Re: TODO item: teach pg_dump about sparsely-stored large objects - Mailing list pgsql-hackers

From Greg Stark
Subject Re: TODO item: teach pg_dump about sparsely-stored large objects
Date
Msg-id CAM-w4HOvd-OKJ=hy72YsRV72SusSMX4rYpbAvt+RvfkJDYtQDw@mail.gmail.com
Whole thread Raw
In response to TODO item: teach pg_dump about sparsely-stored large objects  (Tom Lane <tgl@sss.pgh.pa.us>)
List pgsql-hackers
On Tue, Oct 9, 2012 at 3:40 AM, Tom Lane <tgl@sss.pgh.pa.us> wrote:
> Admittedly, this is no different than what happens when you try to back
> up a sparsely-stored Unix file, at least with simpler backup tools.
> But it seems to me we should try a bit harder.

Fwiw both GNU tar and GNU cp support creating sparse files. They do it
by just detecting blocks of NULs and skipping over them. pg_restore
could do that today without any API changes.

That said, an API to save pg_dump the time and space of reading the
fake zeros out of the database dosen't sonud like a bad thing.

-- 
greg



pgsql-hackers by date:

Previous
From: Greg Stark
Date:
Subject: Re: Add FET to Default and Europe.txt
Next
From: Sébastien Lardière
Date:
Subject: Re: Truncate if exists