There are a few fundamental things we have invented in programming languages,
which was invented at various point of time, particularly by one language and they were later adapted by many other languages.
I am going to list some of them, which have bigger impacts.
DataType The first and foremost thing which we have ever invented is the data type. Computers were just a dumb machine, which could only understand the binary sequence. Binary sequence made no sense till it was grouped to form a DataType.
As you know a DataType, is something which groups the binary sequence together and represents some entity in mathematical word OR real word. All depends on its interpretation.
DataType is not just a grouping of binary sequence but also a set of operations which it posses. I mean a DataType definition just don't end with its binary grouping but also the operation which can be performed on those entities.
Later we have evolved these DataTypes into more complex form. We used mathematics to bring various complex data structures like List, Stack, Queue etc.
Today almost every programming language directly or indirectly have concept of data type.
Pointers With the invention of computers, we've also invented the concept of loading and storing data from Hardware. Computers accesses memory by toggling special bit patterns in the wires (which was called data bus). This led to the invention of Pointers in our high level programming languages.
Pointers specially became popular in C programming language. Pointer is one of the most popular programming concepts ever invented. Other than C language, pointers are supported by languages like C++,C#, Fortran, Pascal. Few dialects of
BASIC also supports pointer.
Structured Programming Most of the programs in the early days were completely relying on GOTO statements, which was a real mess and was making programmers life real hell. Now that's where we have invented the "structured programming " another fundamental programming concept. Through this concept we have brought something called functions, and subsequently we realized the power of abstraction.
OOPs I think OOPs (Object Oriented Programming) is one of the most popular and ever lasting fundamental concept we have invented in history of programming languages. OOps is an umbrella concept. We have brought many concepts under this.Concepts like Data hiding, Inheritance, abstraction, polymorphisms (static & dynamic) were just the beginning of OOps. OOps is available directly or indirectly in almost all modern languages. Languages like C++, java & c# have brought it a long way.
Regular Expression Regular expressions provide a concise and flexible means for identifying patterns in string. This is used for searching and replacing special pattern in a string.If your favourite programming language is supporting Regular Expressions, and you are still thinking of learning,then this is the time to go and learn. Regular Expressions are now supported by many languages (almost all popular programming languages). Additionally Regular expression became standard language for many find and replace system utilities. Unix command (utility) Grep is one of the most popularly known Regular Expression based utility. Regular Expressions became so popular concept that many programming languages made it as the part of there language syntax (construct).Languages like Perl, Ruby and TCL embraced regular expression as their primary language construct.
SQL With the invention of Relational database, where everything is stored in the form of Tables, SQL type of languages evolved. This was mainly developed for data query, data update, schema creation & schema modification. They became so popular that it was extended to make procedural SQL.
With the popularity of SQL, recently Google has brought
GQL to abstract their
Big Table ( a non relational data base). SQL like syntax is also borrowed by yahoo in
YQL, to query data from anywhere on internet.
LINQ in C#.net, is also inspired from SQL.
Managed Heap Managed Heap OR
Smart pointers, was another revolutionary concept which was invented as a hack of OOps concept (classes ) in C++. This was invented by Microsoft in a concept called COM. Smart Pointer, solved the problem of memory leak .
This concept was later adapted as default language semantic in programming languages like Java & C#. Later this was adapted by many programming languages like VB.net and Managed C++.
XPATH
XPath is another programming concept which was developed to access DOM tree, and became a preferred way to access the XML formatted data. This is another programming paradigm which you should be aware of, If by any chance you work with XML.
Duck Typing The Term “
Duck Typing” is invented by Python, though the concept of duck typing is old and was there in few languages earlier than Python.In duck typing, programmer is concerned with just those aspects of an object that are used, rather than with the type of the object itself.
Let's understand this. Let us say we have a real life object Shape, which knows how to draw itself (with a method draw). Now in OOps, We enforce this by creating an interface something like IDraw, any anything which can be drawn on the screen must be of type IDraw(i.e It should be inheriting IDraw). I Duck Typing, object can be drawn on the screen as long as object holds the draw method,irrespective of the type of object. DuckTyping removed the dependency of common interface definition, which are typically shared by client and server modules in OOps languages. Disadvantage of such thing is, programmer will not be able to know at the compile time, that the object is not having the draw method in it. But wait, python does not have compile time, its interpreted language, so all the problem can only be identified during the runtime.
One thing i am sure about DuckTyping is that it is very risky deal when you are building a big (
cathedral like) software. But it is very good when you are writing very small quick and dirty lines of code.
Duck Typing is supported by Python, JavaScript(and similar languages) & C# (for your surprise,
read this nice example).
Duck Typing helps a lot in JavaScript, in fact the concept of “JSON based AJAX” is completely based of duck typing.
Duck typing is a very controversial concept, many
OOPs lovers hate this concept.
Closure Some languages (e.g. JavaScript) allows you to define a function inside another function. Closure is the scope, which inner function is having. Coolest part of closure is, scope remains valid even after outer function have returned.
One of the nice example of closure is this (in JavaScript), Inner function
dofading will still be having access to ‘
Div_InClosureScope’ even after the
Fade have returned.
function Fade(id)
{
var Div_InClosureScope= document.getElementById(id);
var level=0;
function dofading()
{
var hex=level.toString(16);
Div_InClosureScope.style.backgroundColor='#ff'+hex+hex+hex+hex;
if(level<15)
{
level++;
setTimeout(dofading,100);
}
}
setTimeout(dofading,10);
}
In more general term, closure is a special scope, provided to a special instance of a function. In OPPs the member function enjoys closure (data members are in closure scope). Programming language C does not have closure concept at all. Its a most simple and straight language.
Yield I found this technique first in python.This was not something new, but can confuse most the programmers from c/c++ background. This technique somehow stores state of iterator (I will explain latter), and returns different result at different time, and this is something c/c++ programmers are not used to. Any C/C++ programmer assumes a function is a stateless machine, which can return only one result for given set of argument, no matter how many time you going call that function.
A Python function stack can be unwrapped( retuned) in 2 ways, by a return statement or by a yield statement. A return statement stops the execution of a function (same as in c/c++ ). On the other hand an yield statement halts the execution of a function and store the state, so that when it will be invoked later, execution will start from the same point.
Lets take an example of typical Fibonacci number generator.
#An endless generator
def fibonacci():
i = j = 1
while True:
r, i, j = i, j, i + j #respective assignment
yield r
for rabbits in fibonacci():
print rabbits,
if rabbits > 100: break
Output
1 1 2 3 5 8 13 21 34 55 89 144
Python certainly support, C++ like return statement, but using yield is most efficient for this such generators. All those recursive way of writing Fibonacci number generators are really inefficient( though they look simple).
Yield statement is
also supported by C#.net. Here is one of the
nice post about yield in c#. Do not miss!
I love this article, "
Life after loops". somehow, its connected to this blog post
