Be more strict about escape sequences that rename windows or set titles:

ignore any that not valid UTF-8 outright, and for good measure pass the
result through our UTF-8-aware vis(3).

1 files changed, 25 insertions, 0 deletions

diff --git a/utf8.c b/utf8.c
index 3a315749..099541fe 100644
--- a/utf8.c
+++ b/utf8.c
@@ -207,6 +207,31 @@ utf8_stravis(char **dst, const char *src, int flag)
 	return (len);
 }
 
+/* Does this string contain anything that isn't valid UTF-8? */
+int
+utf8_isvalid(const char *s)
+{
+	struct utf8_data	 ud;
+	const char		*end;
+	enum utf8_state		 more;
+	size_t			 i;
+
+	end = s + strlen(s);
+	while (s < end) {
+		if ((more = utf8_open(&ud, *s)) == UTF8_MORE) {
+			while (++s < end && more == UTF8_MORE)
+				more = utf8_append(&ud, *s);
+			if (more == UTF8_DONE)
+				continue;
+			return (0);
+		}
+		if (*s < 0x20 || *s > 0x7e)
+			return (0);
+		s++;
+	}
+	return (1);
+}
+
 /*
  * Sanitize a string, changing any UTF-8 characters to '_'. Caller should free
  * the returned string. Anything not valid printable ASCII or UTF-8 is