No, we definitely need to support all the valid qualifiers in EOF.
I'm not sure if I can fix this WITHOUT rewriting qualifier parsing.
ms
On Jun 12, 2007, at 7:40 PM, Stephen Edwards wrote:
> Mike Schrag wrote:
>> Does this selector actually work properly anywhere but in-memory?
>
> Good question--I'm using it on in-memory filtering in
> this situation, where it does work correctly (and isEqualTo:
> doesn't). My memory (which isn't trustworthy) is that
> doesContain: works when used alone, but when multiple clauses
> are or'ed--which is often the case for me--it fails. The
> EOF-generated SQL for database fetches produces incorrect results.
> The generated SQL produces a simultaneous join on all the
> required entities (good for and, evil for or). Of course,
> I haven't dug around in that stuff for a long time, so I
> could be mistaken ...
>
>> This is one of the api mismatches between cayenne and eof, actually.
>
> Is there something else I should use instead? IIRC, doesContain:
> wasn't actually supported by the query builder in EOModeler (in WO)
> either. Since I much prefer Entity Modeler, I thought I would
> ask. On the one hand, I don't know how to write these fetch
> specs without doesContain:, and I don't want to hard-code them
> in Java either. On the other hand, I don't want to go without
> such an incredibly useful tool as Entity Modeler! It (like everything
> about WOLips) is a nice piece of work.
>
> -- Steve
>
> ----------------------------------------------------------------------
> -----
> Virginia Tech, CS Dept. Web-CAT: Web-based Center for Software
> Testing
> 660 McBryde Hall (0106) Automatic grading using student-written
> tests
> Blacksburg, VA 24061 USA http://web-cat.sourceforge.net/
> (540)-231-5723 http://people.cs.vt.edu/~edwards/
This archive was generated by hypermail 2.0.0 : Wed Jun 13 2007 - 12:43:04 EDT