[sqlalchemy] Multiple JOINs of superclass table and potential discrepancy in chained vs on-clause

2015-08-03 Thread Douglas Russell
Hi again, Full code: https://gist.github.com/dpwrussell/8ecca88f642cca003999 I have an structure linked together like so. A-B is a Many-To-Many and uses an association table. A and B are both subclasses of common base Object. A └── B I also have an object X that can be linked to any type of ob

Re: [sqlalchemy] Multiple JOINs of superclass table and potential discrepancy in chained vs on-clause

2015-08-03 Thread Mike Bayer
all of those syntaxes are supposed to be exactly equivalent so this is a major issue: https://bitbucket.org/zzzeek/sqlalchemy/issues/3505/join-targeting-broken-for-joined-inh On 8/3/15 11:13 AM, Douglas Russell wrote: Hi again, Full code: https://gist.github.com/dpwrussell/8ecca88f642cca003

Re: [sqlalchemy] Multiple JOINs of superclass table and potential discrepancy in chained vs on-clause

2015-08-03 Thread Mike Bayer
On 8/3/15 1:01 PM, Mike Bayer wrote: all of those syntaxes are supposed to be exactly equivalent so this is a major issue: https://bitbucket.org/zzzeek/sqlalchemy/issues/3505/join-targeting-broken-for-joined-inh OK, this is not as much of a bug as I thought, though I might be able to do som

Re: [sqlalchemy] Multiple JOINs of superclass table and potential discrepancy in chained vs on-clause

2015-08-10 Thread Douglas Russell
Thanks. It makes sense to me that you would alias in the JOIN as well as in the WHERE as you say in the issue. I'm tracking that and making use of the from_joinpoint for now. On Monday, 3 August 2015 13:12:16 UTC-4, Michael Bayer wrote: > > > > On 8/3/15 1:01 PM, Mike Bayer wrote: > > all of