Datatypes sizes varies in visual studio 2005 problem confusing

[cfresh]

hi,
i have started learning c for 1 week and i am using VS 2005.
from tutorials i learned that size of char is 1 byte
i am using 32bit system
but i tried this program:

# include<stdio.h>

void main()
{
       char a,b;
a=0xfffffffffffffffe; //now a is holding 8bytes thst is sizeof a double datatype
b=0x1;
b=a+0x1;
printf("size of  varaible A:%d\n",sizeof(a));
printf(value in a is:%x\n value in b is:%x);
return;
}

output was
size of varaible A:1
value in a is:fffffffe
value in b is:ffffffff

my question are:
1)when i added one more f in "a" variable a=0xfffffffffffffffef then when i compiled it showed me the error constant too big(bcoz its more than 32 bit) then why it didnot show earlier this erro message when i assigned value(0xfffffffffffffffe) to char a which greater than it can hold ie 0xff(1 byte size maximum valu that char can hold)
2)
even if "A" has holded the whole 8 bytes in it then why did take only last 4 bytes for calculation?
Or VS 2005 itself expands the datasize from char to float or double ?
Please help

[qny]

First: get into the habit of compiling with all warnings turned on.

Second: the function main should return a value of type int: say so with

int main() { /* ... */ }

Third: the range of values a variable of type can hold varies from computer to computer. In some it it -128 to 127; in some others it is 0 to 255, in yet some (rarer) others it is 0 to 512 or 0 to 65535, etc ... Your computer, very very likely, has the first range, and assigning a value outside the range is a big no no. What happens for strictly positive ranges is that the assigned value is calculated modulo the (maximum value + 1); and for other ranges is Undefined Behaviour. Anyway: writing code that assigns a value with more bytes than the variable can hold does not make the variable large: the excess bytes are ignored.

Fourth: the values you specify in the second printf() are automatically converted to values of type int; but you asked for values of type unsigned int (with the "%x"). The type of the value provided and the type requested do not match: another big no no.

To sum up: if you lie to the compiler, it will get its revenge.

[cfresh]

First: get into the habit of compiling with all warnings turned on.

Second: the function main should return a value of type int: say so with

int main() { /* ... */ }

Third: the range of values a variable of type can hold varies from computer to computer. In some it it -128 to 127; in some others it is 0 to 255, in yet some (rarer) others it is 0 to 512 or 0 to 65535, etc ... Your computer, very very likely, has the first range, and assigning a value outside the range is a big no no. What happens for strictly positive ranges is that the assigned value is calculated modulo the (maximum value + 1); and for other ranges is Undefined Behaviour. Anyway: writing code that assigns a value with more bytes than the variable can hold does not make the variable large: the excess bytes are ignored.

Fourth: the values you specify in the second printf() are automatically converted to values of type int; but you asked for values of type unsigned int (with the "%x"). The type of the value provided and the type requested do not match: another big no no.

To sum up: if you lie to the compiler, it will get its revenge.

Thank you for points i will follw them from now own.