Re: libpq host/hostaddr/conninfo inconsistencies - Mailing list pgsql-hackers

From Tom Lane
Subject Re: libpq host/hostaddr/conninfo inconsistencies
Date
Msg-id 607.1550711575@sss.pgh.pa.us
Whole thread Raw
In response to Re: libpq host/hostaddr/conninfo inconsistencies  (Robert Haas <robertmhaas@gmail.com>)
Responses Re: libpq host/hostaddr/conninfo inconsistencies  (Michael Paquier <michael@paquier.xyz>)
Re: libpq host/hostaddr/conninfo inconsistencies  (Fabien COELHO <coelho@cri.ensmp.fr>)
List pgsql-hackers
Robert Haas <robertmhaas@gmail.com> writes:
>> Robert, any chance you could opine on the doc patch, given that's your
>> suggested direction?

> I find it to be a more change than we really need, and I'm not sure
> how much it helps to clarify the issue at hand. Here is a simpler
> change which seems like it might do the trick (or maybe not, let's see
> what others think).

My only complaint about this is that it makes it sound like you *must*
provide "host", even when giving "hostaddr":

-        Numeric IP address of host to connect to.  This should be in the
+        Numeric IP address that should be used for the server specified
+        by <literal>host</literal>.  This should be in the

The second para explains the cases in which you actually do need to
provide "host", but I'm afraid that the first sentence will have
misled people enough that they won't get the point.

I don't think there's anything very wrong with the existing wording
of this sentence.

Robert's second and third changes seem fine, though.

            regards, tom lane


pgsql-hackers by date:

Previous
From: Peter Geoghegan
Date:
Subject: Re: amcheck verification for GiST
Next
From: Tom Lane
Date:
Subject: Re: Delay locking partitions during INSERT and UPDATE