aliased tables and _deep_deannotate (again)

[Christoph Rauch]

Hello Michael, hello List,

thanks for your previous patch. Most queries of that type do work now.

I am sorry but I have to inform you that I stumbled upon another query which used these selects in a nested fashion, that triggered the bug. :-}

cheers,

Christoph

[Christoph Rauch]

Hello List,

I want to add that for this there exists a workaround for this bug, namely factoring out the select() and directly putting it into the other column_properties. This works.

Christoph

[Michael Bayer]

OK so from the ORM perspective, you're building column_property objects composed from others right ?

--
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/-/0yb7gY4d_BAJ.
To post to this group, send email to sqlal...@googlegroups.com.
To unsubscribe from this group, send email to sqlalchemy+...@googlegroups.com.
For more options, visit this group at http://groups.google.com/group/sqlalchemy?hl=en.
<deep_deannotate_and_aliased_tables.py>

[Michael Bayer]

anyway, this is fixed again by applying my original patch that I thought I wasn't going to need, so thanks for the very nice testing !

[Christoph Rauch]

Hello Michael,

2012/4/3 Michael Bayer <mik...@zzzcomputing.com>

anyway, this is fixed again by applying my original patch that I thought I wasn't going to need, so thanks for the very nice testing !

Ah, nice! I'll apply that one then. Thank *you* for all the awesome work! :)

 

--
Christoph

[Christoph Rauch]

Hello Michael,

2012/4/3 Michael Bayer <mik...@zzzcomputing.com>

OK so from the ORM perspective, you're building column_property objects composed from others right ?

Correct. If I think about it, its totally awesome that this works at all. :)

--
Christoph