Hi,
It's been some time since this topic was created. Has anything changed on
that matter or manually setting column.server_default=FetchedValue() is
still the best way to do it?
Kind regards,
Michał
On Thursday, September 15, 2011 7:58:49 AM UTC+2, Matt Bodman wrote:
Hi,
I am
hey guys,
can’t figure out how to properly construct sqla core delete() construct for
something that should look like (I’m using MySQL):
DELETE FROM a USING a
JOIN b ON (a.id1 = b.id1)
JOIN c ON (a.id2 = c.id2)
WHERE . . .
I can write in it literal SQL and feed to the execute() of course but I
probably (maybe we should improve on our end, though). but when you’re
autoloading, you can set this default up automatically using the column_reflect
event:
http://docs.sqlalchemy.org/en/rel_0_9/core/events.html?highlight=column_reflect#sqlalchemy.events.DDLEvents.column_reflect
we are very slowly getting around to supporting various edge case syntaxes like
these, this one is:
https://bitbucket.org/zzzeek/sqlalchemy/issue/959/support-mysql-delete-from-join
https://bitbucket.org/zzzeek/sqlalchemy/issue/959/support-mysql-delete-from-join
it’s not very often requested
While using 'Inspector' to introspect MySQL and PostgreSQL databases
dialects, I've noticed some slight variation:
For MySQL, a UNIQUE table constraint also shows up as an Index (with
unique=True).
Ditto PostgreSQL (as of 0.9.8).
Looking at the current HEAD, it appears that PostgreSQL has some
the duplicate_constraints marker in MySQL is “duplicates_index”, in the unique
constraint record. MySQL doesn’t actually have a “unique constraint”
construct, they are all ultimately unique indexes.
On Nov 14, 2014, at 2:35 PM, Jon Nelson jnel...@jamponi.net wrote:
While using
Got it, we can live with `blue sky` =) Thanks, Michael.
—
Dima.
From: Michael Bayer mike...@zzzcomputing.commailto:mike...@zzzcomputing.com
Reply-To: sqlalchemy@googlegroups.commailto:sqlalchemy@googlegroups.com
sqlalchemy@googlegroups.commailto:sqlalchemy@googlegroups.com
Date: Friday,
