Re: Small fix for inv_getsize - Mailing list pgsql-patches

From Tom Lane
Subject Re: Small fix for inv_getsize
Date
Msg-id 25612.973183778@sss.pgh.pa.us
Whole thread Raw
In response to Small fix for inv_getsize  (Denis Perchine <dyp@perchine.com>)
Responses Re: Small fix for inv_getsize  (Denis Perchine <dyp@perchine.com>)
List pgsql-patches
Denis Perchine <dyp@perchine.com> writes:
> Just realized that inv_getsize is a little bit wrong :-)). It just get the
> first page, not last
> Here is the patch which will fix the behavior.

No it doesn't; it's a loop, and your patch will change nothing.  Your
original version tried to stop after fetching one tuple, which was
wrong because of visibility considerations.

Now that I think about it, this code could do a two-key scan backwards
and stop after finding the first (last) valid tuple, but that's more
than a one-line change.

            regards, tom lane

pgsql-patches by date:

Previous
From: Tom Lane
Date:
Subject: Re: Patches for REL7_0_PATCHES
Next
From: Denis Perchine
Date:
Subject: Re: Small fix for inv_getsize