scanf (continued) and

Data Types in C

ESC101: Fundamentals of Computing

Nisheeth
How does scanf work ?
HOW WE MUST SPEAK TO MR.
HOW WE USUALLY SPEAK TO A HUMAN
COMPILER

scanf(“%d%d”, &a, &b); Please read one integer. Ignore all

whitespace (spaces,tabs,newlines) after
that till I write another integer. Read that
Format string tells second integer too.
Format string me how you will
Store value of the first integer in a and
write things, and value of second integer in b.
then I am told
where to store
what I have read

Remember Mr. C likes to be told beforehand what all we are going to

ask him to do!
Scanf follows this exact same rule while telling Mr. C how to read
How does scanf work ?
Be a bit careful since Mr C is a bit careless in this matter
He treats allMywhitespace
advice to youcharacters
is to the same when integers are being
input take input one at a time in
Hmm … you are going
the beginning  Try out
scanf will never print anything to write the English
acrobatics in free time
word Hello followed by
scanf(“Hello %d”,&a); space followed by an
integer. I will store the
Use printf to print and scanf to read value of that integer in a
Try out what happens with the following
scanf(“%d %d”,&a,&b); scanf(“%dHello%d”,&a,&b);
scanf(“%d,%d”,&a,&b); scanf(“\“%d%d\“”,&a,&b);
scanf(“%d\n%d”,&a,&b); scanf(“%d\t%d”,&a,&b);
Commenting
Very important programming practice
int main(){ int main(){ int main(){
int a; // My first int int a; /* My first int */ int a; // My first int
int b; // The other int int b; /* The other int */ int b; // The other int
// Assign them values /* Assign them values */ /* Assign them values */
a = 5, b = 4; a = 5, b = 4; a = 5, b = 4;
Yes. In fact /* */ is used
a + b; So I can mix a + b; to comment several a + b;
return 0; return 0; return 0;
and match? lines at once – shortcut!
} } }
Just be a bit careful.
Some compilers don’t
understand // comments 4
More on Comments
Use comments to describe why you int main(){
defined each variable and what each step int a; // My first int
of your code is doing
You will thank yourself for doing this when you are int b; // The other int
looking at your own code before the end sem exams /* Assign them values
Your team members in your company/research group
will also thank you so that I can add
them later on */
Multiline comments very handy. No need
to write // on every line a = 5, b = 4;
a + b;
return 0;
}
5
A Useful Tip While Problem-Solving
Comments can be also used to identify where is error
Error! Okay! Okay!
Mr C will tell you (compile) where he thinks the error is
Commenting out lines can also help identify the error

int main(){ int main(){ int main(){

Aha! I forgot
int a, b; int a, b; int a, b, c;
to declare c
c = a + b; // c = a + b; c = a + b;
a = 5; a = 5; a = 5;
b = 4; b = 4; b = 4;
return 0; return 0; return 0;
} } } 6
A Useful Tip While Solving Problems
Print your solutions to
each one of these pieces
to see where going wrong

Try breaking up
I have no idea the problem into
what is going smaller pieces
wrong here!
7
A Useful Tip While Solving Problems

Equals 0
A Useful Tip While Solving Problems

Replace this part by (2*x*x*x)/3

Basic Data Types in C
 Int: %d specifier
 Integers like 156, -3, etc
 float (short form of “floating point number”) and double: %f specifier
 Real numbers like 3.14, 2.0, -1.3, etc
 double is like float but has larger range
 char (short form of “character”): %c specifier
 Single letter (a-z or A-Z), single digit, or single special character
 A char is always enclosed in inverted single commas
 Some examples: ‘a’, ‘A’, ‘2’, ‘$’, ‘=‘
 These basic data types can also be used with a modifier
 Modifiers change the normal behaviour of a data type (e.g., its range of values) and
memory storage space required (more on next slides)
Type Modifiers in C
 signed (used with int, float/double, char)
 signed means the data type can have positive and negative values
 int, float/double, char are signed by default (no need to write ‘signed’)
 unsigned (used with int, char)
 unsigned means the data type can have only take positive values
 short (used with int)
 short means it uses only half of the memory size of a normal int
 long (used with int)
 long means it uses twice the memory size of a normal int
 Can store a larger range of values of that type
11
Various C Data Types without/with Modifiers
int (signed int) unsigned int short int (short) long int (long)
%d %u %d %ld

short unsigned long unsigned Yes, multiple modifiers

also allowed
%u %lu

float double long double

%f %lf %Lf

char unsigned char

%c %u
12
int
• Can store integers between -2,147,483,648 and 2,147,483,647

#include <stdio.h> Range: -2^31 to (2^31)-1

int main(){
%d signed int uses 32 bits
int a;
scanf(“%d”, &a);
a (4 bytes, 8 bits = 1 byte)
on recent compilers)
printf(“My first int %d”, a); Integer arithmetic applies to
return 0; integers +, -, /, *, %, ()
} Have worked with them a lot so far
Printing well-formatted outputs using printf
 When printing an int value, place a number between % and d (say %5d)
which will specify number of columns to use for displaying that value
Output
int x = 2345, y=123; 2345
printf("%d\n",x); //Usual (and left aligned) 2345
123
printf("%6d\n",x); //Display using 6 columns (right aligned)
2345
printf("%6d\n",y);

printf("%2d\n",x); //Less columns than digits, same as %d

Note: So far, we have only seen how to print integers.

We will see how to print other types of variables later today 14
long int (usually written just long)
• Really long – can store integers between
• -9,223,372,036,854,775,808 and 9,223,372,036,854,775,807
Range: -2^63 to (2^63)-1
#include <stdio.h>
int main(){ long int uses 64 bits
on recent compilers
long a; //long int also
scanf(“%ld”, &a);
a Integer arithmetic applies to long int as
well +, -, /, *, %, ()
printf(“My first long int %ld”, a); Try them out on Prutor
return 0; %ld How does long work with int int + long,
int * long?
}
Will see in next class…
float float uses 32 bits (4 bytes).
Why this range for double?
• int, long allow us to store, do math formulae Will
withseeintegers
reason later
• float allows us to store, do math formulae with reals

#include <stdio.h> Very large range ± 3.4e+38

int main(){ Arithmetic operations apply to
float as well +, -, /, *, ()
float a;
scanf(“%f”, &a);
a %f
Try them out on Prutor

printf(“My first real %f”, a); What happened to

I remember. Remainders
remainder %?
return 0; Did you ever do
remainders with real make sense for integers,
} numbers in school? not for real numbers
 Too many decimal digits
printf revisit: Printing of float/double being printed. Can I just
print one or two?
Correct! Value of a = 123.46
123.456700 1.234567e+02
Value of a = 1.23e+02
• Both %f and %lf work for float and double
• For long double,
Yes. Use %0.2f
%Lf to print
needed Oh right. The usual
2 decimal places rules of rounding apply
• Can use %e if want answer in exponentialhere notation
too. 1.5644 will
#include <stdio.h> Sure. Just like I did it for #include <stdio.h>
become 1.56 if rounded
integer display. See next to 2 places but 1.5654
int main(){ slide  int main(){will become 1.57
double a = 123.4567; double a = 123.4567;
Be printf(“Value Great. So can you also
of a = %0.2f”,
careful, I am rounding %f”, a); a);
while giving printf(“Value
help
of
me
acontrol
= %0.2e”,
%e”,the
a);format
a);
answer correct to 2 decimal places
return123.4567
0;  123.46 return 0; in which float/double is
1.234567  1.23 displayed on screen?
} } 17
printf revisit: Controlled printing of float/double
 Already saw how to use printf for well-formatted int display
 Can also control how to display a float/double using printf
 Can do it using “%a.bf” specifier where a and b are numbers
 Here a is the total field width (number of columns) in which the float will
be displayed, b is the number of digits printed after decimal

float pi = 3.141592; Output

printf("%f\n",pi); //Usual
3.141592
printf("%6.2f\n", pi); //2 decimal
3.14
printf("%0.4f\n",pi); //4 decimal 3.1416
// Note rounding off!
18
double
• Double can also handle real numbers but very large ones
• Similar relation to float as long has to int

#include <stdio.h> Very large range ± 1.79e+308

int main(){ Arithmetic operations apply to
double as well +, -, /, *, ()
double a;
scanf(“%f”, &a);
a There is something called long
double as well
printf(“My first real %f”, a); Use %Lf to work with long
return 0; %lf works doubles
too!
} Try these out on Prutor
char
• Basically, a char is a symbol
• Internally stored as an integer between -128 and 127 (if signed char) or between 0
and 255 (if unsigned char)

Char constants enclosed in ‘ ’

#include <stdio.h> ‘p’ Integer arithmetic applies to char as
int main(){ %c well +, -, /, *, %, ()
char a = ‘p’; a Case sensitive ‘a’, ‘A’ different
printf(“My first char %c\n”, a);
printf(“ASCII value of %c is %d”,a,a);
Various usages (e.g., in arrays of
characters – strings), will see more
return 0; This Will print the ASCII
later
value (integer) of this
} character
 American Standard Code
for Information Interchange

\t

\n

Image courtesy wikipedia.org

ASCII Table with Extended Characters

IMAGE COURTESY: https://theasciicode.com.ar/

Mixing Types in C Expressions
 We can have C expression with variables/constants of several types
 Certain rules exist that decide the type of the final value computed
 Demotion and Promotion are two common rules

• int a = 2/3; // a will be 0 (no demotion/promotion)

• float a = 2/3; // a will be 0.0 (RHS is int with value 0, promoted to float with value 0.0)
• int a = 2/3.0; // a will be 0 (RHS is float with value 0.66, becomes int with value 0)
• float a = 2/3.0; // a will be 0.66 (RHS is float with value 0.66, no demotion/promotion)
• int a = 9/2; // a will be 4 (RHS is int with value 4, no demotion/promotion)
• float a = 9/2; // a will be 4.0 (RHS is int with value 4, becomes float with value 4.0)

 During demotion/promotion, the RHS value doesn’t change, only the

data type of the RHS value changes to the data type of LHS variable 23
Type Casting or Typecasting
Also remember: When
 Converting values of one type to other. assigning values, I always
compute the RHS first
 Example: int to float and float to int (also applies to other types)
 Conversion can be implicit or explicit. Typecasting is the explicit
way Automatic (compiler) By us

 int k =5;
 float x = k; // good implicit conversion, x gets 5.0
 float y = k/10; // poor implicit conversion, y gets 0.0
 float z = ((float) k)/10; // Explicit conversion by typecasting, z gets 0.5
 float z = k/10.0; // this works too (explicit without typecasting), z gets
0.5 24
Typecasting: An Example Program
Also, typecasting But be careful about which
just one variable one you are typecasting
#include <stdio.h> on RHS is enough #include <stdio.h>
Typecasting
int main(){ int main(){ makes it
Several other ways
int total = 100, marks = 50;
also possible, e.g., int total = 100, marks=50;
50.0/100 which
float percentage; Equals 0 proper bracketing. float percentage; equals 0.5
percentage = (marks/total)*100; percentage = (float)marks/total*100;
printf("%.2f",percentage); printf("%.2f",percentage);
return 0; return 0;
0.00 50.00
} }

25
Typecasting is Nice. But Take Care..
#include <stdio.h> Expected #include <stdio.h>
int main(){ conversion int main(){
float x; int y; float x; int y;
x = 5.67; x = 1.0e50; // 10^50
y = (int) x; // typecast (convert) float to int y = (int) x; // typecast (convert) float to int
printf(“%d”,y); printf(“%d”,y);
return 0; 5 return 0; -2147483648
} }

Reverse typecasting error can happen too: No. 1.0e50 is too big to be
Sometimes converting a smaller data type cast as an int (or even Are you
(say int) to larger data type (say float) can long – try yourself) kidding?
also give unexpected results (more on this
later in the semester)
Unexpected!
26