I'll take a stab at it, on first look though it looks like it'll take some serious refactoringf. Wish me luck! :) Doug
On Jan 24, 7:35 pm, Pat Allan <[email protected]> wrote: > Hi Doug > > I don't think anyone's really paid attention to this before... so it's > not a known bug, but you're right, it should be fixed. MySQL will very > likely do the same thing, but everything's abstracted enough that you > won't be restructuring the raw SQL, but re-ordering method calls > instead. If you want to have a go at making a patch, would love to see > it :) > > Cheers > > -- > Pat > > On 25/01/2009, at 3:37 AM, Doug wrote: > > > > > That's a mouthful, so I'll just show an example of what I mean: > > > define_index do > > indexes [:address, amenities.name], :as => :full_text > > end > > > where amenities is some variety of has_many association. In > > postgresql this produces the following sql in the config file: > > > array_to_string(array_accum(COALESCE("properties"."address", '') || ' > > ' || COALESCE("amenities"."name", '')), ' ') > > > Which when run returns a string of the form: "address first_amenity > > address second_amenity address third_amenity" which while function > > ends up making a much larger index than is needed. The correct SQL > > would be: > > > COALESCE("properties"."address", '') || ' ' || array_to_string > > (array_accum(COALESCE("amenities"."name", '')), ' ') > > > which returns: "address first_amenity second_amenity third_amenity", > > what you'd expect. I haven't tried this on Mysql, but it looks like > > it might be doing the same thing. > > > Has anyone else run into this? Is it a known bug? Can someone verify > > that this does the same thing in mysql? I'm happy to write up a > > patch, but I wanted to make sure I was on the right track and no one > > else is working on it right now. > > > Thanks, > > Doug --~--~---------~--~----~------------~-------~--~----~ You received this message because you are subscribed to the Google Groups "Thinking Sphinx" group. To post to this group, send email to [email protected] To unsubscribe from this group, send email to [email protected] For more options, visit this group at http://groups.google.com/group/thinking-sphinx?hl=en -~----------~----~----~----~------~----~------~--~---
