[OPINION] - does language really matter if they all do the samething?

[Dietrich Epp]

On Jan 24, 2004, at 3:24 PM, Paul Prescod wrote:

> function random_sword_magic_power(quality):
>     return choose_random_assoc(
> 	selector = (quality, (poor, medium, good)),
> 	properties = {	(5, 0, 0): glows_in_the_dark(),
> 	              	(3, 3, 0): magically_silent(),
> 			(1, 5, 1): elemental_power(
> 			  choice([earth, water, air, fire])
> 			(0, 2, 4): magical_keen_edge(),
> 			(0, 0, 2): ????}
>
> I don't follow what you are doing with the last line so I didn't 
> duplicate it.

The last line recursively calls random_sword_magic_power() twice and 
concatenates the results.  It's something that I used a lot in this 
program, which I originally wrote in Python, but I was completely 
baffled trying to implement this part.  The reason you can't do it with 
a dict or list in Python is that the list is always evaluated, and 
recurses infinitely.  A dict isn't appropriate because there would be 
duplicate keys.  Think of it as three tables collapsed into one, the 
'poor', 'medium', and 'good' tables, where the relative probabilities 
for each table fall on a different column.

So....

(choose-random-assoc color (red green) ((2 1) foo) ((1 1) bar))

...has a 2/3 chance of producing 'foo' when color=red, but only 1/2 
when color=green.

My original point was that Python is not completely agnostic.  There's 
a method to programming in Python, and some patterns just don't fit 
that method.  I anticipated people asking for examples, and I expected 
that a couple of my examples would fall flat.  While Python is a 
fantastic general-purpose language, it will often be beaten by 
application-specific languages, but only in their domain.  I 
purposefully didn't use any examples of application-specific languages 
because people would tell me that I were not being fair.

I still have no idea how to express the function in Python without 
adding "crutches".