tar/common: get index of first invalid utf-8 char

tar/common/utf8.go

@@ -2,20 +2,21 @@ package common
 
 // IsValidUtf8String checks for in valid UTF-8 characters
 func IsValidUtf8String(s string) bool {
-	for _, r := range s {
-		if int(r) == 0xfffd {
-			return false
-		}
-	}
-	return true
+	return InvalidUtf8Index([]byte(s)) == -1
 }
 
 // IsValidUtf8Btyes checks for in valid UTF-8 characters
 func IsValidUtf8Btyes(b []byte) bool {
-	for _, r := range string(b) {
+	return InvalidUtf8Index(b) == -1
+}
+
+// InvalidUtf8Index returns the offset of the first invalid UTF-8 character.
+// Default is to return -1 for a wholly valid sequence.
+func InvalidUtf8Index(b []byte) int {
+	for i, r := range string(b) {
 		if int(r) == 0xfffd {
-			return false
+			return i
 		}
 	}
-	return true
+	return -1
 }

tar/common/utf8_test.go

@@ -6,27 +6,36 @@ func TestStringValidation(t *testing.T) {
 	cases := []struct {
 		value  string
 		result bool
+		offset int
 	}{
-		{"aä\uFFFD本☺", false},
-		{"aä本☺", true},
+		{"aä\uFFFD本☺", false, 3},
+		{"aä本☺", true, -1},
 	}
 
 	for _, c := range cases {
+		if i := InvalidUtf8Index([]byte(c.value)); i != c.offset {
+			t.Errorf("string %q - offset expected %d, got %d", c.value, c.offset, i)
+		}
 		if got := IsValidUtf8String(c.value); got != c.result {
 			t.Errorf("string %q - expected %v, got %v", c.value, c.result, got)
 		}
 	}
 }
+
 func TestBytesValidation(t *testing.T) {
 	cases := []struct {
 		value  []byte
 		result bool
+		offset int
 	}{
-		{[]byte{0xE4}, false},
-		{[]byte("aä本☺"), true},
+		{[]byte{0xE4}, false, 0},
+		{[]byte("aä本☺"), true, -1},
 	}
 
 	for _, c := range cases {
+		if i := InvalidUtf8Index(c.value); i != c.offset {
+			t.Errorf("bytes %q - offset expected %d, got %d", c.value, c.offset, i)
+		}
 		if got := IsValidUtf8Btyes(c.value); got != c.result {
 			t.Errorf("bytes %q - expected %v, got %v", c.value, c.result, got)
 		}