if ((fd = open(...)) != -1) {
/* do something with fd */
} else {
perror("open");
}
The compiler outputs a warning when you have something like "if (a = b)". If that's what you mean (it sometimes is), you have to write it "if ((a = b))" to silence the warning. int fd; /* fd must be declared in order to be assigned */
if ((fd = open(...)) != -1) {
/* do something with fd */
} else {
perror("open");
}
which would be better written as: int fd = open(...);
if (fd != -1) /* etc */
It would be nice for scoping reasons to be able to write something like: if ((int fd = open(...)) != -1) /* etc */
but if ((int fd == open(...)) != -1) /* etc */
isn't valid code.Anyways, I am a bit torn about the second option. I like the idea of putting the call inside the if clause as it makes for a very explicit error handling but the uninitialized declaration is ugly. What I do in practice tends to depend on the situation but it is rarely satisfying.
Your last suggestion would be ideal, but as you said, it is invalid code unfortunately.
Maybe this
for (int fd = open(...); fd != -1; fd = -1) {
/* do stuff */
}
Just kidding, don't do that. > for (int fd = open(...); fd != -1; fd = -1) {
that doesn't correctly (or rather at all) handle the failure case. You should do (IIRC): #define LET(...) for(__VA_ARGS__,_tmp[0],*_once=_tmp; _once ;_once=0)
/* ^^^ goes in a header file */
LET(int fd = open())
{
if(fd == -1) { /* handle error, return or break */ }
/* do stuff */
} char *strcpy(char *d, char *s) {
char *r = d;
while (*d++ = *s++);
return r;
}
(If you don't need the return value, this becomes even nicer.)while((i = getchar()) != EOF) putchar(i);
This type of thing seems to be "encouraged"...