case label does not reduce to an integer constant

[Rex Mottram]

Can anyone explain this?

% cat t.c
#include <stdio.h>

#define FIRST "first"
#define SECOND "second"

int
main(int argc, char *argv[])
{
switch (argv[1][0]) {
case FIRST[0]:
case 'f':
printf("%s == FIRST\n", argv[1]);
break;
case SECOND[0]:
case 's':
printf("%s == SECOND\n", argv[1]);
break;
}
return 0;
}

% gcc -o t -W -Wall -g t.c
t.c: In function `main':
t.c:10: error: case label does not reduce to an integer constant
t.c:14: error: case label does not reduce to an integer constant

It certainly looks constant to me, and it should be promotable to an
integer exactly as the neighboring case label is. This is gcc 3.4.5 FWIW.

TIA,
RM

[Malcolm McLean]

Re: case label does not reduce to an integer constant


"Rex Mottram" <rexm@not.herew rote in message
news:5JmdnaLkgp QwPJ3VnZ2dnUVZ_ uKpnZ2d@comcast .com...

Can anyone explain this?
>
% cat t.c
#include <stdio.h>
>
#define FIRST "first"
#define SECOND "second"
>
int
main(int argc, char *argv[])
{
switch (argv[1][0]) {
case FIRST[0]:
case 'f':
printf("%s == FIRST\n", argv[1]);
break;
case SECOND[0]:
case 's':
printf("%s == SECOND\n", argv[1]);
break;
}
return 0;
}
>
% gcc -o t -W -Wall -g t.c
t.c: In function `main':
t.c:10: error: case label does not reduce to an integer constant
t.c:14: error: case label does not reduce to an integer constant
>
It certainly looks constant to me, and it should be promotable to an
integer exactly as the neighboring case label is. This is gcc 3.4.5 FWIW.
>

Yes, you've managed to fool it.

However sections of string literals aren't legal in switch statements, only
character constants. It reduces the work the parser has to do.

--
Free games and programming goodies.

http://www.personal.leeds.ac.uk/~bgy1mm

[Pradeep]

Re: case label does not reduce to an integer constant

On Apr 12, 5:49 pm, "Malcolm McLean" <regniz...@btin ternet.comwrote :

"Rex Mottram" <r...@not.herew rote in message
>
news:5JmdnaLkgp QwPJ3VnZ2dnUVZ_ uKpnZ2d@comcast .com...
>

Can anyone explain this?

>

% cat t.c
#include <stdio.h>

>

#define FIRST "first"
#define SECOND "second"

>

int
main(int argc, char *argv[])
{
switch (argv[1][0]) {
case FIRST[0]:
case 'f':
printf("%s == FIRST\n", argv[1]);
break;
case SECOND[0]:
case 's':
printf("%s == SECOND\n", argv[1]);
break;
}
return 0;
}

>

% gcc -o t -W -Wall -g t.c
t.c: In function `main':
t.c:10: error: case label does not reduce to an integer constant
t.c:14: error: case label does not reduce to an integer constant

>

It certainly looks constant to me, and it should be promotable to an
integer exactly as the neighboring case label is. This is gcc 3.4.5 FWIW.

>
Yes, you've managed to fool it.
>
However sections of string literals aren't legal in switch statements, only
character constants. It reduces the work the parser has to do.
>
--
Free games and programming goodies.http://www.personal.leeds.ac.uk/~bgy1mm

Hi,
switch case in c language accepts only integer values.
If you are passing character variable then you have put the same in
single quotations (i.e.,'')
That will indicate ascii value of the character.
But in your program you have written as
case FIRST[0]:
here "FIRST[0]->f" this is character.so you need to put the same in
single quotations.
so replace your code from
-------------------
case FIRST[0]:
-->to
case 'FIRST[0]':
-------------------
similarly chanage
-------------------
case SECOND[0]:
-->to
case 'SECOND[0]':
---------------------

--Pradeep

[Richard Heathfield]

Re: case label does not reduce to an integer constant

Pradeep said:

<snip>

>
switch case in c language accepts only integer values.

Yes, case statements must be an integral constant expression.

If you are passing character variable then you have put the same in
single quotations (i.e.,'')

You can't "pass" a character, or indeed anything, to a switch.

That will indicate ascii value of the character.

No, it won't.

But in your program you have written as
case FIRST[0]:
here "FIRST[0]->f" this is character.so you need to put the same in
single quotations.
so replace your code from
-------------------
case FIRST[0]:
-->to
case 'FIRST[0]':

That isn't even close. In fact, it's so far the wrong side of quite a long
way out that even to see just how wrong it is you'd need extremely
powerful erroculars.

--
Richard Heathfield <http://www.cpax.org.uk >
Email: -http://www. +rjh@
Google users: <http://www.cpax.org.uk/prg/writings/googly.php>
"Usenet is a strange place" - dmr 29 July 1999

[Richard Heathfield]

Re: case label does not reduce to an integer constant

Rex Mottram said:

Can anyone explain this?
>
% cat t.c
#include <stdio.h>
>
#define FIRST "first"
#define SECOND "second"
>
int
main(int argc, char *argv[])
{
switch (argv[1][0]) {
case FIRST[0]:

"An integral constant expression shall have integral type and shall
only have operands that are integer constants, enumeration constants,
character constants, sizeof expressions, and floating constants that
are the immediate operands of casts."

A string literal is none of those. The fact that you're indexing it is
neither here nor there.

--
Richard Heathfield <http://www.cpax.org.uk >
Email: -http://www. +rjh@
Google users: <http://www.cpax.org.uk/prg/writings/googly.php>
"Usenet is a strange place" - dmr 29 July 1999

[Bartc]

Re: case label does not reduce to an integer constant


"Rex Mottram" <rexm@not.herew rote in message
news:5JmdnaLkgp QwPJ3VnZ2dnUVZ_ uKpnZ2d@comcast .com...

Can anyone explain this?
>
% cat t.c
#include <stdio.h>
>
#define FIRST "first"
#define SECOND "second"
>
int
main(int argc, char *argv[])
{
switch (argv[1][0]) {
case FIRST[0]:
case 'f':
printf("%s == FIRST\n", argv[1]);
break;
case SECOND[0]:
case 's':
printf("%s == SECOND\n", argv[1]);
break;
}
return 0;
}
>
% gcc -o t -W -Wall -g t.c
t.c: In function `main':
t.c:10: error: case label does not reduce to an integer constant
t.c:14: error: case label does not reduce to an integer constant
>
It certainly looks constant to me, and it should be promotable to an
integer exactly as the neighboring case label is. This is gcc 3.4.5 FWIW.

Possibly not allowed for the same reason that:

int x[FIRST[0]];

is not allowed, when declared outside a function (inside a function, it
might work as a VLA, but it is still not considered a constant).

I suppose there is a limit to what a compiler can do to determine an
expression is constant, or maybe it considers that the string "first" could
change (I managed to do this on lccwin, but gcc generated a crash when
attempting to change the constant string).

--
Bart