unofficial mirror of guile-devel@gnu.org 
 help / color / mirror / Atom feed
From: Christopher Baines <mail@cbaines.net>
To: Maxime Devos <maximedevos@telenet.be>
Cc: guile-devel@gnu.org
Subject: Re: [PATCH] Define SO_RCVTIMEO and SO_SNDTIMEO.
Date: Sun, 18 Sep 2022 16:11:23 +0200	[thread overview]
Message-ID: <87leqgrcsc.fsf@cbaines.net> (raw)
In-Reply-To: <a0571903-1089-e99c-bcc5-fcf002a925d7@telenet.be>

[-- Attachment #1: Type: text/plain, Size: 1825 bytes --]


Thanks for the feedback Maxime, I obviously lost track of some things
after letting this patch languish for a while.

Maxime Devos <maximedevos@telenet.be> writes:

> On 17-09-2022 10:05, Christopher Baines wrote:
>> +  if (ioptname == SO_RCVTIMEO || ioptname == SO_SNDTIMEO)
>> +    {
>> +      SCM_ASSERT (scm_is_pair (value), value, SCM_ARG4, FUNC_NAME);
>
> How about using SCM_ASSERT_TYPE instead to improve the error message a
> little?

I've switched to using SCM_ASSERT_TYPE now.

>> +      opt_time.tv_sec = scm_to_ulong (SCM_CAR (value));
>> +      opt_time.tv_usec = scm_to_ulong (SCM_CDR (value));
> I think it needs to be documented how to actually use SO_RCVTIMEO /
> SO_SNDTIMEO (more specifically, the types).  For example, the user
> might expect to enter flonums 1.567 as time durations (in seconds)
> instead, or a duration from SRFI 19.  How is the user supposed to know
> the appropriate type, aside from experimenting and reading the source
> code?
>
> Also, it is not documented there is a maximum on the number of
> seconds, I recommend to document that there is some limit, or
> alternatively clip numbers to the maximum.

I've added some documentation now, I'm not sure the interface is
perfect, but it should work.

>> +      optlen = sizeof (opt_time);
>> +      optval = &opt_time;
>> +    }
>> +
>
> I think this needs a #if defined(SO_RCVTIMEO) && defined(SO_SNDTIMEO),
> for systems that don't have those.  Alternatively, if all systems have
> those, it seems to me that the #ifdef in

I guess it should be conditional, so I've added the if in to getsockopt
and setsockopt.

>> +#ifdef SO_RCVTIMEO
>> +  scm_c_define ("SO_RCVTIMEO", scm_from_int (SO_RCVTIMEO));
>> +#endif
>> +#ifdef SO_SNDTIMEO
>> +  scm_c_define ("SO_SNDTIMEO", scm_from_int (SO_SNDTIMEO));
>> +#endif
>
> can be removed.


[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 987 bytes --]

  reply	other threads:[~2022-09-18 14:11 UTC|newest]

Thread overview: 7+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2022-09-17  8:05 [PATCH] Define SO_RCVTIMEO and SO_SNDTIMEO Christopher Baines
2022-09-17  9:40 ` Maxime Devos
2022-09-18 14:11   ` Christopher Baines [this message]
2022-09-18 13:56 ` [PATCH v2] " Christopher Baines
2022-09-18 14:17 ` [PATCH v3] " Christopher Baines
2022-10-12 20:28   ` Ludovic Courtès
2022-10-13 10:28     ` Christopher Baines

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

  List information: https://www.gnu.org/software/guile/

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=87leqgrcsc.fsf@cbaines.net \
    --to=mail@cbaines.net \
    --cc=guile-devel@gnu.org \
    --cc=maximedevos@telenet.be \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).