hi mike,
Michael Bayer wrote:
> hi Kapil -
>
> are you wanting to have multiple copies of history in
> prophistory/historyarraylist ? right now its just "saved version" /
> "new version".
yup, basically on a savepoint operation, it would take a copy.
>
> attribute rollback itself is a pretty
i'm also working on zope/plone integration though at a different layer
then florent, i'm focusing more on mapping to the plone application
schema model, then the zodb object storage. to be opensourced, probably
another 2 weeks.
diving in to hand out 2 cents ;-)
Michael Bayer wrote:
On Feb 2
hi folks,
one of the things i'm interested in doing, is supporting savepoints for
supported databases that support such ( postgres, oracle, firebird). i'd
like to do to do this without clearing the objectstore identity map on a
savepoint rollback. in order to go about this i'd like to modify
Michael Bayer wrote:
On Mar 6, 2006, at 7:38 PM, Kapil Thangavelu wrote:
direction. it would need a table subclass and custom metaclass to get
around the existing singleton table metaclass.
Or you could have two SchemaEngine objects, a "current" and a "previous
version", one of which has
On Mar 6, 2006, at 7:38 PM, Kapil Thangavelu wrote:
direction. it would need a table subclass and custom metaclass to
get around the existing singleton table metaclass.
Or you could have two SchemaEngine objects, a "current" and a
"previous version", one of which has the collection of cu
i've playing with runtime sqlalchemy model management, and was thinking
another way to implement this without using a versioned schema, would be
by instead having a method on the schema/engine for autoloading existing
tables from the database and comparing to the in memory table objects,
and ge
On 3/6/06, Gambit <[EMAIL PROTECTED]> wrote:
So, I'm looking for suggestions on how I can eitherautomatically generate DDL changesets -- not failsafe, but ones I canthen scan for completeness and use as a platform -- or aid me inverifying that I have indeed identified all the changes made between
t
attached is a patch that creates 30 character max column labels,
patched against the latest SVN head. I actually just checked it in
so you could just "svn up" to get it, though im not totally confident
about it as I would have to add some "long column name" unit tests,
but im pretty sure t
I'm trying to use sqlalchemy 0.1.3 with Oracle 9i and cx_Oracle, but can't get
far.
I wrote Table and Mapper objects for an existing schema, It has fairly verbose
table and column identifiers, like SHAMPOO_CLASS.SHAMPOO_OR_CONDITIONER_TYPE
let's say.
Running a simple select() on the ShampooCl
Michael Bayer wrote:
luckily, it was a one liner, so i committed this along with the
examples (short one and the full one) as unit tests in changeset 1104.
Excellent !
Wow, today is already significantly better than yesterday (ahem, where I
attempted to set fire to the kitchen) and I h
Il Sun, 5 Mar 2006 16:42:45 -0500, Michael Bayer ha scritto:
> note that these defaults are "pre-insert" and "pre-update" and only
> correspond to a parameter value of None, so theyre not as flexible as
> a real SQL trigger, which could conceivable fire off any time, pre-
> or post- exec, et
Hi Michael -
Thanks for the interesting read.
I'm not looking for something that will necessarily require a
VersionedTable or the equivalent. The personal fear I'm working
against is that I will neglect to remember/record/identify a change
made in the schema while generating my DDL update sc
i am sure Jonathan will speak up on this one.
or ill just point you to the blog entry http://
spyced.blogspot.com/2006/02/why-schema-definition-belongs-in.html
As for me, I am not *so* opinionated as to say schema management
tools are inherently bad, although I probably wouldnt use them
Hi guys -
I'm looking at various tools for tracking schema changes on the
backend database and generate change scripts for user
upgrades. There's a handful of tools out there, most of them fairly
expensive, but I thought I'd ask if SQLAlchemy could be used for this
kind of purpose. With the
Ive checked this patch in changeset 1089, but I have not yet gotten a
chance to run my oracle server to test against (its usually a several
hours routine since then i have to get all the other unittests
running again, etc). if you want to test it, the unittest is
'engines.py', so run:
luckily, it was a one liner, so i committed this along with the
examples (short one and the full one) as unit tests in changeset 1104.
On Mar 6, 2006, at 1:31 PM, Michael Bayer wrote:
robert -
this is not a backref issue at all. this is, remarkably, an eager
loading bug that has been ther
robert -
this is not a backref issue at all. this is, remarkably, an eager
loading bug that has been there completely from day one ! the
invoice->company->address relationship is: invoice is many-to-one on
company, company is one-to-many on addresses. an eager load chain
all the way t
On 3/6/06, Michael Bayer <[EMAIL PROTECTED]> wrote:
john -how is this different froma.join(a_x_b).join(b).join(a_x_c).join(c).select().execute()?Man, shame on me. No difference. Well, difference is this work! :D
On Mar 6, 2006, at 11:56 AM, Shuo Yang wrote:>
--
john -
how is this different from
a.join(a_x_b).join(b).join(a_x_c).join(c).select().execute()
?
On Mar 6, 2006, at 11:56 AM, Shuo Yang wrote:
---
This SF.Net email is sponsored by xPML, a groundbreaking scripting language
that
robert -
SQLite doenst like the LIMIT without an OFFSET that occurs when you
do get_by. I added an OFFSET 0 to the sqlite module when theres a
LIMIT, now your example works.
- mike
On Mar 5, 2006, at 5:13 PM, Robert Leftwich wrote:
Perhaps I'm misunderstanding what select_by and get_by
of course you are not misunderstanding them. have you tried this test with any other database besides SQLite ? Im looking at the two queries from get() and get_by(), one of which uses a limit, the queries/result sets are as follows:TEST ONE[2006-03-06 02:43:58,008] [engine]: SELECT users.user_nam
Hi:I'm trying to make a patch to make join work with many-to-many relationships via an association table. This works for something like: a join b, but is having ambiguous column name problems if you attempt to do something like: a join b join c.
Patch attached as well as a test script illustrating
22 matches
Mail list logo
