Re: pg_ls_waldir() & pg_ls_logdir()

From: Robert Haas <robertmhaas(at)gmail(dot)com>
To: Dave Page <dpage(at)pgadmin(dot)org>
Cc: PostgreSQL Hackers <pgsql-hackers(at)postgresql(dot)org>
Subject: Re: pg_ls_waldir() & pg_ls_logdir()
Date: 2017-03-15 17:27:30
Message-ID: CA+Tgmoa_PLTthr1GkLsso0Xnb_KasyAkBbf6A_1zCFsdbcCo2g@mail.gmail.com
Views: Raw Message | Whole Thread | Download mbox
Thread:
Lists: pgsql-hackers

On Mon, Feb 20, 2017 at 6:21 AM, Dave Page <dpage(at)pgadmin(dot)org> wrote:
> Patch includes the code and doc updates.

Review:

+ strftime(mtime, 25, "%Y-%m-%d %H:%M:%S %Z",
localtime(&(attrib.st_ctime)));
+ const int n = snprintf(NULL, 0, "%lld", attrib.st_size);
+ char size[n+1];
+ snprintf(size, n+1, "%lld", attrib.st_size);

We don't allow variable declarations in mid-block. You've been
programming in C++ for too long!

The documentation should be updated to say that access to
pg_ls_logdir() and pg_ls_waldir() can be granted via the permissions
system (see the paragraph above the table you updated).

The four existing functions in the documentation table each have a
corresponding paragraph below the table, but the two new functions you
added don't.

+1 for the concept.

--
Robert Haas
EnterpriseDB: http://www.enterprisedb.com
The Enterprise PostgreSQL Company

In response to

Responses

Browse pgsql-hackers by date

  From Date Subject
Next Message Peter Eisentraut 2017-03-15 17:29:56 Re: logical replication access control patches
Previous Message Peter Eisentraut 2017-03-15 17:26:35 Re: Need a builtin way to run all tests faster manner