Re: Unsplitting btree index leaf pages

Martijn van Oosterhout wrote:

> The downsides are probably that it's a pain to make it work
> concurrently and requires writing each index page at least twice. But
> it's a thought...

We already do something similar for page deletions.  Empty pages are not
deleted right away, but they are marked with BTP_DEAD, and then deleted
on a subsequent vacuum.  Or something like that, I don't remember the
exact details.

-- 
Alvaro Herrera                                http://www.CommandPrompt.com/
The PostgreSQL Company - Command Prompt, Inc.