Simple Code Tasks Should Be Simple

I frequently see people do simple things in the most complicated ways with dynamic languages. I suspect much of this is a carry-over from how we teach algorithms and programming in universities. If you want your code to be readable, and you want it to be maintainable long-term, then simple code tasks should be simple.

Merging Two Lists

This is simple. You've got a list of fruits, and a list of vegetables, and you want to merge them into a list called "produce."

In Perl:

my @fruits = qw/apple banana mango/;
my @veggies = qw/broccoli asparagus spinach/;

In JavaScript:

const fruits = ['apple', 'banana', 'mango']
const veggies = ['broccoli', 'asparagus', 'spinach']

Some folks will want to use iteration, or even a push() function of some sort here, but neither is necessary. A simple assignment statement will work just fine.

In Perl:

my @produce = (@fruits, @veggies);

In JavaScript:

const produce = [...fruits, ...veggies];

Not very impressive, I know, but watch what happens when I do the same thing with associative arrays (a.k.a. Objects, a.k.a. hashes). Now we're going to have produce items, with their colors.

In Perl:

my %fruits = (
  apple  => 'red',
  banana => 'yellow',
  mango  => 'light-orange');
my %veggies = (
  broccoli  => 'green',
  asparagus => 'green',
  spinach   => 'green');

my %produce = (%fruits, %veggies);

In JavaScript:

const fruits = {
  apple:  'red',
  banana: 'yellow',
  mango:  'light-orange'}
const veggies = {
  broccoli:  'green',
  asparagus: 'green',
  spinach:   'green'}

const produce = {...fruits, ...veggies};

It's super cool to have slick code that does neat things, but when it comes to squishing data together keeping things simple is always better.

One Exception: When you're using JavaScript, the spread operator (...) is limited to the maximum limit supported by Function.apply(), which (as of the time of this post) is 65,536 total values.

Anyway, I had fun writing this and I hope that your code brings you joy.

Comments

[Matthew O. Persico]

This is why you will hear experienced Perlers (Perlizens?) insist that an array is not a list. If you understand that an array is a list interpreted in numeric indexed order and a hash is a list interpreted in pairwise key value fashion, then you are more likely to perform "list" operations when beneficial. I think the Perl literature needs to do more than it already does to make a "list" a real, live actual data structure, distinct from the array and hash "storage" of lists.

[Mike Stemle]

What does this semantic difference mean for those writing code?

[Matthew O. Persico]

It means they realize they can do list-y type things and not resort to loops, i.e. the answer to your followup question is your article. :-)

[Matthew O. Persico]

Or, more precisely, if the semantic difference was more distinctly taught, you wouldn't have had to write your article.

[Mike Stemle]

Yeah, most dynamic languages seem to have this same difference. I think that for beginners it is semantics, but for those who are starting to come into their own in their coding practice it's a super important concept to understand.

Thanks for this contribution!

[Aleksashko]

Insightful read. I could remember having a headache reading a senior developers code, it took me about a week to fully grasp what code does what. I will definitely keep this mind in my future projects. Nice write up!