Re: psql not responding to SIGINT upon db reconnection - Mailing list pgsql-hackers

From Robert Haas
Subject Re: psql not responding to SIGINT upon db reconnection
Date
Msg-id CA+Tgmoa1Uaqnd_OnV=KrQ5qf3=wYv8JM=3SC1e_=0xo0o9KEjg@mail.gmail.com
Whole thread Raw
In response to Re: psql not responding to SIGINT upon db reconnection  ("Tristan Partin" <tristan@neon.tech>)
Responses Re: psql not responding to SIGINT upon db reconnection
List pgsql-hackers
On Wed, Jan 31, 2024 at 1:07 PM Tristan Partin <tristan@neon.tech> wrote:
> I was looking for documentation of PQsocket(), but didn't find any
> standalone (unless I completely missed it). So I just copied how
> PQsocket() is documented in PQconnectPoll(). I am happy to document it
> separately if you think it would be useful.

As Jelte said back at the end of January, I think you just completely
missed it. The relevant part of libpq.sgml starts like this:

    <varlistentry id="libpq-PQsocket">
     <term><function>PQsocket</function><indexterm><primary>PQsocket</primary></indexterm></term>

As far as I know, we document all of the exported libpq functions in
that SGML file.

I think there's no real reason why we couldn't get at least 0001 and
maybe also 0002 into this release, but only if you move quickly on
this. Feature freeze is approaching rapidly.

Modulo the documentation changes, I think 0001 is pretty much ready to
go. 0002 needs comments. I'm also not so sure about the name
process_connection_state_machine(); it seems a little verbose. How
about something like wait_until_connected()? And maybe put it below
do_connect() instead of above.

--
Robert Haas
EDB: http://www.enterprisedb.com



pgsql-hackers by date:

Previous
From: Nathan Bossart
Date:
Subject: Re: Slow GRANT ROLE on PostgreSQL 16 with thousands of ROLEs
Next
From: Aleksander Alekseev
Date:
Subject: Re: MIN/MAX functions for a record