ghostutils: Treat 0x80 (and above) as non-ASCII

Any ASCII character above 0x7F should be treated as UTF-8 in
ghostutils functions because GLib expects host names to be either
punycode encoded or in valid UTF-8 format.

The checks in gutf8.c already treat 0x80 as non-ASCII, but two checks
in ghostutils.c erroneously check for "great than" not "greater than or
equal to".

Clarify this by adding a new macro which is reused by PUNICODE_IS_BASIC
for better documentation in code.

glib/ghostutils.c

@@ -56,7 +56,8 @@
 #define PUNYCODE_INITIAL_BIAS  72
 #define PUNYCODE_INITIAL_N   0x80
 
-#define PUNYCODE_IS_BASIC(cp) ((guint)(cp) < 0x80)
+#define IS_ASCII(cp) ((guint) (cp) < 0x80)
+#define PUNYCODE_IS_BASIC(cp) IS_ASCII (cp)
 
 /* Encode/decode a single base-36 digit */
 static inline gchar
@@ -258,8 +259,8 @@ contains_non_ascii (const gchar *str,
 
   for (p = str; len == -1 ? *p : p < str + len; p++)
     {
-      if ((guchar)*p > 0x80)
+      if (!IS_ASCII (*p))
-	return TRUE;
+        return TRUE;
     }
   return FALSE;
 }
@@ -505,9 +506,9 @@ g_hostname_to_ascii (const gchar *hostname)
       unicode = FALSE;
       for (p = label; *p && !idna_is_dot (p); p++)
 	{
-	  if ((guchar)*p > 0x80)
+          if (!IS_ASCII (*p))
-	    unicode = TRUE;
+            unicode = TRUE;
-	}
+        }
 
       oldlen = out->len;
       llen = p - label;

glib/tests/hostutils.c

@@ -72,6 +72,7 @@ static const gint num_non_round_trip_names = G_N_ELEMENTS (non_round_trip_names)
 static const gchar *bad_names[] = {
   "disallowed\xef\xbf\xbd" "character",
   "non-utf\x88",
+  "smallest-non-utf-char\x80",
   "xn--mixed-\xc3\xbcp",
   "verylongverylongverylongverylongverylongverylongverylongverylongverylong"
   "verylongverylongverylongverylongverylongverylongverylongverylongverylong"
@@ -331,6 +332,30 @@ test_is_ip_addr (void)
     }
 }
 
+static const gchar *ascii_names[] = {
+  "ascii-\x7F",
+};
+
+static const gchar *non_ascii_names[] = {
+  "disallowed\x80" "character",
+};
+
+static void
+test_hostname_is_non_ascii (void)
+{
+  for (size_t i = 0; i < G_N_ELEMENTS (ascii_names); i++)
+    g_assert_false (g_hostname_is_non_ascii (ascii_names[i]));
+
+  for (size_t i = 0; i < G_N_ELEMENTS (idn_test_domains); i++)
+    g_assert_false (g_hostname_is_non_ascii (idn_test_domains[i].ascii_name));
+
+  for (size_t i = 0; i < G_N_ELEMENTS (non_ascii_names); i++)
+    g_assert_true (g_hostname_is_non_ascii (non_ascii_names[i]));
+
+  for (size_t i = 0; i < G_N_ELEMENTS (idn_test_domains); i++)
+    g_assert_true (g_hostname_is_non_ascii (idn_test_domains[i].unicode_name));
+}
+
 /* FIXME: test names with both unicode and ACE-encoded labels */
 /* FIXME: test invalid unicode names */
 
@@ -362,6 +387,7 @@ main (int   argc,
       return 0;
     }
 
+  g_test_add_func ("/hostutils/hostname_is_non_ascii", test_hostname_is_non_ascii);
   g_test_add_func ("/hostutils/to_ascii", test_to_ascii);
   g_test_add_func ("/hostutils/to_unicode", test_to_unicode);
   g_test_add_func ("/hostutils/is_ip_addr", test_is_ip_addr);