From: Bill Cunningham on

"Bill Cunningham" <nospam(a)nspam.invalid> wrote in message
news:4bb2346b$0$12430
> Now would I also need that header called inet.h or something like that
> for IP information? Posix sure seems easier to me than regular old
> standard C. Not quiute sure why though.

oops. in.h I think I meant.

Bill


From: Bill Cunningham on

"Rick Jones" <rick.jones2(a)hp.com> wrote in message
news:hor5nk$hr1$2(a)usenet01.boi.hp.com...

>> There are the INET and Unix address families (AF_INET, AF_UNIX) and
>> then there are the STREAM and DGRAM style sockets (SOCK_STREAM,
>> SOCK_DGRAM). The former determine the type of addressing you use, and
>> so what you can reach from where. The latter determine the
>> characteristics of the data you send/receive over the socket.
>
> That would be better put as "The latter determine the send/receive
> semantics of the socket. For example a SOCK_STREAM socket will
> present 'byte stream' sematics where message boundaries are up to the
> application. SOCK_DGRAM will present 'message' semantics where
> message boundaries are preserved by the transport."

Would I need all the headers listed under
http://www.opengroup.org/onlinepubs/000095399/idx/networking.html

networking on this page. I'm going to create as simple an ncurses window as
I can but I need to find out about where having a server assign my client a
IP address is going to come from would my system need to broadcast to my
host network's ARP or similar? I shouldn't need bind() on the client side
should I?

Bill


From: Ian Collins on
On 03/31/10 07:27 AM, Bill Cunningham wrote:
> "Rick Jones"<rick.jones2(a)hp.com> wrote in message
> news:hor5nk$hr1$2(a)usenet01.boi.hp.com...
>
>> That would be better put as "The latter determine the send/receive
>> semantics of the socket. For example a SOCK_STREAM socket will
>> present 'byte stream' sematics where message boundaries are up to the
>> application. SOCK_DGRAM will present 'message' semantics where
>> message boundaries are preserved by the transport."
>
> Now would I also need that header called inet.h or something like that
> for IP information? Posix sure seems easier to me than regular old standard
> C. Not quiute sure why though.

Come on Bill, don't run until you can walk.

If you don't have a copy of Stevens' UNIX Network Programming, get one
and read it. Twice.

--
Ian Collins
From: Rick Jones on
Ian Collins <ian-news(a)hotmail.com> wrote:
> If you don't have a copy of Stevens' UNIX Network Programming, get one
> and read it. Twice.

A book on the topic would indeed be goodness. Stevens, Fenner and
Rudoff's UNP is one, I believe there is one from Stallings as well.
Probably others.

rick jones
--
portable adj, code that compiles under more than one compiler
these opinions are mine, all mine; HP might not want them anyway... :)
feel free to post, OR email to rick.jones2 in hp.com but NOT BOTH...
From: Rick Jones on
Bill Cunningham <nospam(a)nspam.invalid> wrote:
> I can but I need to find out about where having a server assign my
> client a IP address is going to come from would my system need to
> broadcast to my host network's ARP or similar? I shouldn't need
> bind() on the client side should I?

As far as writing a sockets program goes, you should be ignoring
question of ARP or how a system is assigned an IP address.

When it comes to the simplest client code:

"You have the right to not code. Anything you code can and will be
used against you by the compiler. You have the right to call bind().
If you cannot afford a bind() call, one will be made for you by
connect()." :)

rick jones
--
I don't interest myself in "why". I think more often in terms of
"when", sometimes "where"; always "how much." - Joubert
these opinions are mine, all mine; HP might not want them anyway... :)
feel free to post, OR email to rick.jones2 in hp.com but NOT BOTH...
First  |  Prev  |  Next  |  Last
Pages: 1 2 3 4
Prev: Not able to change baud rate
Next: Basic Shell concept