Module Private.Utf8_text_chunks

of_utf8 utf8 ~width ~prefer_split_on_spaces splits utf8 into chunks no wider than width characters. chunks_of_utf8 always returns at least one chunk, which may be empty.

If prefer_split_on_spaces = true and such a space exists, utf8 will be split on the last U+0020 SPACE before the chunk becomes too wide. Otherwise, the split happens exactly at width characters.