Re: Errands around AllocateDir()

From: Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>
To: Michael Paquier <michael(dot)paquier(at)gmail(dot)com>
Cc: PostgreSQL mailing lists <pgsql-hackers(at)postgresql(dot)org>
Subject: Re: Errands around AllocateDir()
Date: 2017-12-04 17:05:26
Message-ID: 18531.1512407126@sss.pgh.pa.us
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-hackers

Michael Paquier <michael(dot)paquier(at)gmail(dot)com> writes:
> I had a close look at all the callers of AllocateDir() and noticed a
> couple of unwelcome things (Tom noticed some of those in the thread
> mentioned above, I found others):

The only part of this that seems likely to be controversial is the
decision to get rid of special-case error messages, eg replace
"could not open tablespace directory \"%s\": %m"
with the more generic
"could not open directory \"%s\": %m"

As I said in the previous thread, I don't see anything much wrong
with that; but if anyone doesn't like it, speak now.

Also, I'm inclined to back-patch the exporting of ReadDirExtended,
so that we can rely on it being available if we use it in any
back-patched fixes. I wouldn't back-patch anything else here that's
not a clear bug fix, but we may need public ReadDirExtended anyway
just to do that much (I didn't really check that yet).

regards, tom lane

In response to

Responses

Browse pgsql-hackers by date

  From Date Subject
Next Message Konstantin Knizhnik 2017-12-04 17:18:42 Re: [HACKERS] Secondary index access optimizations
Previous Message Marco Pfatschbacher 2017-12-04 16:55:25 Re: [HACKERS] PATCH: Keep one postmaster monitoring pipe per process