Slide 1: A bit of C programming

Lecture 3

Uli Raich

First Semester 2017/2018

Slide 2: Programming Languages

To make a language a programming language it needs to implement:

• Assignments
• Conditional statements
• loops

Slide 3: C libraries

C uses a large number of code libraries and you can create C libraries yourself.

These libraries may use special data types, which are defined in include files

Before using the library functions

#include <stdio.h>

or

#include “myOwnIncludeFile.h”

Slide 4: The C main program

As a first example people usually write the

“Hello World” program

Let us try to compile and execute this program

First we start the editor, we type and save the program,

then we compile it using the gcc compiler and finally we execute it.

Slide 5: C data types

• C has a number of data types:
• char, short, int, long, float, double
• unsigned char, unsigned short, unsigned int
• can be extended to boolean (in C99 you can #include <stdbool.h>
• No strings! But a pointer to a zero terminated chain of characters
• struct
• union
• enum

And you can define your own data types with typedef

Slide 6: Assignments

We modify the program to do some calculation:

I call the program assignments.c and I compile it with

gcc -o assignments assigments.c

Slide 7: Type casting

Why do we get this strange result for the division? and how can we correct this?

Yes, the reason is that div has got the wrong (namely integer) type

To correct we must first convert a and b to doubles (or floats) before

doing the calculation and div must be a double as well.

Slide 8: Printing formatted output

You have seen that I use the function printf to output

the results of the calculation.

The format string “%d” tells the system to output the

result as a decimal number.

In an exercise this afternoon, where you will implement a

simple calculator and we will use the format %10.4f

There are many additional number formats to output

• strings,
  
  
• decimal,
  
  
• octal and hex numbers …

Look up the man page for printf for an explanation.

Instead of printing on the screen you can also convert the

number to a string using sprintf.

or print to a file using fprintf

(the file must be opened first, see later)

The equivalent call for input also exists and is called scanf.

Slide 9: Conditions: the if statement

Conditions can be tested with if

if (a < b)

printf(“a is smaller than b\n”);

else

printf(“a is bigger or equal to b\n”);

Slide 10: Calculating the Fibonacci numbers

The Fibonacci numbers:

0 1 1 2 3 5 8 13 21 34 …

or xn = xn-1+xn-2

How can we write a program to calculate up to 12 such numbers?

This can easily be done in a for loop

Slide 11: The for loop

Slide 12: The while loop

Can we also do calculations as long as the Fibonacci number

does not exceed a certain value?

Slide 13: Pointers

We can define variables which do not contain the value but

the address of where the value is stored in memory:

char a=5; is the value

char *myText=”Hello World!”;

myText points to the place in memory where Hello World is stored.

Slide 14: Pointer Example

Slide 15: Printing to a file

Instead of printing to stdout with printf you can also print to a file with fprintf:

Slide 16: Command line arguments

The main routine has 2 parameters, which we did not use

yet as well as a return code.

int main(int argc, char ** argv) or

int main(int argc, char *argv[ ]);

int argc is the number of arguments passed

char **argv is a pointer to a list of null terminated C strings.

Slide 17: Command line arguments example

Please note that the argv values are zero terminated strings even

if you give a number! If you expect a number then you have to convert

the string into a number with atoi or atof

Slide 18: Conditional statements

• if, else if, else
• switch (tag)
• Combination of assignment and condition: the ? operator.

Slide 19: Numeric command line arguments and if conditions

Slide 20: The switch statement

Imagine a simple calculator taking 3 parameters:

• 2 numbers
• An mathematical operator: +,-,*,/

We do not want to have too many if statements to find out which operation must be carried out

switch (operator) {
case ‘+’:
do addition;
break;
case ‘-’:
do subtraction;
break;
default:
error;
}

Slide 21: Switch example

Slide 22: Composite variables

We can have a series of values of the same type in an array:

int myArray[10]; the size of the array is fixed!

An array is in reality a pointer to a series of values of the same type.

Or we can define structures where values of different types can be combined

In addition we have enumerations with a certain fixed range of values.

Slide 23: Composite data type example

Slide 24: The ? operator

If (a < b)

c = a

else

c = b;

Can be written in a single statement:

c = (a<b) ? a : b;

Slide 25: The ? operator example

Slide 26: Functions

Up to now the size of our programs did not exceed one page.

Bigger problems must be broken down into smaller, manageable pieces using functions

Example: the operations of our calculator:

double add(double num1, double num2) {

return (num1+num2);

}

The other operations look similar.

These functions may go into the same file as main but also into separate files.

Then they are compiled separately and linked to the main routine

to form an executable program.

Slide 27: Include files

How does the main program know the name of the function and its parameters?

Define these in an include file to be added to the calling program:

and the call itself:

result = add( 5.2, 4.8);

Slide 28: Contents of the include file

This is how the add.h include file could look like:

/*
include file for add.c function
*/

double add(double,double);

Slide 29: Libraries

A big number of libraries are available for use with C

and you can write you own libraries to extend the set.

There are two types of libraries: static libraries and dynamic ones.

The static libraries are named lib name of the library .a

Example libm.a is a library that contains a large set of

mathematical functions

Its dynamic brother is called libm.so

When linking statically the library is added to the executable making

it substantially bigger but independent of library versions

The dynamic library is loaded once into memory and can be used by

several executables. The executable is therefore much smaller but there is a risk

it will not work if it was linked to a wrong library version, the library

is not available or cannot be found.

Slide 30: Linking libraries

Let’s say your program sine.c uses the sine function.

The you must link it to libm

In the static case:

gcc -static -o sine sine.o -lm

In the dynamic case you skip the -static option

gcc -o sine sine.c -lm

-- Uli Raich - 2017-09-07

Comments