On Tuesday, December 27, 2011 7:22:03 AM UTC-8, Michael Bayer wrote:
On Dec 27, 2011, at 4:54 AM, Arturo Sevilla wrote:
Thanks for the code, your example runs without a problem.
I've managed to locate the error. In reality it doesn't have to do
with InstrumentedList. The problem
Ah, OK so what's happening is that _sa_instance_state hasn't yet been
assigned to the User object during the unpickle process, then your __eq__()
is trying to get at self.id which triggers the attribute system and
requires a fully composed User object.
Here you'd need to either have a
Nope, the problem persists, but not with the unpickle process but after it
with composite attributes. For example, I have the first_name attribute on
User, but also contact which is a MutableComposite. If I do the following
(after the unpickle process):
assert user.first_name = 'name'
I don't
On Friday, December 30, 2011 8:26:45 PM UTC-8, Michael Bayer wrote:
On Dec 30, 2011, at 7:46 PM, Arturo Sevilla wrote:
Nope, the problem persists, but not with the unpickle process but after it
with composite attributes. For example, I have the first_name attribute on
User, but also
Thanks for the code, your example runs without a problem.
I've managed to locate the error. In reality it doesn't have to do with
InstrumentedList. The problem occurs when beaker pickle.load()s from its
cache file (a pickled object in a file).
I get the following exception (which makes beaker
I have an error that occurs after an upgrade to 0.7.4 (from 0.6.8). I use
classical mapping but I'm having the following exception when I add it to a
session and save() it:
TypeError: can't pickle function objects
If I directly do pickle.dumps(obj) I get the following exception:
Hi,
I think I have detected two bugs for PostgreSQL databases. I don't think is
a different behavior between 0.6 and 0.7 as there is no problem with sqlite.
In the following code I have a very simple model, which contains a helper
pre-process list class, which just transforms strings into the
As a follow up to this thread, I've seen in the code in version 0.7.2 that
comparator_factory is no longer working for CompositeProperty. Is this a
bug?
Reference:
http://www.sqlalchemy.org/trac/browser/lib/sqlalchemy/orm/descriptor_props.py#L75
You can see that in the constructor there is no
I will see if later today I can provide some tests, or by tomorrow. I think
the basic rationale is that a composite property should behave like a
regular mapped property, and those have a comparator_factory, although I
know this is probable a lame argument.
--
You received this message
Hi,
I have a mapping between a table row and a python field, for which I have a
synonym. I've read the documentation and it says: Each of
column_property(), composite(), relationship(), and comparable_property()
accept an argument called comparator_factory. For which I assume that
synonym()
Hi,
I've been trying on the new futures of composite classes of SQLAlchemy 0.7,
and I don't have any problems with the read-only default mapping. However,
just when I inherit from MutableComposite I start to get an AttributeError
exception when I set the value to None (AttributeError: type
Hi thanks, for answering. Yes I'm aware that setting it to a
non-ContactInformation object will set off the coerce. However, if you put
None to it also sets off the error, something which does not happen when you
just have a non-mutable composite property. Should this be correct
behavior?
Do you mean in 0.7.1 or in the snapshot you put here?
--
You received this message because you are subscribed to the Google Groups
sqlalchemy group.
To view this discussion on the web visit
https://groups.google.com/d/msg/sqlalchemy/-/cT683Za9q90J.
To post to this group, send email to
Great! It worked, I was going to suggest to impolemente coerce() in
MutableComposite, but that's just what the snapshot has :)
Do you know in which minor version should I expect to see this patch?
Because while I can test it and make it work for my machine it would be
better to have the
Nevermind I just realized that it says 0.7.2 in the CHANGE file.
--
You received this message because you are subscribed to the Google Groups
sqlalchemy group.
To view this discussion on the web visit
https://groups.google.com/d/msg/sqlalchemy/-/ri_r-8ywDlkJ.
To post to this group, send email
My only doubt now is whether you can query over the fields of the composite
attribute. In the documentation an example is made of a query with a
comparison of the whole data structure:
session.query(Vertex).filter(Vertex.start == Point(3, 4))
But it also says: As of SQLAlchemy 0.7, composites
On Jan 10, 12:44 pm, Michael Bayer mike...@zzzcomputing.com wrote:
On Jan 10, 2011, at 2:21 PM, Arturo Sevilla wrote:
Hello,
Thanks again for the quick reply!
I tried to isolate all the mapper columns to try to make it less
confusing, now I know that was not a good idea
Hello,
I'm trying to do the following:
'home': orm.composite(
Address,
user.c.HomeAddress_Street,
# The following column is a UUID but is a foreign key to a
mapped
# table in SQLAlchemy, ideally would be to say
relationship(City)
, Arturo Sevilla wrote:
Hello,
I'm trying to do the following:
'home': orm.composite(
Address,
user.c.HomeAddress_Street,
# The following column is a UUID but is a foreign key to a
mapped
# table in SQLAlchemy, ideally would be to say
Hello,
I've been trying to do unit testing with SQLAlchemy and PostgreSQL. In
my model/tables I have two entities state and country, state has a
foreign key relating to country (no problems there).
In my setUp() method I call create_all() for my metadata while in
tearDown() I call drop_all().
of the
tearDown().
http://www.sqlalchemy.org/docs/orm/session.html#closing
On Jan 9, 2011, at 6:24 AM, Arturo Sevilla wrote:
Hello,
I've been trying to do unit testing with SQLAlchemy and PostgreSQL. In
my model/tables I have two entities state and country, state has a
foreign
21 matches
Mail list logo
