Storing numbers in a character array

[KeilanS]

I'm writing a program that needs to take in a 2000 x 2000 multidimensional array.

All the entries in the array are numbers between -1 and 100.

So I use a nested loop to take all the numbers and store them in the array. But when I output it gives me all single characters, or it will give me the ASCII character. Not the number.

[CPD]

I'm writing a program that needs to take in a 2000 x 2000 multidimensional array.

That's a lot of space. Do you have to have all of the numbers in memory at once?

But when I output it gives me all single characters, or it will give me the ASCII character. Not the number.

I guess you're using an array of char to hold the numbers? char is treated specially on the assumption that you want a character and not a small integer, so you need to do some casting to int when printing.

[KeilanS]

I do have all the numbers in memory at once. I realize it's taking up about 4MB of memory.

And I will try the cast thing. Thanks.

[KeilanS]

Yep, that helped! I also found out another problem. The program was storing the -1 as 2 characters.

The first one was '-' and the second was '1'. So I fixed that by inputting into an int, then storing the int as a char.

[dargueta]

A nonstandard way of fixing that would be to declare the array as an array of __int8. I'm pretty sure that's just a Microsoft extension.

char myData[2000][2000]; //requires typecasting
__int8 myOtherData[2000][2000]; //no typecasting

__int8 is signed by default.

[KeilanS]

My computer prof requires us to compile and run our programs under Fedora Linux. Thanks though, I'll keep that in mind.

[TkTech]

From my base/types.h file, these are guaranteed to be the proper size by C/C++

typedef unsigned int   u32int;
typedef          int   s32int;
typedef unsigned short u16int;
typedef          short s16int;
typedef unsigned char  u8int;
typedef          char  s8int;

[dargueta]

C99 specifies minimum widths. These could be off.

[MeTh0Dz]

Check out 7.18.1.1 and 7.18.1.2 for information on the exact specifications. Although it is basically implementation defined, but required if the width exists.

Format: intN_t and uintN_t ; where N is the width

[KeilanS]

Sorry guys, these last 3 messages are over my head.

Beyond what I've learnt in the class so far.