Re: Progress on fast path sorting, btree index creation time - Mailing list pgsql-hackers

From Peter Geoghegan
Subject Re: Progress on fast path sorting, btree index creation time
Date
Msg-id CAEYLb_UCYjNL3kPG+2p3JHA86rwf=vaZOBz6F+eBmReWSHBS0A@mail.gmail.com
Whole thread Raw
In response to Re: Progress on fast path sorting, btree index creation time  (Bruce Momjian <bruce@momjian.us>)
List pgsql-hackers
On 9 February 2012 17:16, Bruce Momjian <bruce@momjian.us> wrote:
>> Yes, I am. The main reason that the loops exist in those functions
>> (which is the only way that they substantially differ) is because they
>> each have to get the other keys through various ways that characterise
>> the tuple class that they encapsulate (index_getattr(),
>> heap_getattr(), etc).
>
> Does this help all types for sorting, including strings?

Yes, it does, though of course that's not expected to make too much
difference with text when the C locale isn't used, because the
comparisons are inherently much more expensive, and there isn't a
whole lot we can do about that, at least here.

--
Peter Geoghegan       http://www.2ndQuadrant.com/
PostgreSQL Development, 24x7 Support, Training and Services


pgsql-hackers by date:

Previous
From: Robert Haas
Date:
Subject: Re: Add protransform for numeric, varbit, and temporal types
Next
From: Tom Lane
Date:
Subject: Re: Notify system doesn't recover from "No space" error