Pig Latin Translator

Which Pig Latin rules does the translator follow?

Words that begin with a vowel get "way" added (apple becomes appleway). Words that begin with consonants have the whole leading consonant group moved to the end, followed by "ay" (string becomes ingstray). "Qu" travels as a unit, so queen becomes eenquay. Y counts as a consonant, so yellow becomes ellowyay, and a word with no vowel at all just gets "ay" (rhythm becomes rhythmay). Other conventions exist, such as adding "yay" instead of "way", but this tool uses the way/ay pair.

What happens to capital letters and punctuation?

A capitalised word keeps a capital on its new first letter (Hello becomes Ellohay), and a word typed entirely in capitals stays in capitals (WOW becomes OWWAY). Punctuation before or after a word stays where it was, so (yellow) becomes (ellowyay) and world! becomes orldway!. Punctuation inside a word, such as the apostrophe in don’t, is treated as part of the word: it becomes on’tday.

Are numbers and mixed tokens translated?

No. Any token that contains a digit is copied through untouched, so 5pm. stays 5pm. and 3 stays 3. Hyphenated words are handled part by part: well-known becomes ellway-ownknay. Accented Latin letters count as letters, so café becomes afécay.

Why does Pig Latin to English only claim to be approximate?

Because the original word boundary is lost. Water becomes aterway, but the reverse step reads anything ending in "way" as a word that began with a vowel, so aterway decodes to ater, and world (orldway) decodes to orld. Words whose result ends in "ay" do better: it moves the trailing consonant group back, choosing the longest common English opening (str, thr, sch, th and so on), which gets string, school, queen, thank and day right but can misjudge unusual words.

Does it keep line breaks and spacing?

Yes. The input is split on whitespace but the whitespace itself is preserved, so paragraphs, line breaks and double spaces come out where they went in. The word-by-word breakdown lists each word with letters in it, and the count in its heading skips tokens that are only numbers or symbols.

Can it translate other languages?

It is built around English spelling. Letters from any alphabet count as letters, but only a, e, i, o, u and their accented forms are treated as vowels, so words in a script without those letters are only given the "ay" ending. Results for non-English words follow the same rules and are rarely meaningful.

Is there a length limit, and is my text sent anywhere?

The code sets no length limit. The calculate function that does the translation contains no network request; it only transforms the string you typed.