Horton

Re: B-tree parent pointer and checkpoints

Greg Stark <gsstark@mit.edu>

View thread
From: Greg Stark <gsstark@mit.edu>
To: Heikki Linnakangas <heikki.linnakangas@enterprisedb.com>
Cc: Tom Lane <tgl@sss.pgh.pa.us>, Teodor Sigaev <teodor@sigaev.ru>, PostgreSQL-development <pgsql-hackers@postgresql.org>, Oleg Bartunov <oleg@sai.msu.su>
Date: 2010-11-12T22:34:51Z
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. Rewrite the GiST insertion logic so that we don't need the post-recovery

On Fri, Nov 12, 2010 at 7:20 PM, Heikki Linnakangas
<heikki.linnakangas@enterprisedb.com> wrote:
> I think we can work around that with a small modification to the page split
> algorithm. In a nutshell, when the child page is split, put a flag on the
> left half indicating that the rightlink must always be followed, regardless
> of the NSN. When the downlink is inserted to the parent, clear the flag.
> Setting and clearing of these flags need to be performed during WAL replay
> as well.
>

Does this not cause duplicate results? Or does GIST already have to be
prepared to deal with duplicate results?


-- 
greg