THE C++ BLOG

File extensions - a confusing affair

2009-02-14T05:22:00.000-08:00

Did you wonder what ".GOOK" and ".GIFAR" could mean? nope, you didn't!! But as a geek you need to know all the exotic extensions..
I got into this mess when I installed a script editor and it changed the default setting for a host of my other files like .xml and .class files. The next time I tried to compile and view my code, I got a response that no suitable program to open such extension found!
Loads of hours on the internet later I found this great site that has a registry fixer software. My recommendation - install this software asap - it even fixed my File Extension BIN that had started pointing to different places.. I use Avira anti-virus software which also uses BIN file extension and this extension is so very common but when I forgot to uncheck "always use this program to open this kind of file" on the Open With dialogue, I struck disaster..

This registry fixer software is just great. All registry errors fixed now.. and I can go on with my coding without worrying about how to fix my registry files :)

If you want to know about the full forms and which company devised the file formats then that information is too very nicely given here..
One very interesting trivia I found about BIN files here is that they can be opened in 12 different ways.. that means there are 12 different uses of the same file format.

You can also get info about file extensions through the alphabetical file index on the same site. :p

Implementing Dictionary in C++

2008-08-11T07:18:00.000-07:00

Working with C#, I loved the feature of the data-type : Dictionary. I was wondering how we can implement a Dictionary in C++.

Heres what could be a way out...

1. Following the C# implementation as the model, we create a class having two elements : key and value, and implement it as a doubly linked-list. This will optimize structure parsing of the dictionary.

2. We can use Templates to make it a case with generic key and value types.

3. Another internal optimization would be to use hash.

4. Sort the dictionary based on the key values.

5. Provide functions to set and unset key-value pair.
--As I noticed that dictionary is immutable during access, provide a stack say _keys_to_delete on top of which we can push the keys of entries to be deleted. Implement a garbage collection of sorts that runs a loop to delete the entries based on the keys found in the stack _keys_to_delete after the dictionary has finished iterative access (so that it again becomes mutable)...

Missing The 'String' data type in C++

2008-07-21T13:24:00.000-07:00

After working with languages like Java and C# for some time, one tends to assume to comfort of the 'String' data type. Coming back to some hard-core programming in C++ after some time, I wished that C++ be evolved into something that is more programmer friendly but at the same time maintains the speed and efficiency that is currently provides..

With many OpenSource and other wise initiatives to use modern languages for high-end development like in writing the new-gen OSes, the forte of C++, pointers, seems to be a mixed blessing.

Overloading the NEW operator - Part II

2008-03-15T08:19:00.000-07:00

The NEW operator has, among many advantages, then ability to be overloaded class-wise. In other words, you can overload the NEW operator on a per-class basis, so that your new operator is specifically tailored to respond to what type of object you are applying the NEW operator to.

So you can overload the NEW operator for each class to perform in special situations like handling the case when memory runs out and so on..

Sample code:

class myClass
{
char abc;
int def;
public:
void *operator new(size_t bytes)
{
.....specific code.......
}

};
class otherClass
{
char a;
int b;
public:
void *operator new(size_t bytes)
{
.....specific code.......
}
};
void main()
{
myClass obj1= new myClass();
otherClass obj2 = new otherClass();
.......................
.......................
}

Overloading the NEW and DELETE operators in C++

2008-03-06T08:11:00.000-08:00

Operator Overloading - One of the important pillars of the Object Oriented Concept is a very useful tool in the hands of the right programmer.

The fact that you can even Overload the NEW operator and the DELETE operator as well adds convenience to the custom memory management.

1. Overloading the NEW Operator:

void *operator new(size_t size, int setvalue)
{
void *pointer;

p = malloc(size);
if(p==NULL)
outofMemory( );
memset(pointer,setvalue,size);
return(p);
}

void outofMemory( )
{
cout<<"OOPS ran out of free memory..";
exit(1);
{.....perform other memory freeing operations here and try to re-allocate the memory....}
}

To set the _new_handler pointer to default

2008-03-01T09:58:00.000-08:00

In my last post, I talked about using the _new_handler pointer.. one thing I forgot to mention was to un-set the _new_handler pointer back to default value after pointing it to a user defined function.

For that, use -

set_new_handler(0);

This will return the pointer to its default state containing NULL.

What to do when your C++ memory underruns..

2008-02-26T07:24:00.000-08:00

While playing with Pointers, the worst fears of any programmer are underruns and overflows....

But there is a very elegant solution to memory exhaustion in C++, wont provide you more memory but will at-least ensure that this exception is well handled..Here is the way out..

Suppose that your new operator used for creating new objects in the memory runs out of free memory so what to do then?

C++ has an internal function pointer called the _new_handler. Usually it contains a NULL which is returned by new operator when it fails to allocate memory for the requested object.
Now there is another special construct called the set_new_handler ( ) that lets you set the _new_handler to point to a user defined function which will be called in-case the new operator fails to allocate the memory. So that particular function will be called when such a case occurs.

Here is a code snippet to explain better:

void main( )
{
void outpfMemory( );
set_new_handler(outofMemory);

char *pointer = new char[..some large value like 64000u...];
}
void outofMemory( )
{
cout<<"OOPS Ran out of memory..";
exit(1);
}

The function outofMemory will be called when the char Pointer assignment fails due to lack to available memory in the free store.

The C++ Memory Free Store

2008-02-19T07:08:00.000-08:00

It is a concept that is very conveniently skipped while learning C++ at the beginner's level. Having used the 'new' and 'delete' operators, it is imperative to understand how the C++ memory is managed.

1. What does the 'new' operator actually do??

Ans. It basically does two things - allocate memory for the new object and call the constructor of that object and thereby in effect 'create' the object.

The C++ Memory Heap a.k.a the 'free-store' is used to assign chunks of memory size as requested by each constructor that is called by the 'new' operator on that object.

In comparison to the malloc() function, the new operator returns a type-safe pointer while th malloc() return a pointer of type 'void' which needs to be explicity type-casted by the programmer.

For the special case when the memory store available with the free store is exhausted, we can test if the new operator returns a NULL.. But there is a more elegant way to handle this case.. look for my next post on how to handle the situation of free store exhuastion.

Some project ideas in C++

2008-01-29T10:47:00.000-08:00

Here is a collection of project ideas that can be easily implemented in c++ to try your hands at the game.. I will list them in order of difficulty (low to high)

Basic operator overloading - try creating as many as possible combinations of function name and rest of the function signature - purely syntactical exercise.
Create a function that can convert an "int" type into a "string" containing the Binary representation and vice-versa.
Implement a rotated array solver - a rotated array means the elements of a "sorted" array are rotated and wrapped around from last position to 0th by rotating factor - 'r'

So after rotation, the array's element Ai will be at ( ( i + r ) % arraylength ). Now suppose you are given such an array, which is already rotated by some unknown rotation factor - 'r', create a function to bring the array to original sorted form without using sorting directly. Keep in mind performance parameters in terms of speed and memory and elegance of code.

A sneak peek on Data Structures

2006-08-28T01:30:00.000-07:00

Data Stuctures...hmm when i started writing this post i thought why am i jumping to data structures so early? But then i thought since i was being a lazy bum for quite a while so why not Data structures; it's better than nothing! (and i can't not think of anything else at the moment)

So , Data structures are here... The first question that one might ask is; seemingly what are data structures?? But there is an obvious answer in the name of the topic itself!

Data- it means any set of raw values in any format which we are interested in so that we can collectively process it to yield useful information!

Structure- it means any way of organizing. Be it data, variables or any heirarchy! It can be anything. Did you know that orgzanizing residential space can also be considered a sturcture, of course, high rise buildings are called super-structures in general terms..
Ok, so we have data and structures; 1+1=2. Yes, it is as simple as that! Organizing data according to specific parameters to yield useful information is creating a data structure.

Types: Some simple types are- Que, Linked list, Stack, Circular que, Tree etc;

-> que is simple, its just like data standing in line, with the first data packet getting processed first.
-> Linked list; when one data packet links to other, it is called a linked list. Its like elephants moving in circus holding each others' tail. ;)
-> Stack, just like plates stacked on top of each other.
-> Circular cue; what starts also ends at the same point.
-> Tree; the first thing that comes to one's mind when they hear the word tree is 'branch'. So data branching out towards other data is called a tree structure.

Just keep these simple examples in mind and you will have a very good understanding of D.S when you read about it's implementation.

Calling Polly - Polymorphism

2006-07-16T11:15:00.000-07:00

'POLYMORPHISM' - The computer geek's buzz word.
one thing that drives the software industry is polymorphism! It is a very strong concept. To say, it is not as hard to master as it sounds, and at the same time it is not as easy to master as it seems from this statement. Boggled ?? Well dont be ! Lets make polymorphism simple.
POLYMORPHISM -> 1. FUNCTION OVERLOADING.
2. SCOPE RESOLUTION.
An easy way to understand things is to divide them into logical parts, ans as they say divide and rule....
1. Function overloading - this is what packs most of the punch.

Function overloading means using the same function name to call different functions according to need. It can be understood like this, your dining table functions as a table/surface on which food is served and eaten, while some times it also serves as table/surface on which the family board games are played. The world is full of such implementations of overloading. Your dad uses the computer to do his office work, you use it for playing games.
The point is that you use the same thing for different roles as per need.

int myfunc (int a, int b)
{
return (a+b);
}
int myfunc (char a, char b)
{
cout<}

here the functions have the same name but different parameter list and different body. With practice and further analysis you will understand it better.

2. Scope resolution - same variable name, but different variable (in distinct scope)

C++ provides a way to declare two variables having the same name but different their root scopes are disctinct, although they can overlap.

{ int a=1;
{
int a=2;
{
int a=3;
cout< }
}
}

Note: some programmers dont feel that scope res. should be called a part of polymorphism.

One difference between C and C++

2006-05-13T00:38:00.000-07:00

It is worth noting that one major difference between C and C++ is the way the code is documented, that is to say that the layout of various items like declaration and initialization as well as the position of the main function is fixed in C.

while C++ provides a more flexible approach in the same matter. while coding in c++ you can initialize anywhere in between the code, and declare and define any function keeping in mind just that it is not called before it's declaration. so that is an added benefit on coding in C++.
actually this can be thought of to be a direct consequence of the ability of a C++ code to perform late binding. SO you can call the free store of memory any time you want.

example:
In C the following code is invalid, but valid in C++:

void main( )
{
int a;
clrscr( );
int b;
}
here initialzing the interger 'b' after calling the function clrscr( ) would lead you to an error.
while at the same time, it's a perfect little code in case of C++.

that is why in case of C++ you can write the for loop like this:

for(int i=0; i<10;>

Keep this fact in mind to take your programming skills to a better level.

Knowing data abstraction

2006-03-10T11:11:00.000-08:00

Data abstraction is an important part of OOP- Object Oriented Programming.
Herein a group of data holding members are thought of as Objects having their own behaviour and interaction protocols towards other objects and the global environment.
These objects exhibit their behaviour through functions which can access them.
So basically Abstraction is the process of hiding the information that is at work in the background. Example: A car is known as a system(object) comprising of four wheels(functions to interact with th road), a steering wheel (a function that interacts with you, the user) ... the gearstick(function that interacts with the user again), but you do not know the nitty-gritty of the complex gear system yet you can easily use it to your advantage. Now this is a working and moving example of abstraction....

BASICS TO KNOW

2005-11-02T07:28:00.000-08:00

THE INCLUDE SECTION:

1> WHAT ARE LIBRARY FUNCTIONS?

They are the routinely used functions which are made use of in almost all the programs.
They are stored in the header files which are included in the standard package in the folder generally named as ' Include'. Thewy header files store the defination of these library functions so you can use them by simply including those header files in your program and then calling those functions with whatever arguments needed.

example : #include

This header file is better known as : Input-Output-Stream.header

So, as the name suggests, it includes the defination of the commonly used input and output functions.

Thus, this facility helps you to avoid writing the code for the same functions again and again.
And the good news is that you can even create your own header files which define commonly used functions, classes and variales used generously all over your proejct.

STRUCTURE OF A GOOD 'C/C++' PROGRAM

2005-09-22T08:48:00.000-07:00

There are certain basic aspects that should be kept in mind; such as the format of the code, indentation, comments, smart variables etc..

As for the basic layout, here is a tip:

1> Documentation:

This should contain the objective of the program, the creator's name and date if neccessary and other details about the program. It should be in the form of a multi-line comment.

Example: /* Program to add two numbers.
created on: 22 sept-2005, by:urjit */

2>Linking of files and Inclusion:

This section should be used to link to all header files, external files and the library functions. Macros could also be placed in this section.

Example: #include
#include
define PI 3.14

3>Main program:

This section should ideally contain the classes (in case of C++), global variables, filestreams, and the body of the main( ) function, containing all the sub-functions, local variables, objects etc..

4>Functions and extra comments:

This section should contain the extra information and notes about debugging if required (untill the code is perfected), other functions that are to be used in the main program (for defining the functions here, they have to be declared above their control entry; i.e., generally above main).

Example:
/*program to add two numbers. created by:xuxux created on: 30/6/05 */

IDEOLOGY OF A PROGRAMMING LANGUAGE

2005-08-25T09:15:00.000-07:00

There are three important aspects of any programming language(and maybe just any language for that matter).

First: What are the ways in which it stores data?

This involves concepts of size of data units, compatibilty of stored data on various platforms where it is to be used and the storage media used (usually the main memory for most direct data storing functions and calculations), format of the stored data and most importantly the indexing or pre-management of data to be stored.

Second: How it accomplishes input and output.

This relates to the methods of storing data and the usage of the various input and output streams to accomplish the task, feasiblity and polarity of using the input and output streams as well as the time taken to store/retrive and rewrite the data.

Third: How and what operators it uses to transform and combine data.

This covers the likes of methods of data standardization so that data can be compared and transformed from one main standard form to subsidiaries, operations to be performed on the data, the operation specifies etc.

These are the Three kinds of building blocks of the basic ideology behind a programming language. I will cover each of the three aspects in detail in the future posts.

INTRODUCTION TO C

2005-08-22T10:48:00.000-07:00

Before we venture into actual C++, let me talk a little about 'C' the ancestor of C++. It would be of interest to find out how C came into existance.

What is C?

C is a programming language which was developed at AT & T's Bell Laboratories of USA in 1972. It was designed and written by Dennis Ritchie.

In the late Seventies C began to replace the more familiar languages of that time like PL/I, ALGOL etc. No one Pushed C. Although C wasn't made the 'official' Bell Labs Language, without any advertisement C's reputation spread and its pool of users grew. Ritchie seems to have been rather surprised that so many programmers preffered C to older languages like FORTRAN or PL/I, or the newer ones Pascal and APL. But that's waht certainly happened!

That is how we got C++ in our hands.

INTRODUCTION TO C++

2005-08-17T05:12:00.000-07:00

The C++ language was developed at AT&T Bell Laboratories in the early 1980s by Bjarne Stroustrup. The name C++ was coined by: Rick Mascitti, where '++' is the increment operator.

The maturation of the C++ language is attested to by two recent events:

1] the foramation of an ANSI (American National Standard Institute) C++ committee and
2] the publication of the Annotated C++ Reference Manual by Ellis and Stroustrup.

The major reason for the success and popularity of C++ is that it supports the Object Oriented technology, the latest in the software development and the most near to the real world.

THE C++ BLOG

2005-08-15T05:17:00.000-07:00

HI I AM STARTING THIS NEW C++ PROGRAMMING BLOG TODAY.