Horton

Re: Alpha4 release blockers (was Re: wrapping up this CommitFest)

Tom Lane <tgl@sss.pgh.pa.us>

View thread
From: Tom Lane <tgl@sss.pgh.pa.us>
To: Peter Eisentraut <peter_e@gmx.net>
Cc: pgsql-hackers@postgresql.org, Andres Freund <andres@anarazel.de>, Rémi Zara <remi_zara@mac.com>, Stefan Huehner <stefan@huehner.org>
Date: 2011-03-06T16:54:48Z
Lists: pgsql-hackers

Commits

Same data as JSON: GET /api/v1/messages/:b64id/commits the thread's linked commits as JSON, with link sources. API reference →

  1. Further refine patch for commenting operator implementation functions.

  2. Fix citext's upgrade-from-unpackaged script to set its collation correctly.

  3. Mark operator implementation functions as such in their comments.

  4. Add KNNGIST support to contrib/btree_gist.

  5. Fix plpython breakage detected on certain Fedora machines on buildfarm.

Andres Freund <andres@anarazel.de> writes:
> Ah. Finally after trying to stare down the code for some more time the issue 
> is pretty simple.

> -		fmgr_info_collation(irel->rd_index->indcollation.values[attnum-1],
> +		fmgr_info_collation(irel->rd_indcollation[attnum-1],

Good catch ... but while I was poking around to make sure that there
were no other similar errors, I started to get pretty desperately
unhappy with the state of this code altogether.  What exactly is
indcollation supposed to mean?  Does it have any meaning for unordered
indexes?  If it means something about the index's sort order, why is it
that it's not consulted while building the pathkeys that represent the
sort order?  (It isn't.)

			regards, tom lane