Michael Glaesemann <grzm@myrealbox.com> writes:
> I'm interested in hearing what others have to say, especially along the
> lines of implementation. I haven't seen much discussion of distributed
> keys on the lists (other than Josh Berkus), so perhaps there isn't much
> interest. Would there be any foreseeable opposition if I (and/or
> others) worked on this?
This seems to have considerable overlap with the problem of indexing
inheritance hierarchies (so that constraints on tables with children
would work as people expect). It may be that it's sufficient to solve
it for inheritance cases, and not try to support the generic case of
constraints applied across arbitrary sets of tables. ISTM that the
latter could introduce a bunch of extra definitional and practical
issues beyond what you'd have to solve to do the former.
You're discussing it on the wrong list though --- try -hackers.
regards, tom lane