Horton

Re: range_agg with multirange inputs

Greg Stark <stark@mit.edu>

View thread View on postgr.es
From: Greg Stark <stark@mit.edu>
To: Chapman Flack <chap@anastigmatix.net>
Cc: Paul Jungwirth <pj@illuminatedcomputing.com>, PostgreSQL Hackers <pgsql-hackers@lists.postgresql.org>
Date: 2022-03-28T20:17:34Z
Lists: pgsql-hackers 
Fwiw the cfbot is failing due to a duplicate OID. Traditionally we
didn't treat duplicate OIDs as reason to reject a patch because
they're inevitable as other patches get committed and the committer
can just renumber them.

I think the cfbot kind of changes this calculus since it's a pain lose
the visibility into whether the rest of the tests are passing that the
cfbot normally gives us.

If it's not to much of a hassle could you renumber resubmit the patch
with an updated OID?

[10:54:57.606] su postgres -c "make -s -j${BUILD_JOBS} world-bin"
[10:54:57.927] Duplicate OIDs detected:
[10:54:57.927] 8000

Commits

  1. Add range_agg with multirange inputs

  2. Change some internal error messages to elogs

  3. Additional tests for range_intersect_agg(anymultirange)

  4. doc: Document range_intersect_agg(anymultirange)