Bump liberty, test UTF-8 sanitization

1 files changed, 14 insertions, 6 deletions

diff --git a/degesch.c b/degesch.c
index d96567f..9f03dcd 100644
--- a/degesch.c
+++ b/degesch.c
@@ -3035,6 +3035,9 @@ irc_to_utf8 (const char *text)
 {
 	if (!text)
 		return NULL;
+
+	// XXX: the validation may be unnecessarily harsh, could do with a lenient
+	//   first pass, then replace any errors with the replacement character
 	size_t len = strlen (text) + 1;
 	if (utf8_validate (text, len))
 		return xstrdup (text);
@@ -7940,7 +7943,7 @@ irc_sanitize_cut_off_utf8 (char **line)
 	const char *p = *line, *end = strchr (p, 0);
 	int32_t codepoint;
 	while ((codepoint = utf8_decode (&p, end - p)) >= 0
-		&& codepoint <= 0x10FFFF /* TODO: move this check into a function */)
+		&& utf8_validate_cp (codepoint))
 		;
 	if (codepoint != -2)
 		return;
@@ -13984,10 +13987,15 @@ test_wrapping (void)
 }
 
 static void
-test_utf8_prefix (void)
+test_utf8 (void)
 {
 	static const char *a[] = { "fřoo", "Fřooř", "fřOOŘ" };
 	hard_assert (utf8_common_prefix (a, N_ELEMENTS (a)) == 5);
+
+	char *cut_off = xstrdup ("ё\xD0");
+	irc_sanitize_cut_off_utf8 (&cut_off);
+	hard_assert (!strcmp (cut_off, "ё\xEF\xBF\xBD"));
+	free (cut_off);
 }
 
 int
@@ -13995,10 +14003,10 @@ main (int argc, char *argv[])
 {
 	struct test test;
 	test_init (&test, argc, argv);
-	test_add_simple (&test, "/config",      NULL, test_config);
-	test_add_simple (&test, "/aliases",     NULL, test_aliases);
-	test_add_simple (&test, "/wrapping",    NULL, test_wrapping);
-	test_add_simple (&test, "/utf8-prefix", NULL, test_utf8_prefix);
+	test_add_simple (&test, "/config",   NULL, test_config);
+	test_add_simple (&test, "/aliases",  NULL, test_aliases);
+	test_add_simple (&test, "/wrapping", NULL, test_wrapping);
+	test_add_simple (&test, "/utf8",     NULL, test_utf8);
 	return test_run (&test);
 }