"Ross J. Reedstrom" <reedstrm@rice.edu> writes:
> So perhaps it was more of the "This code is less ready than I thought
> it was, but now that I've spent the time understanding it and the
> problem, the shortest way out is forward".
Yeah, exactly. By the time I really understood how incomplete the
collation patch was, I'd done most of the work to fix it; and insisting
on backing it out of 9.1 didn't seem productive (even assuming that I
could have won that argument, which was by no means a given).
I'm still fairly troubled by the potential overhead in the form of extra
lookups during parse time, but have not had the time to try to measure
that. Too bad we haven't got a performance-test farm.
regards, tom lane