in SA? What I want to do is this, create my database as such: CREATE TABLE songs( id INTEGER PRIMARY KEY ON CONFLICT REPLACE, name TEXT ); Now, when I do a bunch of INSERTs without specifying 'id' field, everything works as you would expect. But, when I issue a INSERT with a 'id' that allready exists, SQLite automatically and automagicaly does an UPDATE instead. Now, this is very handy because I don't have to do a SELECT prior to modifying a table. Is there a way to use this SQLite feature from SA? Or somehow simulate it without too much gory details? Thanks guys, klm. --~--~-~--~~~---~--~~ You received this message because you are subscribed to the Google Groups sqlalchemy group. To post to this group, send email to sqlalchemy@googlegroups.com To unsubscribe from this group, send email to [EMAIL PROTECTED] For more options, visit this group at http://groups.google.com/group/sqlalchemy?hl=en -~--~~~~--~~--~--~---
[sqlalchemy] Re: SQLite and ON CONFLICT clause?
On Mar 19, 2007, at 12:04 PM, Karlo Lozovina wrote: Btw, is it probable that SA will some day include SQLites' ON CONFLICT clause (both CREATE TABLE and INSERT OR forms)? It's rather ugly to have to specify my tables manually using .execute(). the issue is that every database has dozens if not hundreds of custom, non-SQL syntaxes when creating tables. some method of allowing these syntaxes generically would have to be devised, but even that will not allow everyone's syntaxes to become available. i definitely cant just add flags for every kind of option to the Table API, thered be hundreds. --~--~-~--~~~---~--~~ You received this message because you are subscribed to the Google Groups sqlalchemy group. To post to this group, send email to sqlalchemy@googlegroups.com To unsubscribe from this group, send email to [EMAIL PROTECTED] For more options, visit this group at http://groups.google.com/group/sqlalchemy?hl=en -~--~~~~--~~--~--~---
[sqlalchemy] Re: cascading question
yes, thats ticket #249. ive expanded the scope of that ticket since
this particular pattern is much more general usage than what I
initially thought it was.
anyway this will be the next ticket i fix since its pretty major.
On Mar 19, 2007, at 5:54 AM, Alexandre CONRAD wrote:
Hello Mike,
Here's a simple test case script that creates tables in sqlite
(memory).
To resume, we have site_table table and option_table objects as a
many-to-many relationship through a secondary weak table
options_has_sites.
I want to be able to:
- delete a site without deleting the attached options.
- delete an option without deleting the attached sites.
- make sure the weak table is beeing cleaned up correctly depending
if a
site or an option has been removed.
Regards,
--
Alexandre CONRAD
Michael Bayer wrote:
the rows in the M2M table should be deleted automatically.it
*might* require that your instances are present in the session but
its not supposed to. can you make me a short test case for this
one ? its not the first time ive heard about it.
technically you can just put ON DELETE CASCADE on the tables too but
id like to fix this issue.
On Mar 16, 2007, at 12:19 PM, Alexandre CONRAD wrote:
Humm, this doesn't help as if a site is deleted, it deletes the
options
that where related to that option.
I want to be able to:
- delete an option without deleting a site
- delete a site without deleting an option
just delete (clean up) the related rows inside the weak
options_has_sites table.
I just can't figure it out...
Michael Bayer wrote:
youd need to add some delete cascades to your relationship, maybe
on just the backref (using the backref() function), e.g.
option_mapper = assign_mapper(ctx, Option, option_table,
properties={
'sites':relation(Site, backref=backref(options,
cascade=save-update, delete),
secondary=options_has_sites, cascade=save-update),
},
order_by=option_table.c.name,
)
On Mar 16, 2007, at 10:38 AM, Alexandre CONRAD wrote:
Hello,
I have a many-to-many relation between an option table and a site
table.
Deleting an option correctly deletes the related rows in
options_has_sites table.
But when I delete a site, I have some orphan rows in the
options_has_sites table. How can I avoid this ?
# SITE TABLE ---
site_table = Table('sites', meta,
Column('id', Integer, primary_key=True),
Column('name', Unicode(20), nullable=False, unique=True),
)
class Site(object):
pass
site_mapper = assign_mapper(ctx, Site, site_table,
order_by=site_table.c.name,
)
# OPTION TABLE
option_table = Table('options', meta,
Column('id', Integer, primary_key=True),
Column('name', Unicode(20), unique=True, nullable=False),
)
options_has_sites = Table('sites_has_options', meta,
Column('id_site', None, ForeignKey('sites.id'),
primary_key=True),
Column('id_option', None, ForeignKey('options.id'),
primary_key=True),
)
class Option(object):
pass
option_mapper = assign_mapper(ctx, Option, option_table,
properties={
'sites':relation(Site, backref=options,
secondary=options_has_sites, cascade=save-update),
},
order_by=option_table.c.name,
)
Should I play with backref() ?
Regards,
--
Alexandre CONRAD
--
Alexandre CONRAD - TLV FRANCE
Research Development
-
--
Texte inséré par Platinum 2007:
S'il s'agit d'un mail indésirable (SPAM), cliquez sur le lien
suivant pour le reclasser : http://127.0.0.1:6083/Panda?
ID=pav_31925SPAM=true
-
--
from sqlalchemy import *
meta = DynamicMetaData()
# Sites
site_table = Table('sites', meta,
Column('id', Integer, primary_key=True),
Column('name', Unicode(20), nullable=False, unique=True),
Column('email', Unicode(100)),
)
class Site(object):
pass
site_mapper = mapper(Site, site_table,
order_by=site_table.c.name,
)
# Options
option_table = Table('options', meta,
Column('id', Integer, primary_key=True),
Column('name', Unicode(20), unique=True, nullable=False),
Column('description', Unicode(40)),
)
options_has_sites = Table('options_has_sites', meta,
Column('id_site', None, ForeignKey('sites.id'), primary_key=True),
Column('id_option', None, ForeignKey('options.id'),
primary_key=True),
)
class Option(object):
def __repr__(self):
return Option: %s % repr(self.name)
option_mapper = mapper(Option, option_table,
properties={
'sites':relation(Site, backref=options,
secondary=options_has_sites, cascade=save-update),
},
order_by=option_table.c.name,
)
meta.connect(sqlite://, echo=True)
meta.create_all()
# Make session.
session = create_session()
#
[sqlalchemy] Re: eager load with polymorphic
Thanks Michael, Michael Bayer написа: On Mar 19, 2007, at 8:29 AM, che wrote: Maybe the problem is that somehow polymorhic and eager cannot be combined? please advice thats correct, in many cases eager loading will degrade to lazy loading, particularly when it detects a self-referential table relationship (which happens often with polymorphic relationships). if you turn on sqlalchemy.orm logging you should see this in the logs to confirm. unfortunately you are right - in the case of only Address I get in the logs: ... * SA: DEBUG eagerload scalar instance on [EMAIL PROTECTED] ... and in the MailAddress case: ... * SA: DEBUG degrade to lazy loader on [EMAIL PROTECTED] * SA: DEBUG set instance-level lazy loader on [EMAIL PROTECTED] [* SA: DEBUG Executing lazy callable on [EMAIL PROTECTED] * SA: DEBUG lazy load attribute address on instance [EMAIL PROTECTED] ... Is it possible to workaround this behaviour somehow now (or some plans for the SA future)? regards, Stefan --~--~-~--~~~---~--~~ You received this message because you are subscribed to the Google Groups sqlalchemy group. To post to this group, send email to sqlalchemy@googlegroups.com To unsubscribe from this group, send email to [EMAIL PROTECTED] For more options, visit this group at http://groups.google.com/group/sqlalchemy?hl=en -~--~~~~--~~--~--~---
[sqlalchemy] Re: Inconsistent results in session.flush()
but sure give me the logs with sqlalchemy.orm == logging.DEBUG On Mar 19, 2007, at 6:49 AM, svilen wrote: Michael Bayer wrote: So you can take the post_update out and update to rev 2424. i have a problem with this 2424, before that was ok. As i can't separate a short case now, here the conditions: multi-table-inheritance, polymorphic. Nothing else too fancy. table_Entity has primary db_id, hand-made/sequence obj_id and other data; While populating the database: ... * SA: INFO INSERT INTO Entity (disabled, obj_id, atype) VALUES (?, ?, ?) * SA: INFO [0, 1, 'Person'] * SA: INFO INSERT INTO Person (name_id, db_id) VALUES (?, ?, ) * SA: INFO [2, 1] and now 2423 does: * SA: INFO COMMIT while 2424 does: * SA: INFO INSERT INTO Entity (disabled, obj_id, db_id, atype) VALUES (?, ?, ?, ?) * SA: INFO [0, 1, 1, 'Person'] * SA: INFO ROLLBACK for whatever reason it decides that the same object is dirty AND new, and attempts to reinsert it once more. Any idea? do u want logs? --~--~-~--~~~---~--~~ You received this message because you are subscribed to the Google Groups sqlalchemy group. To post to this group, send email to sqlalchemy@googlegroups.com To unsubscribe from this group, send email to [EMAIL PROTECTED] For more options, visit this group at http://groups.google.com/group/sqlalchemy?hl=en -~--~~~~--~~--~--~---
[sqlalchemy] Possible to detect if a mapper property has been loaded or not?
I've got a sort of odd situation where I would like to detect if a (lazy) relation property of a mapper has been already loaded or not without triggering an automatic lazy load. Is it possible to do that? Thanks, Rick --~--~-~--~~~---~--~~ You received this message because you are subscribed to the Google Groups sqlalchemy group. To post to this group, send email to sqlalchemy@googlegroups.com To unsubscribe from this group, send email to [EMAIL PROTECTED] For more options, visit this group at http://groups.google.com/group/sqlalchemy?hl=en -~--~~~~--~~--~--~---
[sqlalchemy] Re: Possible to detect if a mapper property has been loaded or not?
i think if you just look in the object's __dict__ for the attribute and its not there, the lazy load didnt happen yet. On Mar 19, 2007, at 1:04 PM, Rick Morrison wrote: I've got a sort of odd situation where I would like to detect if a (lazy) relation property of a mapper has been already loaded or not without triggering an automatic lazy load. Is it possible to do that? Thanks, Rick --~--~-~--~~~---~--~~ You received this message because you are subscribed to the Google Groups sqlalchemy group. To post to this group, send email to sqlalchemy@googlegroups.com To unsubscribe from this group, send email to [EMAIL PROTECTED] For more options, visit this group at http://groups.google.com/group/sqlalchemy?hl=en -~--~~~~--~~--~--~---
[sqlalchemy] Re: Possible to detect if a mapper property has been loaded or not?
Yep, that seems to work, thanks Mike. On 3/19/07, Michael Bayer [EMAIL PROTECTED] wrote: i think if you just look in the object's __dict__ for the attribute and its not there, the lazy load didnt happen yet. On Mar 19, 2007, at 1:04 PM, Rick Morrison wrote: I've got a sort of odd situation where I would like to detect if a (lazy) relation property of a mapper has been already loaded or not without triggering an automatic lazy load. Is it possible to do that? Thanks, Rick --~--~-~--~~~---~--~~ You received this message because you are subscribed to the Google Groups sqlalchemy group. To post to this group, send email to sqlalchemy@googlegroups.com To unsubscribe from this group, send email to [EMAIL PROTECTED] For more options, visit this group at http://groups.google.com/group/sqlalchemy?hl=en -~--~~~~--~~--~--~---
[sqlalchemy] Re: cascading question
ok that ticket is fixed in the trunk
On Mar 19, 2007, at 5:54 AM, Alexandre CONRAD wrote:
from sqlalchemy import *
meta = DynamicMetaData()
# Sites
site_table = Table('sites', meta,
Column('id', Integer, primary_key=True),
Column('name', Unicode(20), nullable=False, unique=True),
Column('email', Unicode(100)),
)
class Site(object):
pass
site_mapper = mapper(Site, site_table,
order_by=site_table.c.name,
)
# Options
option_table = Table('options', meta,
Column('id', Integer, primary_key=True),
Column('name', Unicode(20), unique=True, nullable=False),
Column('description', Unicode(40)),
)
options_has_sites = Table('options_has_sites', meta,
Column('id_site', None, ForeignKey('sites.id'), primary_key=True),
Column('id_option', None, ForeignKey('options.id'),
primary_key=True),
)
class Option(object):
def __repr__(self):
return Option: %s % repr(self.name)
option_mapper = mapper(Option, option_table,
properties={
'sites':relation(Site, backref=options,
secondary=options_has_sites, cascade=save-update),
},
order_by=option_table.c.name,
)
meta.connect(sqlite://, echo=True)
meta.create_all()
# Make session.
session = create_session()
# Inject sites and options.
for i in range(1, 4):
o = Option()
o.name, o.description = opt%d % i, This is option %d % i
session.save(o)
s = Site()
s.name, s.email = site%d % i, [EMAIL PROTECTED] % i
session.save(s)
session.flush()
session.clear()
print \n### Now, let's query for site 1.
s = session.query(Site).get(1)
opts = session.query(Option).select()
print \n### Add options from 1 to 3 to the site.
s.options = opts[0:3] # Put option 1, 2 and 3
session.flush()
session.clear()
print \n### Now, let's query for site 1 again.
s = session.query(Site).get(1)
print \n### Check the site has the options 1, 2 and 3.
print s.options, !!!we should have opt 1, 2 and 3 here!!!
session.clear()
print \n### If it has, we should have pairs of (id_site, id_opt)
in options_has_sites.
print options_has_sites.select().execute().fetchall(), !!!we
should have 3 pairs here!!!
print \n### Now, let's query for option 1.
o = session.query(Option).get(1)
print \n### Now remove option 1.
session.delete(o)
session.flush()
session.clear()
print \n### Now, let's query for site 1 again.
s = session.query(Site).get(1)
print \n### Check what options has the site. Option 1 should be
removed from sites.
print s.options, !we should only have opt2 and opt3 in
here!
session.clear()
print \n### Now check that the row (id_site, id_opt) for option
1 should be removed from table options_has_sites.
print options_has_sites.select().execute().fetchall(), !!!we
should only have 2 pairs now that opt1 was removed!!!
print \n### Let's query for site 1.
s = session.query(Site).get(1)
print \n### Now let's delete the site.
session.delete(s)
session.flush()
session.clear()
print \n### We should still have option 2 and 3, even if the
site was deleted.
print session.query(Option).select()
print \n### The rows (id_site, id_opt) for option 2 and 3 of the
site should be removed from options_has_sites.
print options_has_sites.select().execute().fetchall(), !We
want this cleaned up and empty now that we have removed the site.
How to do that ??!
--~--~-~--~~~---~--~~
You received this message because you are subscribed to the Google Groups
sqlalchemy group.
To post to this group, send email to sqlalchemy@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at
http://groups.google.com/group/sqlalchemy?hl=en
-~--~~~~--~~--~--~---
[sqlalchemy] Re: pyodbc and tables with triggers
Sorry for the delay ;-) No objections. I was just trying to figure out
what was happening. (didn't get at first that module selection is a
bit ugly also meant that it didn't work if you specified the module)
But you already have a patch it seems. I'll try and test it out.
Steven
On 15 mrt, 15:43, Rick Morrison [EMAIL PROTECTED] wrote:
Sorry, Stephen, I replied too early; your second email arrived before the
first. A whole day before the first.
So until we get a real cleanup, you're looking to try modules in this order:
['pyodbc', 'adodbapi', 'pymssql']
Sounds OK to me -- any objections out there?
Rick
On 3/14/07, Rick Morrison [EMAIL PROTECTED] wrote:
It's the second case, that is, it sniffs out what modules are installed.
As I said before, this
(along with other modules that effectively do the same thing), is up for a
clean-up soon, see ticket #480.
Rick
On 3/14/07, polaar [EMAIL PROTECTED] wrote:
{'pyodbc': use_pyodbc, 'adodbapi': use_adodbapi, 'pyodbc':
use_pyodbc}.get(module.__name__, use_default)()
Sorry, should be pymssql instead of pyodbc twice, but I guess you got
that...
--~--~-~--~~~---~--~~
You received this message because you are subscribed to the Google Groups
sqlalchemy group.
To post to this group, send email to sqlalchemy@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at
http://groups.google.com/group/sqlalchemy?hl=en
-~--~~~~--~~--~--~---
[sqlalchemy] Re: server_side_cursors
OK, ive added ticket #514 to remind me to evaluate/test the patch
out..will try to commit by tomorrow.
On Mar 18, 2007, at 3:16 PM, Steve Huffman wrote:
Here's a patch that add's a DelayedMetaResultProxy(ResultProxy) class.
In the case of fetchone and fetchmany, it overwrites itself with the
ResultProxy's version after the first call. Actually, fetchone calls
the super class's initially and simply sets the metadata before
returning. I would have done that for all of the affected functions
with a decorator, but the close() call in the others causes trouble.
On 3/17/07, Michael Bayer [EMAIL PROTECTED] wrote:
no its a psycopg thing, its like this:
# server side cursor (giving it a name makes it server side.
psycopg2 devs think thats a good API.)
cursor = connection.cursor(x)
# execute. cursor now has pending results.
cursor.execute(select * from table)
SQLAlchemy's result wrapper, ResultProxy, then calls:
metadata = cursor.metadata
in order to get information about the columns in the result set. but
*this fails*, because we are using server side cursors, and
metadata apparently does not know to fetch its info from the
server. doesnt happen on my machine. not sure if this changes due
to psycopg2 versions, PG setup, or what. if we can determine its a
psycopg2 version issue, then everyone can just upgrade.
anyway what psycopg wants you to do is:
row = cursor.fetchone()
metadata = cursor.metadata
then it works. a tweaked out ResultProxy is needed to handle this
extra complexity in this case.
if youre saying we shouldnt use metadata at all, that doesnt work
because ResultProxy is globally used for all results regardless of
their source, textual, column-based, and otherwise. also the
databases will often not give you back the same column name as what
you gave it (case conventions, etc) and in some cases we dont even
have a column name to start with (like select some_function()).
On Mar 17, 2007, at 8:41 PM, Steve Huffman wrote:
I may be missing something fundamental here, but why doesn't it
already know the metadata since I defined the columns in which I'm
interested?
thing_table = sa.Table(thing, md, sa.Column('id', sa.Integer,
primary_key = True))
On 3/17/07, Michael Bayer [EMAIL PROTECTED] wrote:
the cursor metadata often cannot be read until fetchone() is
called
first. the current result set implementation we have doesnt call
fetchone() before it tries to get the metadata, and normally it
shouldnt (since the result set doesnt even know if its the result
of a
select/insert/whatever). id like an alternate result set class
to go
into effect when PG/server side cursors/select is used to do
this, i
think someone was supposed to send a patch. its hard for me to
develop since my version of PG 8.1 doesnt seem to reproduce the
issue.
On Mar 17, 8:14 pm, [EMAIL PROTECTED]
[EMAIL PROTECTED] wrote:
I was excited to see the server_side_cursors option that was added
recently.
I saw the reports of it not working with autoload = True, but I've
been having trouble getting it to work at all.
When attempting to select a row using:
t2.select().execute().fetchone()
I get:
INFO sqlalchemy.engine.base.Engine.0x..d0 SELECT thing.id FROM
thing
INFO sqlalchemy.engine.base.Engine.0x..d0 {}
Traceback (most recent call last): File stdin, line 1, in ?
File build/bdist.linux-x86_64/egg/sqlalchemy/engine/base.py,
line
811, in __repr__
File build/bdist.linux-x86_64/egg/sqlalchemy/engine/base.py,
line
671, in _get_col
File build/bdist.linux-x86_64/egg/sqlalchemy/engine/base.py,
line
659, in _convert_key
sqlalchemy.exceptions.NoSuchColumnError: Could not locate
column in
row for column '0'
This query runs fine without server_side_cursors = True
Any suggestions?
postgres_delay_metadata.patch
--~--~-~--~~~---~--~~
You received this message because you are subscribed to the Google Groups
sqlalchemy group.
To post to this group, send email to sqlalchemy@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at
http://groups.google.com/group/sqlalchemy?hl=en
-~--~~~~--~~--~--~---
[sqlalchemy] How to model an odd many-to-many relationship?
I have a many-to-many relationship between two entities, call them A and B. This is easy enough to model in sqlalchemy using Elixir. However, to complicate things, I need an integer column somewhere called 'priority'. In the relationship between an A and multiple Bs, I want the Bs to be ordered by the value of the 'priority' column. In the relationship between a B and multiple As, the value of 'priority' is irrelevant. The problem is, I don't know what entity needs to have the 'priority' column in it, or how to model this in the relationship between A and B. I thought that maybe the 'priority' column should be in the secondary table that handles the many-to-many relationship between A and B, but I'm not sure how to set that up, and I'm not sure that Elixir can handle that at all. I'd appreciate it if someone can point me in the right direction on this. --~--~-~--~~~---~--~~ You received this message because you are subscribed to the Google Groups sqlalchemy group. To post to this group, send email to sqlalchemy@googlegroups.com To unsubscribe from this group, send email to [EMAIL PROTECTED] For more options, visit this group at http://groups.google.com/group/sqlalchemy?hl=en -~--~~~~--~~--~--~---
[sqlalchemy] Re: Issue using rev 2425
OK the bad thing is that theres no way ppl are going to be able to
figure out a relation like this one unless I add crazy amounts of
docs...anyway, the answer is:
mapper(Section, section_table, properties = {
'items': relation(Item, backref='section'),
'keywords':relation(Keyword,
primaryjoin=and_(
keyword_table.c.uuid==itemkeyword_table.c.keyword_uuid,
item_table.c.uuid==itemkeyword_table.c.item_uuid,
section_table.c.id==item_table.c.section_id),
viewonly=True, foreign_keys=[keyword_table.c.uuid], remote_side=
[item_table.c.section_id])
}
)
the foreignkey argument is deprecated, and in the trunk the entire
system of calculating what does it mean when we say A relates to B
has been highly clarified and strictified, as well as the way it
calculates the lazy clause which is the part that was breaking for
you. that logic doesnt find the things it wants to in that big join
condition you gave it since it cant reconcile section_table against
keyword_table in such a way that it also knows where to put the ?
for the lazy clause (in previous versions, it made a very good guess,
which guessed right for things like the above but guessed wrong for a
lot of other stuff. now it doesnt guess so much). of course the
improvment to make here is to antipicate that particular error and
raise a nicer message earlier on.
anyway in the above, you give it the foreign key (or however many you
want to describe) in foreign_keys, and then to tell the lazy loader
which columns to put the ? in you use remote_side. the above values
are a little artificial to make the right thing happen.
alternatively, you can just force your own lazy loader, perhaps I
should add a recipe for this...its just as functional (except wont
work for an eager load) and is much more blunt in its intention:
from sqlalchemy.orm.session import attribute_manager
def load_keywords(instance):
def lazyload():
session = object_session(instance)
return session.query(Keyword).select(and_(
keyword_table.c.uuid==itemkeyword_table.c.keyword_uuid,
item_table.c.uuid==itemkeyword_table.c.item_uuid,
instance.id==item_table.c.section_id))
return lazyload
attribute_manager.register_attribute(Section, 'keywords',
uselist=True, callable_=load_keywords)
of course thats a variant on whats in the docs, which is just to use
property to make your own function (which IMHO makes the viewonly
flag not really needed).
On Mar 19, 2007, at 9:56 PM, Steve Zatz wrote:
import uuid as uuid_
from sqlalchemy import *
def get_uuid():
a = str(uuid_.uuid4())
a = a.strip('{}')
return a.replace('-','')
engine = create_engine('sqlite://')
metadata = BoundMetaData(engine)
engine.echo = True #True
item_table = Table('item',metadata,
Column('id', Integer, primary_key=True),
Column('uuid',String(32), unique=True, nullable=False),
Column('parent_uuid', String(32), ForeignKey
('item.uuid'), nullable=True),
Column('name',String(150)),
Column('section_id', Integer, ForeignKey('section.id'))
)
section_table = Table('section', metadata,
Column('id', Integer, primary_key=True),
Column('name', String(25), unique=True,
nullable=False),
)
keyword_table = Table('keyword', metadata,
Column('uuid',String(32), primary_key=True),
Column('name', String(25), unique=True,
nullable=False)
)
itemkeyword_table = Table('item_keyword', metadata,
Column('item_uuid', String(32),ForeignKey
('item.uuid'), primary_key=True),
Column('keyword_uuid', String(32), ForeignKey
('keyword.uuid'), primary_key=True),
)
metadata.create_all()
# class definitions
class Item(object):
def __init__(self, name=None, id=None, uuid=None, **kw): #?
**kw for x,y in kw, setattr(x) = y
self.name = name
self.id = id
if uuid:
self.uuid = uuid
else:
self.uuid = get_uuid()
for k in kw:
setattr(self, k, kw[k])
def _get_keywords(self):
return [ik.keyword for ik in self.itemkeywords]
keywords = property(_get_keywords)
class Section(object):
def __init__(self, name=None, id=None):
self.name = name
self.id = id
class Keyword(object):
def __init__(self, name=None, id=None, uuid=None):
self.name = name
self.id = id
if uuid:
self.uuid = uuid
else:
self.uuid = get_uuid()
class ItemKeyword(object):
def __init__(self, keyword=None):
if keyword:
self.keyword = keyword
mapper(Section, section_table, properties = {'items': relation
(Item, backref='section'), 'keywords':relation(Keyword,
[sqlalchemy] Replicating a transaction
Hi, I'm wondering if it would be possible to have a situation where transactions coming in from the ORM system could be sent to multiple database servers at once. If I were to look into adding this, where would be the best place to start? Many thanks. -- Benno Rice [EMAIL PROTECTED] http://jeamland.net/ --~--~-~--~~~---~--~~ You received this message because you are subscribed to the Google Groups sqlalchemy group. To post to this group, send email to sqlalchemy@googlegroups.com To unsubscribe from this group, send email to [EMAIL PROTECTED] For more options, visit this group at http://groups.google.com/group/sqlalchemy?hl=en -~--~~~~--~~--~--~---
