The following program compiles (using MS Visual Studio 2010) with no errors or warnings and runs without any error, why? Thanks in advance for explanation.
#include <stdio.h>
int f[3] = { 1,2,3 };
int g[4] = { 12,45,55,65 };
int main(int argc, char *argv[])
{
printf("%d\n", f[5]);
return 0;
}
The output is: 55
f and g are next to each other in memory and when you output f[5] it simply looks for an address that is 5*sizeof(int) bytes from the start of f and prints out the value there (which happens to be the 55 from array g). You should avoid doing this as it is dangerous.
And use code tags [code][/code].
"Owners of dogs will have noticed that, if you provide them with food and water and shelter and affection, they will think you are god. Whereas owners of cats are compelled to realize that, if you provide them with food and water and shelter and affection, they draw the conclusion that they are gods."
-Christopher Hitchens
The output is actually garbage from memory, in the general case. C has no array-index out of bounds checking built in, so you can use an index off the end of an array without the compiler complaining, but the memory you are accessing when you do it could hold anything.
In your case, the program managed to put the two arrays next to each other in memory, so that f[5]==g[2]. This is more likely to happen for small arrays, but it is not guaranteed.
In the future, please use code tags around your code
EDIT: beaten ^_^
C is fun
C has no bound checking... So this is valid as f and g are allocated memory continuously with f below g...Originally Posted by rassul
So f[5] is valid...
Hope this helps...
I would caution against this statement as it is not clear what is meant by "valid" in this context. f[5] is syntactically correct and should compile, but it is logically incorrect and results in undefined behaviour. How f and g are positioned in memory is implementation defined, so you should not rely on this.
Look up a C++ Reference and learn How To Ask Questions The Smart Way
No, it's certainly not valid, just lucky.
It could just as easily output 12, 45, -858993460, or anything else, even just simply crash.
My homepage
Advice: Take only as directed - If symptoms persist, please see your debugger
Linus Torvalds: "But it clearly is the only right way. The fact that everybody else does it some other way only means that they are wrong"
Ok I got my mistake as f and g are global arrays...
But I want to know that is my statement correct if they would have been defined in main function as then memory will be assigned to them in stack ???
Please update me..
Regards
C_Enthusiast
Which statement are you referring to?
Look up a C++ Reference and learn How To Ask Questions The Smart Way
In this case you have a defined array of 3 elements. By definition only elements 0, 1 and 2 are valid all the rest are out of bounds. As the others have pointed out this leads to undefined behavior since you have no guarantee what that spot in memory holds...
Also, consider the example given above... where f[5] overlaps with g[2]... what does f[5] = 0; do to the other variable...
One of the most important lessons in life says: "Just because you _can_ do something, does not mean you _should_".
Either you put it on the stack or not, you still access an out-of-range. This can be one of the hardest bug to catch.
"All that we see or seem
Is but a dream within a dream." - Poe
I am referring to this statement
f and g are allocated memory continuously with f below g...
rassul's observation implies that that appears to be the case. As to whether your statement will be correct in some other case: it depends on the implementation.
Look up a C++ Reference and learn How To Ask Questions The Smart Way
