On May 18, 1:21 am, Martin Bonner <martinfro...@[EMAIL PROTECTED]
> wrote:
> On May 17, 9:29 am, nickf3 <nic...@[EMAIL PROTECTED]
> wrote:
>
> > On May 16, 8:27 pm, Alan McKenney <alan_mckenn...@[EMAIL PROTECTED]
> wrote:
....
>
> > > A. // only works if host is big-endian
>
> > > unsigned char *buf = ....
> > > unsigned long uint32value;
>
> > > memcpy( &uint32value, buf+ipos, 4 );
>
> > > B. // works regardles of endianness of your host
>
> > > uint32value = ( buf[ipos] << 24 ) |
> > > ( buf[ipos+1] << 16 ) |
> > > ( buf[ipos+2] << 8 ) |
> > > buf[ipos+3] ;
>
> > > I generally use B, to avoid having to worry
> > > about endianness.
>
> > Please don't reinvent the wheel, use standard library:
>
> > #include <arpa/inet.h>
>
> > uint32_t
> > htonl(uint32_t hostlong);
>
> > uint16_t
> > htons(uint16_t hostshort);
>
> > uint32_t
> > ntohl(uint32_t netlong);
>
> > uint16_t
> > ntohs(uint16_t netshort);
>
> > These are noops on big-endian machines, so there's no overhead.
>
> Except that they are /not/ part of the standard library unless you can
> assume Posix. Also they have a horrible interface. The code Alan
> wrote took a buffer of unsigned chars (presumably containing octets
> from the communications protocol) and converted it to a long in a way
> that will work on any compiler that implements the C++ standard (and a
> fair few that don't). ntohl et-al assumes all sorts of things about
> the representations (which may not be true, eg, on a DSP)
>
Let's look at the (open) source, shall we:
/* $OpenBSD: ntohl.c,v 1.6 2005/08/06 20:30:03 espie Exp $ */
/*
* Written by J.T. Conklin <jtc@[EMAIL PROTECTED]
>.
* Public domain.
*/
#include <sys/types.h>
#include <machine/endian.h>
#undef ntohl
u_int32_t
ntohl(u_int32_t x)
{
#if BYTE_ORDER == LITTLE_ENDIAN
u_char *s = (u_char *)&x;
return (u_int32_t)(s[0] << 24 | s[1] << 16 | s[2] << 8 |
s[3]);
#else
return x;
#endif
}
Taken directly from OpenBSD cvs. Looks familiar, doesn't it?
Where's that horror you speak of?
Then, there's arch-specific code for i386:
.... BSD license
/* hostorder = ntohl(netorder) */
ENTRY(ntohl)
movl 4(%esp),%eax
rorw $8,%ax
roll $16,%eax
rorw $8,%ax
ret
Then, sparc:
.... BSD license
/* hostorder = ntohl(netorder) */
ENTRY(ntohl)
retl
nop
And many other archs.
It's perfectly fine (I'd say, required) to _understand_
how to write ****table code, but would you want every
programmer on your team doing this stuff by hand, or
would you want them _reusing_ the existing libs?
Yes, it's not a part of the C++ std lib, but neither is that
socket API you use to get to the "communication channel", nor
whatever API you use to talk to a DSP (which doesn't have
C interface, only C++, right?)
Unless you elaborate on that last statement in
your post, I call it FUD.
--
Nikolai
--
[ See http://www.gotw.ca/resources/clcm.htm
for info about ]
[ comp.lang.c++.moderated. First time posters: Do this! ]
|
