Re: Improve the granularity of PQsocketPoll's timeout parameter?

Robert Haas <robertmhaas@gmail.com>

From: Robert Haas <robertmhaas@gmail.com>
To: Tom Lane <tgl@sss.pgh.pa.us>
Cc: pgsql-hackers@lists.postgresql.org, Tristan Partin <tristan@partin.io>, Dominique Devienne <ddevienne@gmail.com>
Date: 2024-06-12T18:45:35Z
Lists: pgsql-hackers
On Wed, Jun 12, 2024 at 2:25 PM Tom Lane <tgl@sss.pgh.pa.us> wrote:
> Robert Haas <robertmhaas@gmail.com> writes:
> > On Wed, Jun 12, 2024 at 1:53 PM Tom Lane <tgl@sss.pgh.pa.us> wrote:
> >> * I decided to invent a typedef
> >> typedef pg_int64 PGusec_time_t;
>
> > I don't think it's a bad idea to have a typedef, but that particular
> > one is pretty unreadable. Mmm, let's separate some things with
> > underscores and others by a change in the capitalization conventIon!
>
> "PG" as a prefix for typedefs in libpq-fe.h is a pretty ancient
> precedent.  I'm not wedded to any of the rest of it --- do you
> have a better idea?

Hmm, well, one thing I notice is that most of the other typedefs in
src/interfaces/libpq seem to do PGWordsLikeThis or PGwordsLikeThis
rather than PGwords_like_this. There are a few that randomly do
pg_words_like_this, too. But I know of no specific precedent for how a
microsecond type should be named.

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



Commits

  1. Improve the granularity of PQsocketPoll's timeout parameter.