"Simple Pig Latin" codewars kata not passing tests

[shivajikobardan]

TL;DR Summary

codewars kata not passing tests

I'm doing this kata.
https://www.codewars.com/kata/520b9d2ad5c005041100000f/train/javascript

function pigIt(str) {
  const words = str.split(" ");
  for (var i = 0; i < words.length; i++) {
    if (words[i] === "!") {
      return "!";
    }

    var firstLetterSliced = words[i].slice(0, 1);
    var remainingLetterSliced = words[i].slice(1, words[i].length);
    return (remainingLetterSliced + firstLetterSliced + "ay");
  }
}

pigIt('Pig latin is cool'); // igPay atinlay siay oolcay

My code isn't passing a test. Please help fix the issue.

It works as expected so why is it not passing the tests?

 

[Filip Larsen]

You may want to dig into understanding what return mean.

 

[Mark44]

Your input string apparently was "This is my string". Your code produced "hisTay" and failed to process the other words in that string.

 

[Vanadium 50]

@shivajikobardan , a meta-comment. You are asking a LOT of questions, sometimes several in one day. There is no evidence that yolu are generalizing the answers you get to other pieces of code.

What's your goal? If it's to program on your own without PF help, I don't see you are making progress towards this goal. As I said, you need to generalize answers to one question to new situations.

 

[Filip Larsen]

By the way, this training exercise is tagged with regular expressions, so to solve it in the right spirit one should probably aim for using regular expressions. Combined with functional programming this can be solved with a "one-line" transformation pipeline.

 

[jedishrfu]

One other point is true pig-latin strips the consonants from the word so that "this" becomes "isthay" not "hisTay".

https://en.wikipedia.org/wiki/Pig_Latin

Any pig-roman will notice you are a foreigner, if you spoke it that way.

This may be why they suggest you use regular expressions for parsing the string.

 

[Vanadium 50]

Any pig-roman

Isn't that porcoroman?

 

[Mark44]

One other point is true pig-latin strips the consonants from the word so that "this" becomes "isthay" not "hisTay".

Based on the posted image of the test results -- "Expected 'hisThay' to equal 'hisTay siay ymay tringsay' -- the testing mechanism isn't expecting dipthongs such as 'th', 'sh', 'ch', etc. to be separated out.

The only problem I can see with the OP's solution is that his code stopped after the first word of the input string.

 

[Ibix]

Perhaps this is Wild Boar Latin, a different dialect.

But indeed, the OP might want to print out the values of i and reflect on what values he expects it to take and which ones it actually takes.

 

[jedishrfu]

Or wolf latin as spoken by the parents of Omulusray and emusRay.

 

[shivajikobardan]

@shivajikobardan , a meta-comment. You are asking a LOT of questions, sometimes several in one day. There is no evidence that yolu are generalizing the answers you get to other pieces of code.

What's your goal? If it's to program on your own without PF help, I don't see you are making progress towards this goal. As I said, you need to generalize answers to one question to new situations.

yeah I'm questioning that as well. I'm making no progress at all. Any suggestions?

 

[pbuk]

By the way, this training exercise is tagged with regular expressions, so to solve it in the right spirit one should probably aim for using regular expressions.

Absolutely, doing it any other way misses the point (and is much harder to boot).

Edit: there is a bug in the solution below: I have rewritten the solution below at https://www.physicsforums.com/threa...s-kata-not-passing-tests.1049368/post-6849397.

To demonstrate just how easy it is to solve, I will work through the solution to the harder problem posed by @jedishrfu:

One other point is true pig-latin strips the consonants from the word so that "this" becomes "isthay" not "hisTay".

• So we want to capture successive consonants: the easiest way to do this is to capture one or more succesive instances of anything that isn't a non-alphanumeric character or a vowel. \W matches a non-alphanumeric character so we have ([^\Waeiouy]+).
• We then want to capture the rest of the word; \w matches an alphanumeric ("word") character so we now have ([^aeiouy]+)(\w*).
• Add a g flag so the regexp acts on the whole string ("global", otherwise it will only look for the first match) and a u flag so it is case insensitive and we have /([^\Waeiouy]+)(\w*)/gu as the regular expression to search for.
• Now in a JavaScript string's replace method with a regular expression search pattern, we can use the captured parts of the string in the replacement: $1 will give us the first captured group etc. So '$2$1ay' will reverse the order of the groups and add 'ay', and because of the 'global' flag on the regular expression it will do this for each match.

Putting it all together we have:

function pigIt(str) {
  return str.replace(/([^\Waeiouy]+)(\w*)/gu, '$2$1ay');
}
pigIt('Pig latin is cool')
// 'igPay atinlay isay oolcay'
pigIt('This is my string')
// 'isThay isay ymay ingstray'
pigIt('Test that, even with punctuation, it works!\nReally!!!')
// 'estTay atthay, eenvay ithway unctuationpay, itay orksway!\neallyRay!!!'

 

[Mark44]

Shouldn't "even" be transformed to "veneay" rather than "eenvay"?

 

[pbuk]

Shouldn't "even" be transformed to "veneay" rather than "eenvay"?

Yeah, that's a bug (although I think it should actually just be "evenay").

 

[pbuk]

Reworked solution to the harder problem posed by @jedishrfu:

One other point is true pig-latin strips the consonants from the word so that "this" becomes "isthay" not "hisTay".

• So we want to capture successive consonants: the easiest way to do this in English is to capture anything that isn't a vowel: ([^aeiouy]*) (we will deal with uppercase characters later).
• We then want to capture the rest of the word; \w matches an alphanumeric ("word") character so we now have ([^aeiouy]*)(\w+). Note the + selector rather than *: this is to fix a bug in my first attempt where it added an extra "ay" at the end of the string.
• Add a g flag so the regexp acts on the whole string ("global", otherwise it will only look for the first match) and a u flag so it is case insensitive and we have /([^aeiouy]*)(\w+)(\W*)/gu as the regular expression to search for.
• Now in a JavaScript string's replace method with a regular expression search pattern, we can use the captured parts of the string in the replacement: $1 will give us the first captured group etc. So '$2$1ay$3' will reverse the order of the first two groups and add 'ay', followed by any punctuation or whitespace after the end of the word. Because of the 'global' flag on the regular expression it will do this for each match.

Putting it all together we have:

const pigIt = (str) => {
  return str.replace(/([^aeiouy]*)(\w+)(\W*)/gu, '$2$1ay$3');
};

console.log(pigIt('Pig latin is cool'));
// 'igPay atinlay isay oolcay'
console.log(pigIt('This is my string'));
// 'isThay isay ymay ingstray'
console.log(pigIt('Test that, even with punctuation, it works!\nReally!!!'));
// 'estTay atthay, evenay ithway unctuationpay, itay orksway!\neallyRay!!!'

 

[Vanadium 50]

Any suggestions?

You might limit yourself to one question per day. That will force you to think hard about the responses you are getting and what parts need clarification.

It may sound slower, but I think we can all agree "here's my code; debug it for me" isn't working.