// Doing this in a separate pass (rather than slicing the string s
	// and collecting the result substrings right away) is significantly
	// more efficient, possibly due to cache effects.
	start := -1 // valid span start if >= 0
	for i := 0; i < len(s); {
		size := 1
		r := rune(s[i])
		if r >= utf8.RuneSelf {
			r, size = utf8.DecodeRune(s[i:])
		}
		if f(r) {
			if start >= 0 {