Casting return value to fit within constraints of error values (portability)

[clayne]

I realize this may work for the majority of platforms. But I am
concerned about the aspects of sign extension and portability.

Is this portable:

{
unsigned long bw;
[...]

if ((signed long)(bw = read(s, buf, sizeof(buf))) == -1) {
error_handling();
}
}

With the goal being able to utilize larger unsigned values but still
being able to check for a return value of "-1" (since this is the
mechanism read() uses) to handle errors?

[clayne]

clayne wrote:

unsigned long bw;
[...]

if ((signed long)(bw = read(s, buf, sizeof(buf))) == -1) {

In retrospect this is probably a pointless "workaround" as read() won't
ever be returning more than signed size_t (ssize_t) in the first place,
so no need to declare 'bw' as unsigned long or size_t, etc.

[clayne]

clayne wrote:

In retrospect this is probably a pointless "workaround" as read() won't
ever be returning more than signed size_t (ssize_t) in the first place,

Interesting though that I already see (Solaris for instance):

ssize_t read(int fildes, void *buf, size_t nbyte);

The read() function attempts to read nbyte bytes from the
file associated with the open file descriptor, fildes, into
the buffer pointed to by buf.

[Mark McIntyre]

On 27 Jan 2006 20:57:31 -0800, in comp.lang.c , "clayne"
<cl****@anodized.com> wrote:

ssize_t read(int fildes, void *buf, size_t nbyte);

you /are/ aware that read() is offtopic here? Its a posix function,
probably best discussed in a unix group.
Mark McIntyre
--
"Debugging is twice as hard as writing the code in the first place.
Therefore, if you write the code as cleverly as possible, you are,
by definition, not smart enough to debug it."
--Brian Kernighan

----== Posted via Newsfeeds.Com - Unlimited-Unrestricted-Secure Usenet News==----
http://www.newsfeeds.com The #1 Newsgroup Service in the World! 120,000+ Newsgroups
----= East and West-Coast Server Farms - Total Privacy via Encryption =----

[Jack Klein]

On 27 Jan 2006 20:40:48 -0800, "clayne" <cl****@anodized.com> wrote in
comp.lang.c:

I wonder whether it is worth replying, since you have done a
commendable job of replying to yourself.

And just for the heck of it, I will point out that there is no "read"
function in C, it is a POSIX extension and hence off-topic here. But
your underlying problem is similar to one that could happen when you
call time(), which could return (time_t)-1.

I realize this may work for the majority of platforms. But I am
concerned about the aspects of sign extension and portability.

Is this portable:

{
unsigned long bw;
[...]

if ((signed long)(bw = read(s, buf, sizeof(buf))) == -1) {
Why, oh why, would you want to do this?
error_handling();
}
}

With the goal being able to utilize larger unsigned values but still
being able to check for a return value of "-1" (since this is the
mechanism read() uses) to handle errors?

Let's replace read with time(), to keep it topical.

#include <time.h>

void func(void)
{
time_t now;
if ((now = time(NULL)) == (time_t)-1)
{
/* error handling */
}
}

Don't case the returned value to another type, cast the comparison
value constant to the type of the return value.

--
Jack Klein
Home: http://JK-Technology.Com
FAQs for
comp.lang.c http://c-faq.com/
comp.lang.c++ http://www.parashift.com/c++-faq-lite/
alt.comp.lang.learn.c-c++
http://www.contrib.andrew.cmu.edu/~a...FAQ-acllc.html