Horton

Re: recovery modules

Michael Paquier <michael@paquier.xyz>

View thread View on postgr.es
From: Michael Paquier <michael@paquier.xyz>
To: Nathan Bossart <nathandbossart@gmail.com>
Cc: Andres Freund <andres@anarazel.de>, pgsql-hackers@postgresql.org
Date: 2023-01-17T05:32:03Z
Lists: pgsql-hackers 
On Mon, Jan 16, 2023 at 02:40:40PM -0800, Nathan Bossart wrote:
> On Mon, Jan 16, 2023 at 04:36:01PM +0900, Michael Paquier wrote:
> > Once this issue was fixed, nothing else stood out, so applied this
> > part.
> 
> Thanks!  I've attached a rebased version of the rest of the patch set.

When it comes to 0002, the only difference between the three code
paths of shell_recovery_end(), shell_archive_cleanup() and
shell_restore() is the presence of BuildRestoreCommand().  However
this is now just a thin wrapper of replace_percent_placeholders() that
does just one extra make_native_path() for the xlogpath.

Could it be cleaner in the long term to remove entirely
BuildRestoreCommand() and move the conversion of the xlogpath with
make_native_path() one level higher in the stack?
--
Michael

Commits

  1. Redesign archive modules

  2. Remove uses of AssertVariableIsOfType() obsoleted by f2b73c8

  3. Rename contrib module basic_archive to basic_wal_module

  4. Refactor code for restoring files via shell commands

  5. Refactor code in charge of running shell-based recovery commands