bin/176713: [patch] nc(1) closes network socket too soon

[Ronald F. Guilmette]

In message <CAGwOe2Z+cS5zRx4NomiuXb5f04tp5WmKUXw67CQToEzWiHwuwQ at mail.gmail.com>
=?ISO-8859-1?Q?Fernando_Apestegu=EDa?= <fernando.apesteguia at gmail.com> wrote:

>Yes, that is what I tested. Behavior before (truncated output) and after
>(correct output) applying the patch.

OK, good.  Thanks.

>If this is going to be the final version of the patch, i.e. if it is going
>to include the -q flag, then the patch needs to be extended to reflect that
>in nc(1) man page.

I am in complete agreement that _if_ a new option is implemented within nc
then it really must also be properly documented in the associated man page.

As I have expressed however, it is my hope that whoever decides these
things will decide to simply fix the bug in nc and to _not_ even bother
to introduce an option which might help to preserve ``backward compatability''
with the current (broken) behavior of nc.

I simply do not believe that the current (arguably "broken") behavior of nc
is of any particular value to anybody.  The only reason I proposed a patch
that included an option to elicit the (non-broken) behavior was because I
have the humility to admit that I am not actually omniscient with respect
to other people's needs.


Regards,
rfg