如何计算字符串中日文和拉丁文字符的数量？

Question

I need to write a function that properly follows Japanese counting of characters and therefore returns length of characters in a string given these conditions:

• 1 for Full-width char (Japanese kanji, katakana, and hiragana)
• 0.5 for Half-width char (0-9, AZ).

here is my unit test I wrote:

 describe('#getCaptionLength', () => { it('should return correct caption length for japanase', () => { const text = 'を取り外すコネクタと考えてください'; const result = getCaptionLength(text); expect(result).toBe(17) }); it('should return correct caption length for japanase mixed with latin', () => { const text = 'を取り外すコネクタと考えてください hello world'; const result = getCaptionLength(text); expect(result).toBe(17 + 6); }); });

Can you please help me write this function that would pass my unit test? Thanks!

Answer 1

It's not very complicated if you want to just pass these two tests you could write something like this:

 function getCaptionLength(text) { // find all latin characters with RexExp let re = new RegExp('[A-Za-z0-9 ]+', 'g'); let found = text.match(re); // get length of latin part let latinLength = found? found.join('').length: 0; // japanese part is just full string - latin part let japaneseCharactersLength = text.length - latinLength; // calculate and return the final result return japaneseCharactersLength + latinLength * 0.5; }

But this of course would also count as 1 everytihng that is not either Japanese character OR latin character such as emoji, special characters and what not.