first commit

text/clean.go

@@ -0,0 +1,109 @@
+package text
+
+import (
+	"strings"
+	"unicode/utf8"
+)
+
+func CleanMOTD(s string) string {
+	if s == "" {
+		return ""
+	}
+
+	s = strings.ReplaceAll(s, "\r\n", "\n")
+	s = strings.ReplaceAll(s, "\r", "\n")
+	s = stripColorCodes(s)
+	s = stripControl(s)
+	s = normalizeSpacesKeepNewlines(s)
+	return strings.TrimSpace(s)
+}
+
+func stripColorCodes(s string) string {
+	var b strings.Builder
+	b.Grow(len(s))
+
+	for i := 0; i < len(s); {
+		r, size := utf8.DecodeRuneInString(s[i:])
+		if r == utf8.RuneError && size == 1 {
+			i++
+			continue
+		}
+
+		if r == '§' || r == '&' {
+			if i+2 < len(s) && (s[i] == '§' || s[i] == '&') && (s[i+1] == 'x' || s[i+1] == 'X') {
+				j := i + 2
+				ok := true
+				for k := 0; k < 6; k++ {
+					if j+2 >= len(s) {
+						ok = false
+						break
+					}
+					if s[j] != s[i] {
+						ok = false
+						break
+					}
+					c := s[j+1]
+					if !isHex(c) {
+						ok = false
+						break
+					}
+					j += 2
+				}
+				if ok {
+					i = j
+					continue
+				}
+			}
+
+			i += size
+			if i < len(s) {
+				_, nextSize := utf8.DecodeRuneInString(s[i:])
+				i += nextSize
+			}
+			continue
+		}
+
+		b.WriteRune(r)
+		i += size
+	}
+
+	return b.String()
+}
+
+func isHex(c byte) bool {
+	return (c >= '0' && c <= '9') ||
+		(c >= 'a' && c <= 'f') ||
+		(c >= 'A' && c <= 'F')
+}
+
+func stripControl(s string) string {
+	var b strings.Builder
+	b.Grow(len(s))
+
+	for _, r := range s {
+		if r == '\n' || r == '\t' {
+			b.WriteRune(r)
+			continue
+		}
+		if r < 32 || r == 127 {
+			continue
+		}
+		b.WriteRune(r)
+	}
+	return b.String()
+}
+
+func normalizeSpacesKeepNewlines(s string) string {
+	lines := strings.Split(s, "\n")
+	for i := range lines {
+		lines[i] = strings.Join(strings.Fields(lines[i]), " ")
+	}
+
+	out := strings.Join(lines, "\n")
+	out = strings.Trim(out, "\n")
+
+	for strings.Contains(out, "\n\n\n") {
+		out = strings.ReplaceAll(out, "\n\n\n", "\n\n")
+	}
+	return out
+}