Hi,
On Fri, Sep 02, 2022 at 10:37:56AM +0200, Hiltjo Posthuma wrote:
> On Thu, Sep 01, 2022 at 01:56:04PM +0000, Tom Schwindl wrote:
> > On Tue Aug 30, 2022 at 10:43 AM CEST, Petr Vaněk wrote:
> > > ---
> > > ii.1 | 4 ++--
> > > 1 file changed, 2 insertions(+), 2 deletions(-)
> > >
> > > diff --git a/ii.1 b/ii.1
> > > index 8e06af7..db05ab6 100644
> > > --- a/ii.1
> > > +++ b/ii.1
> > > _AT_@ -29,8 +29,8 @@ and ii creates a new channel directory with in and out file.
> > > .IR nickname ]
> > > .RB [ \-f
> > > .IR realname ]
> > > -.RB < \-u
> > > -.IR sockname >
> > > +.RB [ \-u
> > > +.IR sockname ]
> > > .SH OPTIONS
> > > .TP
> > > .BI \-s " servername"
> > > --
> > > 2.35.1
> >
> > While we're at it, I think we should also remove the angle brackets
> > around the `-s' option and sync the names of the arguments with the
> > ones used in the usage() function.
> > Additionally, the `-u' option is missing in the usage message and a lot
> > of unnecessary angle brackets are used, too.
The -u is not missing in usage, see ii output from most recent commit:
% ./ii
usage: ./ii <-s host> [-i <irc dir>] [-p <port>] [-u <sockname>] [-n <nick>] [-k <password>] [-f <fullname>]
^~~~~~~~~~~~~~~
> I agree.
But yeah, I can remove angle brackets. I also see that -k is not
synchronized between man page and usage, the "password" in usage message
is quite misleading.
> I don't fully like the man page, because using either the -s or -u option is
> mandatory.
>
> Maybe something like: <-s server> | <-u socket>.
Well, this is not true, only -s is mandatory. It is always used in irc
protocol in PING and USER messages. However, it is also used for opening
tcp connection if -u is not given.
Therefore I would stick with: <-s host> [-u sockname]
Petr
Received on Fri Sep 02 2022 - 11:19:37 CEST