Re: [bug-inetutils] ping, --count=NUMBER

[Alfred M. Szmidt]

   Alfred M. Szmidt <address@hidden> ha escrit:

   > It would work the same way as POSIXLY_CORRECLY.

   OK, then, to make it clear:

   1) BSD_ME_HARDER=0

     Ping default number of times and exit.

   2) BSD_ME_HARDER=1

     Ping until user breaks by C-c.

   3) BSD_ME_HARDER is undefined

     Same as 2)

   If that's what you mean, then it's OK with me.  The (3) is
   important to not break the default behaviour.

Yes, that is what I mean.

  However the variable name seems strange to me.

It is just a pun on POSIX_ME_HARDER, I'd prefer something more sane.
Maybe BSDLY_CORRECTLY..

   Speaking about environment variables, I'd rather propose, instead
   of BSD_ME_HARDER, to introduce the PING_COUNT variable with the
   following meaning:

   1) PING_COUNT=0 or undefined
     Ping until user breaks by C-c.

   2) PING_COUNT=N, where N is an integer number
     Ping that number of times and exit (unless overridden by -c).

   How about that?  

--count=0 defaults to infinity, so I find the above confusing.  It
would also mean that we would have to introduce a FOO_BAR variable for
each FOO program, and BAR feature.