Need someone to help me understand these two solutions - map, reduce, match, regex

Description: I've been playing around with Codingames today, and came across this problem in a Clash of Code.

Problem: Given a word, output the sum of the word's index according to it's place in the ABC.

Example: word = "HI" result = 15

Explanation: H = 7 and I = 8, the sum of that is 15. This is because the H and I are at the 7th and 8th index in the alpahabet, if we determine that A = 0;

My solution: I've got it working with a nested loop, but it's not very efficient. Other solution 1:

print(readline().split("").reduce((a,b)=>b.match(/[A-Z]/)?a+b.charCodeAt(0)-65:a,0))

Other solution 2:

s=0
readline().split``.map(e=>/[A-Z]/.test(e)?s+=e.charCodeAt(0)-'A'.charCodeAt(0):0)
print(s)

Can someone explaing these codes to me? I've tried AI but I didn't get the hang of it. I also welcome suggestions of websites etc. that helps me understand these codes better.

Thank you!

Solution

Using normal split, reduce and match plus a ternary:

const res = "HeLLo"
  .split("")
  .reduce((a,b)=> 
    b.match(/[A-Z]/)? // does the current letter match an uppercase?
    a+b.charCodeAt(0)-65 // return accumulator plus current letter value minus A's letter value of 65
    :a  // otherwise return the accumulator
    ,0) // start with an accumulator of 0

console.log(res);

Using tagged template split, a map (no need for a map here) and a ternary

s=0;
"HeLLo"
  .split``  // split on anything
  .map(e=>/[A-Z]/.test(e)? // if from A-Z
  s+=e.charCodeAt(0)-'A'.charCodeAt(0) // add each charcode minus the charCode for A
  :0 // otherwise 0
  )
console.log(s);

Don't use a map when you need a forEach - and a ternary as above:

s=0;
const res = "HeLLo"
  .split``  // split on anything 
  .forEach(e=> s+= /[A-Z]/.test(e) ? e.charCodeAt(0)-65 : 0)
console.log(s);