`number_of_words` doesn't support languages that don't use space separator.

number_of_words doesn't support languages that don't use space as separator.

English

    {{'The quick brown fox jumps over the lazy dog' | size}} //43
    {{'The quick brown fox jumps over the lazy dog' | number_of_words}} //9

Arabic I'm not sure if الكسول` counts as 1 or 6 word(s).

    {{'الثعلب البني السريع يقفز فوق الكلب الكسول' | size}} //41
    {{'الثعلب البني السريع يقفز فوق الكلب الكسول' | number_of_words}}  //7

Japanese

    {{'クイックブラウンキツネは怠惰な犬の上を飛ぶ' | size}} //21
    {{'クイックブラウンキツネは怠惰な犬の上を飛ぶ' | number_of_words}} //1

Korean. I'm not sure if 빠른 갈색 counts as 2 or 4 words.

    {{'빠른 갈색 여우는 게으른 개를 뛰어 넘습니다' | size}} //24
    {{'빠른 갈색 여우는 게으른 개를 뛰어 넘습니다' | number_of_words}} //7

Chinese.

    {{'敏捷的棕毛狐狸从懶狗身上跃过' | size}} //14
    {{'敏捷的棕毛狐狸从懶狗身上跃过' | number_of_words}} //1

Maybe worth looking at https://github.com/jpatokal/script_detector.

Start at the number_of_words filter and compare its current results with the English, Arabic, Japanese, Korean, and Chinese examples in the issue. Clarify with maintainers how words should be defined for languages without spaces, then determine the expected behavior and add coverage for the agreed cases.