Re: [sqlalchemy] twophase error sqlalchemy
Ok, that was my fault.
I was calling rollback() on a sqla.orm.session, instead of calling
transaction.abort()
Although the documentation only forbids calling commit, it does not say
anything about rolling back an individual session :(
Thanks!
Pau.
2013/6/25 Michael Bayer mike...@zzzcomputing.com
looks like a zope.transaction issue to me, or at least one where they'd
have to show me how they need to use the Session. They seem to be calling
upon session.transaction.prepare() directly at a time when that
session.transaction is no longer associated with the parent session.
You'd need to take it up with the zope.transaction folks first.
On Jun 25, 2013, at 5:49 AM, Pau Tallada tall...@pic.es wrote:
Hi!
I think I've stepped on the same or similar bug.
I have a software which uses zodb transactions to synchronize operations
in two databases. In the end, the two databases are the same (we are using
the exact same postgresql connection url for both). The software runs fine
if only one instance is launched.
But... if I launch more than one instance at a time, there is a great
possibility that I get an Exception like this:
Traceback (most recent call last):
File
/nfs/pau/PAUdm/codes/tonello/ENV/packaging_test/lib/python2.6/site-packages/brownthrower/dispatcher/static/__init__.py,
line 238, in _run_job
transaction.commit()
File
/nfs/pau/PAUdm/codes/tonello/ENV/packaging_test/lib/python2.6/site-packages/transaction/_manager.py,
line 111, in commit
return self.get().commit()
File
/nfs/pau/PAUdm/codes/tonello/ENV/packaging_test/lib/python2.6/site-packages/transaction/_transaction.py,
line 280, in commit
reraise(t, v, tb)
File
/nfs/pau/PAUdm/codes/tonello/ENV/packaging_test/lib/python2.6/site-packages/transaction/_transaction.py,
line 271, in commit
self._commitResources()
File
/nfs/pau/PAUdm/codes/tonello/ENV/packaging_test/lib/python2.6/site-packages/transaction/_transaction.py,
line 417, in _commitResources
reraise(t, v, tb)
File
/nfs/pau/PAUdm/codes/tonello/ENV/packaging_test/lib/python2.6/site-packages/transaction/_transaction.py,
line 394, in _commitResources
rm.tpc_vote(self)
File
/nfs/pau/PAUdm/codes/tonello/ENV/packaging_test/lib/python2.6/site-packages/zope/sqlalchemy/datamanager.py,
line 153, in tpc_vote
self.tx.prepare()
File
/nfs/pau/PAUdm/codes/tonello/ENV/packaging_test/lib/python2.6/site-packages/sqlalchemy/orm/session.py,
line 313, in prepare
if self._parent is not None or not self.session.twophase:
AttributeError: 'NoneType' object has no attribute 'twophase'
I'm completely lost on this :(
Any clues?
BTW, sessions are created like this:
engine = create_engine('postgresql://foo')
session_maker = scoped_session(sessionmaker(
bind = engine,
twophase = True,
extension = ZopeTransactionExtension()
))
Thank you very much in advance :)
Pau.
2013/2/20 Michael Bayer mike...@zzzcomputing.com
you'd have to illustrate a working example of the code that causes that.
On Feb 20, 2013, at 6:18 AM, Christian Démolis
christiandemo...@gmail.com wrote:
Hi,
AttributeError: 'NoneType' object has no attribute 'twophase'
243.
244.
245.
246.
247.
248.
249.
250.
251.
252.
Session already has a Connection associated for the
given Connection's Engine)
else:
conn = bind.contextual_connect()
if self.session.twophase and self._parent is None:
transaction = conn.begin_twophase()
elif self.nested:
transaction = conn.begin_nested()
else:
Anybody understand this error ?
--
You received this message because you are subscribed to the Google Groups
sqlalchemy group.
To unsubscribe from this group and stop receiving emails from it, send an
email to sqlalchemy+unsubscr...@googlegroups.com.
To post to this group, send email to sqlalchemy@googlegroups.com.
Visit this group at http://groups.google.com/group/sqlalchemy?hl=en.
For more options, visit https://groups.google.com/groups/opt_out.
--
You received this message because you are subscribed to the Google Groups
sqlalchemy group.
To unsubscribe from this group and stop receiving emails from it, send an
email to sqlalchemy+unsubscr...@googlegroups.com.
To post to this group, send email to sqlalchemy@googlegroups.com.
Visit this group at http://groups.google.com/group/sqlalchemy?hl=en.
For more options, visit https://groups.google.com/groups/opt_out.
--
--
Pau Tallada Crespí
Dep. d'Astrofísica i Cosmologia
Port d'Informació Científica (PIC)
Tel: +34 93 586 8233
--
--
You received this message because you are subscribed to the Google Groups
sqlalchemy group.
To unsubscribe from this group and stop receiving emails from it, send an
email to
Re: [sqlalchemy] HSTORE serialize/de-serialize incorrectly handles backslashes
On Tue, Jun 06/25/13, 2013 at 02:47:18PM -0400, Michael Bayer wrote: On Jun 25, 2013, at 2:13 PM, Ryan Kelly rpkell...@gmail.com wrote: There are also some other parsing problems that I consider to be corner cases and broken as implemented in PostgreSQL, such as: (postgres@[local]:5432 14:05:43) [dev] select 'a=,b='::hstore; (postgres@[local]:5432 14:05:47) [dev] select 'a=, b='::hstore; (postgres@[local]:5432 14:06:45) [dev] select 'a= , b='::hstore; (postgres@[local]:5432 14:06:48) [dev] select 'a= ,b='::hstore; (postgres@[local]:5432 14:06:50) [dev] select 'a=,'::hstore; (postgres@[local]:5432 14:10:12) [dev] select ',=,'::hstore; None of which are parsed by SQLAlchemy but some of which are parsed by PostgreSQL. Posgresql or psycopg2 ? isn't the serialization here normally done natively if you're on a more recent psycopg2? Well these are really de-serialization. But yes, serialization and de-serialization are handled natively by psycopg2. In some circumstances we actually need to parse/write the values directly ourselves (usually when reading/writing files in the COPY format). Unfortunately, psycopg2's parser is much stricter than the documented format as it requires quoted keys (which in some ways make sense, as the server will never return anything unquoted) so none of the above examples work. -Ryan P. Kelly -- You received this message because you are subscribed to the Google Groups sqlalchemy group. To unsubscribe from this group and stop receiving emails from it, send an email to sqlalchemy+unsubscr...@googlegroups.com. To post to this group, send email to sqlalchemy@googlegroups.com. Visit this group at http://groups.google.com/group/sqlalchemy. For more options, visit https://groups.google.com/groups/opt_out.
[sqlalchemy] Unable to store datetime.datetime.max using SQLAlchemy==0.8.1 with the mysql-python==1.2.4 driver
Hi,
not sure if I should post this issue here or if its a mysql-python one but
perhaps someone with better knowledge can help me figure that out?
A more pretty print version of this issue is available
here:
http://stackoverflow.com/questions/17315422/unable-to-store-datetime-datetime-max-using-sqlalchemy-0-8-1-with-the-mysql-pyt
I've noticed a change in behavior for storing datetime.datetime.max via
`SQLAlchemy==SQLAlchemy==0.8.1` and going from `mysql-python==1.2.3` to
`mysql-python==1.2.4`. By only changing the driver from 1.2.3 to 1.2.4 I go
from being able to store to being unable to store it.
Where do I turn to for help in this matter? SQLAlchemy or mysql-python? Is
this expected behaviour or a bug or do I have a bad setup? I fear that a
change like this will break a lot of systems out there.
This is my SQLAlchemy setup:
from sqlalchemy import create_engine, Integer, DateTime, Column
from sqlalchemy.orm import sessionmaker
from sqlalchemy.ext.declarative import declarative_base
from datetime import datetime
engine = create_engine('mysql://root@localhost/test_database',
echo=True)
Base = declarative_base()
class User(Base):
__tablename__ = 'users'
id = Column(Integer, primary_key=True)
age = Column(DateTime, default=datetime.max)
Base.metadata.create_all(engine)
session = sessionmaker(bind=engine)()
u = User()
session.add(u)
session.commit()
I also have a virtualenv called test. This is what happens when I run the
code above.
(test)➜ ~ pip install MySQL-python==1.2.3
(test)➜ ~ python test.py
2013-06-26 10:29:18,885 INFO sqlalchemy.engine.base.Engine SELECT
DATABASE()
2013-06-26 10:29:18,885 INFO sqlalchemy.engine.base.Engine ()
2013-06-26 10:29:18,887 INFO sqlalchemy.engine.base.Engine SHOW
VARIABLES LIKE 'character_set%%'
2013-06-26 10:29:18,887 INFO sqlalchemy.engine.base.Engine ()
2013-06-26 10:29:18,891 INFO sqlalchemy.engine.base.Engine SHOW
VARIABLES LIKE 'sql_mode'
2013-06-26 10:29:18,891 INFO sqlalchemy.engine.base.Engine ()
2013-06-26 10:29:18,896 INFO sqlalchemy.engine.base.Engine DESCRIBE
`users`
2013-06-26 10:29:18,896 INFO sqlalchemy.engine.base.Engine ()
2013-06-26 10:29:18,904 INFO sqlalchemy.engine.base.Engine BEGIN
(implicit)
2013-06-26 10:29:18,905 INFO sqlalchemy.engine.base.Engine INSERT INTO
users (age) VALUES (%s)
2013-06-26 10:29:18,905 INFO sqlalchemy.engine.base.Engine
(datetime.datetime(, 12, 31, 23, 59, 59, 99),)
2013-06-26 10:29:18,908 INFO sqlalchemy.engine.base.Engine COMMIT
And the database (test_database) looks like this:
mysql select * from users;
++-+
| id | age |
++-+
| 1 | -12-31 23:59:59 |
++-+
1 row in set (0.00 sec)
This is my expected result so nothing strange here.
However, by simply switching the driver to mysql-python==1.2.4 I get this
result.
(test)➜ ~ pip install MySQL-python==1.2.4
(test)➜ ~ python test.py
2013-06-26 10:33:39,544 INFO sqlalchemy.engine.base.Engine SELECT
DATABASE()
2013-06-26 10:33:39,544 INFO sqlalchemy.engine.base.Engine ()
2013-06-26 10:33:39,546 INFO sqlalchemy.engine.base.Engine SHOW
VARIABLES LIKE 'character_set%%'
2013-06-26 10:33:39,546 INFO sqlalchemy.engine.base.Engine ()
2013-06-26 10:33:39,546 INFO sqlalchemy.engine.base.Engine SHOW
VARIABLES LIKE 'sql_mode'
2013-06-26 10:33:39,546 INFO sqlalchemy.engine.base.Engine ()
2013-06-26 10:33:39,547 INFO sqlalchemy.engine.base.Engine DESCRIBE
`users`
2013-06-26 10:33:39,547 INFO sqlalchemy.engine.base.Engine ()
2013-06-26 10:33:39,551 INFO sqlalchemy.engine.base.Engine BEGIN
(implicit)
2013-06-26 10:33:39,552 INFO sqlalchemy.engine.base.Engine INSERT INTO
users (age) VALUES (%s)
2013-06-26 10:33:39,552 INFO sqlalchemy.engine.base.Engine
(datetime.datetime(, 12, 31, 23, 59, 59, 99),)
/Users/pelle/.virtualenvs/test/lib/python2.7/site-packages/sqlalchemy/engine/default.py:324:
Warning: Datetime function: datetime field overflow
cursor.execute(statement, parameters)
/Users/pelle/.virtualenvs/test/lib/python2.7/site-packages/sqlalchemy/engine/default.py:324:
Warning: Out of range value for column 'age' at row 1
cursor.execute(statement, parameters)
2013-06-26 10:33:39,553 INFO sqlalchemy.engine.base.Engine COMMIT
And the database looks like this.
mysql select * from users;
++-+
| id | age |
++-+
| 1 | -00-00 00:00:00 |
++-+
1 row in set (0.00 sec)
So now all of the sudden I receive a warning `Warning: Datetime function:
datetime field overflow` and I end up with a nullable value in my database.
--
You received this message because you are subscribed to the Google Groups
Re: [sqlalchemy] HSTORE serialize/de-serialize incorrectly handles backslashes
On Jun 26, 2013, at 6:57 AM, Ryan Kelly rpkell...@gmail.com wrote: On Tue, Jun 06/25/13, 2013 at 02:47:18PM -0400, Michael Bayer wrote: On Jun 25, 2013, at 2:13 PM, Ryan Kelly rpkell...@gmail.com wrote: There are also some other parsing problems that I consider to be corner cases and broken as implemented in PostgreSQL, such as: (postgres@[local]:5432 14:05:43) [dev] select 'a=,b='::hstore; (postgres@[local]:5432 14:05:47) [dev] select 'a=, b='::hstore; (postgres@[local]:5432 14:06:45) [dev] select 'a= , b='::hstore; (postgres@[local]:5432 14:06:48) [dev] select 'a= ,b='::hstore; (postgres@[local]:5432 14:06:50) [dev] select 'a=,'::hstore; (postgres@[local]:5432 14:10:12) [dev] select ',=,'::hstore; None of which are parsed by SQLAlchemy but some of which are parsed by PostgreSQL. Posgresql or psycopg2 ? isn't the serialization here normally done natively if you're on a more recent psycopg2? Well these are really de-serialization. But yes, serialization and de-serialization are handled natively by psycopg2. In some circumstances we actually need to parse/write the values directly ourselves (usually when reading/writing files in the COPY format). Unfortunately, psycopg2's parser is much stricter than the documented format as it requires quoted keys (which in some ways make sense, as the server will never return anything unquoted) so none of the above examples work. OK just trying to get my head around the issue - psycopg2's parser does not have the backslashing issue right ? -Ryan P. Kelly -- You received this message because you are subscribed to the Google Groups sqlalchemy group. To unsubscribe from this group and stop receiving emails from it, send an email to sqlalchemy+unsubscr...@googlegroups.com. To post to this group, send email to sqlalchemy@googlegroups.com. Visit this group at http://groups.google.com/group/sqlalchemy. For more options, visit https://groups.google.com/groups/opt_out. -- You received this message because you are subscribed to the Google Groups sqlalchemy group. To unsubscribe from this group and stop receiving emails from it, send an email to sqlalchemy+unsubscr...@googlegroups.com. To post to this group, send email to sqlalchemy@googlegroups.com. Visit this group at http://groups.google.com/group/sqlalchemy. For more options, visit https://groups.google.com/groups/opt_out.
Re: [sqlalchemy] Unable to store datetime.datetime.max using SQLAlchemy==0.8.1 with the mysql-python==1.2.4 driver
On Jun 26, 2013, at 10:18 AM, Pelle Almquist pe...@wrapp.com wrote: Hi, not sure if I should post this issue here or if its a mysql-python one but perhaps someone with better knowledge can help me figure that out? A more pretty print version of this issue is available here: http://stackoverflow.com/questions/17315422/unable-to-store-datetime-datetime-max-using-sqlalchemy-0-8-1-with-the-mysql-pyt I've noticed a change in behavior for storing datetime.datetime.max via `SQLAlchemy==SQLAlchemy==0.8.1` and going from `mysql-python==1.2.3` to `mysql-python==1.2.4`. By only changing the driver from 1.2.3 to 1.2.4 I go from being able to store to being unable to store it. Where do I turn to for help in this matter? SQLAlchemy or mysql-python? Is this expected behaviour or a bug or do I have a bad setup? I fear that a change like this will break a lot of systems out there. Well it seems like MySQL-python doesn't like trying to store a date with the year here, it probably changed something about how it renders dates. So yeah you'd need to take it up with Python-MySQL. I would note that there are some good alternatives to MySQL-Python on the scene now, there's OurSQL, PyMySQL, and MySQL-connector-python seems to be working decently now, in case that helps. -- You received this message because you are subscribed to the Google Groups sqlalchemy group. To unsubscribe from this group and stop receiving emails from it, send an email to sqlalchemy+unsubscr...@googlegroups.com. To post to this group, send email to sqlalchemy@googlegroups.com. Visit this group at http://groups.google.com/group/sqlalchemy. For more options, visit https://groups.google.com/groups/opt_out.
[sqlalchemy] Declarative Sequences
Apologies if I'm posting a question to this group incorrectly or with bad
etiquette. I've been reading the emails from this group for a few years and now
have a question myself.
I am automating the creation of our models and business objects by reading a
schema definition from YAML. We use the declarative base, some metaclasses and
superclasses, a couple of mixes.. and ultimately PostgreSQL 9.x (exclusively)
I create a custom type using a metaclass that creates the primary key column
along with anything else, and decorate anything that needs `@declared_attr`
appropriately, for the declarative mechanism to discover it.
Long story short,
I would like to create custom sequences using Sequence() with a start value,
using a declared Column, and not having to construct a Table itself, as
described in the docs here
http://docs.sqlalchemy.org/en/latest/dialects/postgresql.html
I would like to do something like this:
class User(object):
id = Column(Integer, Sequence('user_id_seq', start=1),
primary_key=True)
This does work, and emits the CREATE SEQUENCE, but does not set the owned table
or the column to user.id
I get around this by using the default SERIAL, using DDL events, by updating
the normal ALTER SEQUENCE [user_id_seq] RESTART WITH [value].
However, I may have just missed something or misunderstood something deep in
declarative.
My goal is to automate as much of the schema, model, and business object
creation as possible using our DSL, which is leveraging the power of
SQLAlchemy, GeoAlchemy2, PostgreSQL, and PostGIS.
Much obliged for any help,
Mat
--
You received this message because you are subscribed to the Google Groups
sqlalchemy group.
To unsubscribe from this group and stop receiving emails from it, send an email
to sqlalchemy+unsubscr...@googlegroups.com.
To post to this group, send email to sqlalchemy@googlegroups.com.
Visit this group at http://groups.google.com/group/sqlalchemy.
For more options, visit https://groups.google.com/groups/opt_out.
Re: [sqlalchemy] Declarative Sequences
On Jun 26, 2013, at 12:02 PM, Mat Mathews m...@miga.me wrote:
I would like to do something like this:
class User(object):
id = Column(Integer, Sequence('user_id_seq', start=1),
primary_key=True)
This does work, and emits the CREATE SEQUENCE, but does not set the owned
table or the column to user.id
I'm not able to reproduce, even assigning the same sequence name to two
different tables simultaneously produces the correct result. Can you modify
the test below to illustrate your issue?
from sqlalchemy import *
from sqlalchemy.orm import *
from sqlalchemy.ext.declarative import declarative_base, declared_attr
Base = declarative_base()
class IdMixin(object):
id = Column(Integer, Sequence('user_id_seq', start=1), primary_key=True)
class A(IdMixin, Base):
__tablename__ = 'a'
class B(IdMixin, Base):
__tablename__ = 'b'
e = create_engine(postgresql://scott:tiger@localhost/test, echo=True)
Base.metadata.drop_all(e)
Base.metadata.create_all(e)
sess = Session(e)
a1 = A()
sess.add(a1)
b1 = B()
sess.add(b1)
sess.commit()
assert a1.id == 1
assert b1.id == 10001
--
You received this message because you are subscribed to the Google Groups
sqlalchemy group.
To unsubscribe from this group and stop receiving emails from it, send an email
to sqlalchemy+unsubscr...@googlegroups.com.
To post to this group, send email to sqlalchemy@googlegroups.com.
Visit this group at http://groups.google.com/group/sqlalchemy.
For more options, visit https://groups.google.com/groups/opt_out.
Re: [sqlalchemy] Declarative Sequences
Thanks for the quick reply.
If I modify the name of the sequence in the test, it reproduces what I have
experienced.
When I check the details on the sequence in postgres, there is no table owner
or column specified by the sequence 'test_user_id_seq'.. and I would expect to
see both tables `a` and `b`.
class IdMixin(object):
id = Column(Integer, Sequence('test_user_id_seq', start=1),
primary_key=True)
I will write a complete self contained test, that anyone can run. But wanted to
give my immediate feedback.
Thanks so much,
Mat
On Jun 26, 2013, at 6:37 PM, Michael Bayer mike...@zzzcomputing.com wrote:
On Jun 26, 2013, at 12:02 PM, Mat Mathews m...@miga.me wrote:
I would like to do something like this:
class User(object):
id = Column(Integer, Sequence('user_id_seq', start=1),
primary_key=True)
This does work, and emits the CREATE SEQUENCE, but does not set the owned
table or the column to user.id
I'm not able to reproduce, even assigning the same sequence name to two
different tables simultaneously produces the correct result. Can you modify
the test below to illustrate your issue?
from sqlalchemy import *
from sqlalchemy.orm import *
from sqlalchemy.ext.declarative import declarative_base, declared_attr
Base = declarative_base()
class IdMixin(object):
id = Column(Integer, Sequence('user_id_seq', start=1),
primary_key=True)
class A(IdMixin, Base):
__tablename__ = 'a'
class B(IdMixin, Base):
__tablename__ = 'b'
e = create_engine(postgresql://scott:tiger@localhost/test, echo=True)
Base.metadata.drop_all(e)
Base.metadata.create_all(e)
sess = Session(e)
a1 = A()
sess.add(a1)
b1 = B()
sess.add(b1)
sess.commit()
assert a1.id == 1
assert b1.id == 10001
--
You received this message because you are subscribed to the Google Groups
sqlalchemy group.
To unsubscribe from this group and stop receiving emails from it, send an
email to sqlalchemy+unsubscr...@googlegroups.com.
To post to this group, send email to sqlalchemy@googlegroups.com.
Visit this group at http://groups.google.com/group/sqlalchemy.
For more options, visit https://groups.google.com/groups/opt_out.
--
You received this message because you are subscribed to the Google Groups
sqlalchemy group.
To unsubscribe from this group and stop receiving emails from it, send an email
to sqlalchemy+unsubscr...@googlegroups.com.
To post to this group, send email to sqlalchemy@googlegroups.com.
Visit this group at http://groups.google.com/group/sqlalchemy.
For more options, visit https://groups.google.com/groups/opt_out.
Re: [sqlalchemy] Declarative Sequences
On Jun 26, 2013, at 1:26 PM, Mat Mathews m...@miga.me wrote:
Thanks for the quick reply.
If I modify the name of the sequence in the test, it reproduces what I have
experienced.
When I check the details on the sequence in postgres, there is no table owner
or column specified by the sequence 'test_user_id_seq'.. and I would expect
to see both tables `a` and `b`.
class IdMixin(object):
id = Column(Integer, Sequence('test_user_id_seq', start=1),
primary_key=True)
I will write a complete self contained test, that anyone can run. But wanted
to give my immediate feedback.
oh, you're looking for a linkage in PG's information schema I guess?
Sequence() doesn't have that functionality. My understanding was that PG's
SERIAL created the linkage as a server default, so you can get this by adding
this server default yourself, see below for demo:
from sqlalchemy import *
from sqlalchemy.orm import *
from sqlalchemy.ext.declarative import declarative_base, declared_attr
Base = declarative_base()
class IdMixin(object):
id = Column(Integer,
Sequence('some_id_seq', start=1),
server_default=text(nextval('some_id_seq')),
primary_key=True)
class A(IdMixin, Base):
__tablename__ = 'a'
class B(IdMixin, Base):
__tablename__ = 'b'
e = create_engine(postgresql://scott:tiger@localhost/test, echo=True)
Base.metadata.drop_all(e)
Base.metadata.create_all(e)
sess = Session(e)
a1 = A()
sess.add(a1)
b1 = B()
sess.add(b1)
sess.commit()
assert a1.id == 1
assert b1.id == 10001
sess.execute(INSERT INTO b DEFAULT VALUES)
assert sess.execute(SELECT * FROM b WHERE id=10002).scalar()
Thanks so much,
Mat
On Jun 26, 2013, at 6:37 PM, Michael Bayer mike...@zzzcomputing.com wrote:
On Jun 26, 2013, at 12:02 PM, Mat Mathews m...@miga.me wrote:
I would like to do something like this:
class User(object):
id = Column(Integer, Sequence('user_id_seq', start=1),
primary_key=True)
This does work, and emits the CREATE SEQUENCE, but does not set the owned
table or the column to user.id
I'm not able to reproduce, even assigning the same sequence name to two
different tables simultaneously produces the correct result. Can you modify
the test below to illustrate your issue?
from sqlalchemy import *
from sqlalchemy.orm import *
from sqlalchemy.ext.declarative import declarative_base, declared_attr
Base = declarative_base()
class IdMixin(object):
id = Column(Integer, Sequence('user_id_seq', start=1),
primary_key=True)
class A(IdMixin, Base):
__tablename__ = 'a'
class B(IdMixin, Base):
__tablename__ = 'b'
e = create_engine(postgresql://scott:tiger@localhost/test, echo=True)
Base.metadata.drop_all(e)
Base.metadata.create_all(e)
sess = Session(e)
a1 = A()
sess.add(a1)
b1 = B()
sess.add(b1)
sess.commit()
assert a1.id == 1
assert b1.id == 10001
--
You received this message because you are subscribed to the Google Groups
sqlalchemy group.
To unsubscribe from this group and stop receiving emails from it, send an
email to sqlalchemy+unsubscr...@googlegroups.com.
To post to this group, send email to sqlalchemy@googlegroups.com.
Visit this group at http://groups.google.com/group/sqlalchemy.
For more options, visit https://groups.google.com/groups/opt_out.
--
You received this message because you are subscribed to the Google Groups
sqlalchemy group.
To unsubscribe from this group and stop receiving emails from it, send an
email to sqlalchemy+unsubscr...@googlegroups.com.
To post to this group, send email to sqlalchemy@googlegroups.com.
Visit this group at http://groups.google.com/group/sqlalchemy.
For more options, visit https://groups.google.com/groups/opt_out.
--
You received this message because you are subscribed to the Google Groups
sqlalchemy group.
To unsubscribe from this group and stop receiving emails from it, send an email
to sqlalchemy+unsubscr...@googlegroups.com.
To post to this group, send email to sqlalchemy@googlegroups.com.
Visit this group at http://groups.google.com/group/sqlalchemy.
For more options, visit https://groups.google.com/groups/opt_out.
Re: [sqlalchemy] Declarative Sequences
That absolutely solves my issue. I had a serious suspicion I was missing
something, and it was to look at the Column options, not just the Sequence.
Thanks a ton.
On Jun 26, 2013, at 7:44 PM, Michael Bayer mike...@zzzcomputing.com wrote:
On Jun 26, 2013, at 1:26 PM, Mat Mathews m...@miga.me wrote:
Thanks for the quick reply.
If I modify the name of the sequence in the test, it reproduces what I have
experienced.
When I check the details on the sequence in postgres, there is no table
owner or column specified by the sequence 'test_user_id_seq'.. and I would
expect to see both tables `a` and `b`.
class IdMixin(object):
id = Column(Integer, Sequence('test_user_id_seq', start=1),
primary_key=True)
I will write a complete self contained test, that anyone can run. But wanted
to give my immediate feedback.
oh, you're looking for a linkage in PG's information schema I guess?
Sequence() doesn't have that functionality. My understanding was that PG's
SERIAL created the linkage as a server default, so you can get this by
adding this server default yourself, see below for demo:
from sqlalchemy import *
from sqlalchemy.orm import *
from sqlalchemy.ext.declarative import declarative_base, declared_attr
Base = declarative_base()
class IdMixin(object):
id = Column(Integer,
Sequence('some_id_seq', start=1),
server_default=text(nextval('some_id_seq')),
primary_key=True)
class A(IdMixin, Base):
__tablename__ = 'a'
class B(IdMixin, Base):
__tablename__ = 'b'
e = create_engine(postgresql://scott:tiger@localhost/test, echo=True)
Base.metadata.drop_all(e)
Base.metadata.create_all(e)
sess = Session(e)
a1 = A()
sess.add(a1)
b1 = B()
sess.add(b1)
sess.commit()
assert a1.id == 1
assert b1.id == 10001
sess.execute(INSERT INTO b DEFAULT VALUES)
assert sess.execute(SELECT * FROM b WHERE id=10002).scalar()
Thanks so much,
Mat
On Jun 26, 2013, at 6:37 PM, Michael Bayer mike...@zzzcomputing.com wrote:
On Jun 26, 2013, at 12:02 PM, Mat Mathews m...@miga.me wrote:
I would like to do something like this:
class User(object):
id = Column(Integer, Sequence('user_id_seq', start=1),
primary_key=True)
This does work, and emits the CREATE SEQUENCE, but does not set the owned
table or the column to user.id
I'm not able to reproduce, even assigning the same sequence name to two
different tables simultaneously produces the correct result. Can you
modify the test below to illustrate your issue?
from sqlalchemy import *
from sqlalchemy.orm import *
from sqlalchemy.ext.declarative import declarative_base, declared_attr
Base = declarative_base()
class IdMixin(object):
id = Column(Integer, Sequence('user_id_seq', start=1),
primary_key=True)
class A(IdMixin, Base):
__tablename__ = 'a'
class B(IdMixin, Base):
__tablename__ = 'b'
e = create_engine(postgresql://scott:tiger@localhost/test, echo=True)
Base.metadata.drop_all(e)
Base.metadata.create_all(e)
sess = Session(e)
a1 = A()
sess.add(a1)
b1 = B()
sess.add(b1)
sess.commit()
assert a1.id == 1
assert b1.id == 10001
--
You received this message because you are subscribed to the Google Groups
sqlalchemy group.
To unsubscribe from this group and stop receiving emails from it, send an
email to sqlalchemy+unsubscr...@googlegroups.com.
To post to this group, send email to sqlalchemy@googlegroups.com.
Visit this group at http://groups.google.com/group/sqlalchemy.
For more options, visit https://groups.google.com/groups/opt_out.
--
You received this message because you are subscribed to the Google Groups
sqlalchemy group.
To unsubscribe from this group and stop receiving emails from it, send an
email to sqlalchemy+unsubscr...@googlegroups.com.
To post to this group, send email to sqlalchemy@googlegroups.com.
Visit this group at http://groups.google.com/group/sqlalchemy.
For more options, visit https://groups.google.com/groups/opt_out.
--
You received this message because you are subscribed to the Google Groups
sqlalchemy group.
To unsubscribe from this group and stop receiving emails from it, send an
email to sqlalchemy+unsubscr...@googlegroups.com.
To post to this group, send email to sqlalchemy@googlegroups.com.
Visit this group at http://groups.google.com/group/sqlalchemy.
For more options, visit https://groups.google.com/groups/opt_out.
--
You received this message because you are subscribed to the Google Groups
sqlalchemy group.
To unsubscribe from this group and stop receiving emails from it, send an email
to sqlalchemy+unsubscr...@googlegroups.com.
To post to this group, send email to sqlalchemy@googlegroups.com.
Visit this group at http://groups.google.com/group/sqlalchemy.
For more options, visit https://groups.google.com/groups/opt_out.
