## An OOP Pattern Discovered by Accident

Expand Messages
• After I implemented the Math::GrahamFunction code to calculate the Graham Function using Perl ( http://search.cpan.org/~shlomif/Math-GrahamFunction-0.01/ ), I
Message 1 of 2 , Jan 26, 2007
After I implemented the Math::GrahamFunction code to calculate the Graham
Function using Perl (
http://search.cpan.org/~shlomif/Math-GrahamFunction-0.01/ ), I decided to
translate the code to Common Lisp in order to learn Lisp better. So I started
translating the methods of the classes, one by one.

Now, I have an SqFacts class that manages squaring factors, which for the
purpose of our discussion are an increasing series of natural numbers.
Furthermore, there's a Dipole class that inherits from it and manages two
squaring factors - a result and a composition. Now in my Dipole Perl code I
had the following declaration for the factors method:

<<<<<<<<<
sub factors
{
my \$self = shift;

return \$self->result()->factors();
}
>>>>>>>>>

And besides it similar declarations for some other functions:

<<<<<<<<<<
sub is_square
{
my \$self = shift;
return \$self->result()->is_square();
}

sub exists
{
my (\$self, \$factor) = @_;

return \$self->result()->exists(\$factor);
}

sub first
{
my \$self = shift;

return \$self->result()->first();
}
>>>>>>>>>>>

Now when I implemented it in my lisp code, I implemented the SqFacts class,
and tested it and then starting implementing the Dipole class. The factors()
method was one of the first I implemented, and then I wrote two tests for the
is_square predicate method, for both true and false returns. Surprisingly,
they both passed without even implementing is_sqaure.

As it turned out, is_square and friends for the SqFacts class, used the
factors() method and since factors() was already implemented and called the
factors of the "result" property, then in Dipole, they just did the same for
the factors of the "result". So I didn't have to implement the rest of the
methods.

Thinking about it, I realised one can consciously use this scheme by working
on an accessor and then over-riding this accessor in the derived class to
point to the accessor of the containing class.

So we can do:

<<<<<<<<<<<<
class Account
{
function account() { return self; }

# Other functions that operate on account.
}

class B inherit from Account
{
Account myaccount;

function account() { return myaccount; }

}
>>>>>>>>>>

Is this a known pattern in Object Oriented Programming and Design, and if so
how is it called?

Regards,

Shlomi Fish

---------------------------------------------------------------------
Shlomi Fish shlomif@...
Homepage: http://www.shlomifish.org/

Chuck Norris wrote a complete Perl 6 implementation in a day but then
destroyed all evidence with his bare hands, so no one will know his secrets.
• I did not completly understand what is the situation you are describing but when you seem to describe the Template Method Design Pattern . In this pattern you
Message 2 of 2 , Jan 26, 2007
I did not completly understand what is the situation you are describing but
when you seem to describe the "Template Method Design Pattern".
In this pattern you have the base algorithm, for example (in loose pseudo code
that looks like java):

abstract class SomeAlgorithm{

final public calcIT(){
this.calc1() + this.calc2()
}
abstract public calc1();
public calc2(){
return 0; //by default.
}

}

and the concrete classes who supply the actual implementation of parts
of the algorithm. At any time you can also override the calc2() which is
optional.

class concreteAlgorithm1 extends SomeAlgorithm{
public calc1(){
return 1;
}
}

class concreteAlgorithm2 extends SomeAlgorithm{
public calc1(){
return 2;
}
public calc2(){
return 3;
}
}

On Friday 26 January 2007 18:15, Shlomi Fish wrote:
> After I implemented the Math::GrahamFunction code to calculate the Graham
> Function using Perl (
> http://search.cpan.org/~shlomif/Math-GrahamFunction-0.01/ ), I decided to
> translate the code to Common Lisp in order to learn Lisp better. So I
> started translating the methods of the classes, one by one.
>
> Now, I have an SqFacts class that manages squaring factors, which for the
> purpose of our discussion are an increasing series of natural numbers.
> Furthermore, there's a Dipole class that inherits from it and manages two
> squaring factors - a result and a composition. Now in my Dipole Perl code I
> had the following declaration for the factors method:
>
> <<<<<<<<<
> sub factors
> {
> my \$self = shift;
>
> return \$self->result()->factors();
> }
>
>
> And besides it similar declarations for some other functions:
>
> <<<<<<<<<<
> sub is_square
> {
> my \$self = shift;
> return \$self->result()->is_square();
> }
>
> sub exists
> {
> my (\$self, \$factor) = @_;
>
> return \$self->result()->exists(\$factor);
> }
>
> sub first
> {
> my \$self = shift;
>
> return \$self->result()->first();
> }
>
>
> Now when I implemented it in my lisp code, I implemented the SqFacts class,
> and tested it and then starting implementing the Dipole class. The
> factors() method was one of the first I implemented, and then I wrote two
> tests for the is_square predicate method, for both true and false returns.
> Surprisingly, they both passed without even implementing is_sqaure.
>
> As it turned out, is_square and friends for the SqFacts class, used the
> factors() method and since factors() was already implemented and called the
> factors of the "result" property, then in Dipole, they just did the same
> for the factors of the "result". So I didn't have to implement the rest of
> the methods.
>
> Thinking about it, I realised one can consciously use this scheme by
> working on an accessor and then over-riding this accessor in the derived
> class to point to the accessor of the containing class.
>
> So we can do:
>
> <<<<<<<<<<<<
> class Account
> {
> function account() { return self; }
>
> # Other functions that operate on account.
> }
>
>
> class B inherit from Account
> {
> Account myaccount;
>
> function account() { return myaccount; }
>
> }
>
>
> Is this a known pattern in Object Oriented Programming and Design, and if
> so how is it called?
>
> Regards,
>
> Shlomi Fish
>
> ---------------------------------------------------------------------
> Shlomi Fish shlomif@...
> Homepage: http://www.shlomifish.org/
>
> Chuck Norris wrote a complete Perl 6 implementation in a day but then
> destroyed all evidence with his bare hands, so no one will know his
> secrets.

--
Regards,
Tzahi.
--