RogerBW's Blog

I’ve been doing the Weekly Challenges. The latest involved combining lists and squashing unicode characters. (Note that this is open until 16 October 2022.)

Task 1: Zip List

You are given two list @a and @b of same size.

Create a subroutine sub zip(@a, @b) that merge the two list as shown in the example below.

(In other words, f([1, 3, 5], [2, 4, 6]) == [1, 2, 3, 4, 5, 6].)

I only met this relatively recently; it's not really something I've tended to want to do in my programming life. Some of the languages I'm using have a built-in function to do this; others don't. As an example of one that doesn't, Lua:

function zip(a, b)
   local out = {}
   for i = 1,#a do
      table.insert(out,a[i])
      table.insert(out,b[i])
   end
   return out
end

and similarly with JavaScript and PostScript. Kotlin has a zip but you can't trivially flatten its List of Pairs into a single List, so I end up doing it the hard way.

(This is made simpler because we're told the two input lists are the same length; otherwise we'd have to decide what to do in case of a length mismatch, whether to stop early or put in something like an undef.)

For the other languages, I tried to use the built-function when it was available. Perl doesn't have a built-in, but List::MoreUtils has mesh which I just wrap for simplicity.

sub zip($a, $b) {
  return [mesh(@{$a}, @{$b})];
}

Rust gets a bit fiddly, and there's probably a better way to do it with flatten()…

fn ziplist(a: Vec<&str>, b: Vec<&str>) -> Vec<String> {
    a.iter()
        .zip(b.iter())
        .map(|x| vec![x.0.to_string(), x.1.to_string()])
        .collect::<Vec<Vec<String>>>()
        .concat()
}

Python needs itertools for chain.

def ziplist(a, b):
  return list(chain(*zip(a,b)))

Ruby is, as often, the most straightforward:

def ziplist(a,b)
  return a.zip(b).flatten
end

Raku:

sub ziplist(@a, @b) {
  return Array(zip(@a,@b)[*;*]);
}

Task 2: Unicode Makeover

You are given a string with possible unicode characters.

Create a subroutine sub makeover($str) that replace the unicode characters with ascii equivalent. For this task, let us assume it only contains alphabets.

This is a hard thing to do right. So my answer, as with date software, is not to do it myself wherever possible. The canonical answer for years has been UniDecode (for Perl, Text::UniDecode); now there's also anyascii, that has more glyph coverage. Both of these are implemented for multiple languages and leave me not writing any code at all.

Perl:

use utf8;
use Text::Unidecode;

is(unidecode("ÃÊÍÒÙ"), "AEIOU", 'example 1');

is(unidecode("âÊíÒÙ"), "aEiOU", 'example 2');

Rust:

use any_ascii::any_ascii;

#[test]
fn test_ex1() {
    assert_eq!(any_ascii("ÃÊÍÒÙ"), "AEIOU");
}

#[test]
fn test_ex2() {
    assert_eq!(any_ascii("âÊíÒÙ"), "aEiOU");
}

Full code on github.