pgsql: Use access() to check file existence in GetNewRelFileNode()

Started by Michael Paquieralmost 8 years ago4 messagescomitters
Jump to latest
#1Michael Paquier
michael@paquier.xyz

Use access() to check file existence in GetNewRelFileNode()

Previous code used BasicOpenFile() and close() just to check for a file
collision, while there is no need to hold open a file descriptor but
that's an overkill here.

Author: Paul Guo
Reviewed-by: Peter Eisentraut, Michael Paquier
Discussion: /messages/by-id/CABQrizcUtiHaquxK=d4etBX8GF9kbZB50Nt1gO9_aN-e9SptyQ@mail.gmail.com

Branch
------
master

Details
-------
https://git.postgresql.org/pg/commitdiff/677da8c15d19c11465d78f18bfd5ceb5d6fc3af1

Modified Files
--------------
src/backend/catalog/catalog.c | 15 ++++-----------
1 file changed, 4 insertions(+), 11 deletions(-)

#2Andrew Dunstan
andrew@dunslane.net
In reply to: Michael Paquier (#1)
Re: pgsql: Use access() to check file existence in GetNewRelFileNode()

On 07/08/2018 06:00 AM, Michael Paquier wrote:

Reviewed-by: Peter Eisentraut, Michael Paquier

I think if you're the committer you're assumed to have reviewed the
patch :-)

cheers

andrew

--
Andrew Dunstan https://www.2ndQuadrant.com
PostgreSQL Development, 24x7 Support, Remote DBA, Training & Services

#3Andres Freund
andres@anarazel.de
In reply to: Andrew Dunstan (#2)
Re: pgsql: Use access() to check file existence in GetNewRelFileNode()

Hi,

On 2018-07-08 10:05:53 -0400, Andrew Dunstan wrote:

On 07/08/2018 06:00 AM, Michael Paquier wrote:

Reviewed-by: Peter Eisentraut, Michael Paquier

I think if you're the committer you're assumed to have reviewed the patch
:-)

I personally think it makes sense to put oneself there. At least I order
the Reviewed-By etc lists roughly by the amount of work done. Sometimes
you come across a patch where others have done most of the review work,
so it's sensible to put yourself last. In other cases you might have
spent months of effort - so first.

Greetings,

Andres Freund

#4Michael Paquier
michael@paquier.xyz
In reply to: Andres Freund (#3)
Re: pgsql: Use access() to check file existence in GetNewRelFileNode()

On Sun, Jul 08, 2018 at 10:27:33AM -0700, Andres Freund wrote:

On 2018-07-08 10:05:53 -0400, Andrew Dunstan wrote:

I think if you're the committer you're assumed to have reviewed the patch
:-)

I personally think it makes sense to put oneself there. At least I order
the Reviewed-By etc lists roughly by the amount of work done. Sometimes
you come across a patch where others have done most of the review work,
so it's sensible to put yourself last. In other cases you might have
spent months of effort - so first.

Yeah, that's my line of thoughts as well. So it is not strange to me
either to put oneself in any lists if adapted.
--
Michael