Replace the stray bytes of text that is not UTF-8

A commit whose encoding header names something iconv cannot convert,
or whose text is labelled UTF-8 without being it, reached a page
declared UTF-8 with its bytes untouched. The atom feed already
replaced them, so its UTF-8 reader moves to shared.c and every ident,
subject, message and tag text passes through it.

1 file changed, 18 insertions, 9 deletions

diff --git a/source/parsing.c b/source/parsing.c
index 655973b..8aa0ee4 100644
--- a/source/parsing.c
+++ b/source/parsing.c
@@ -92,8 +92,11 @@ static int end_of_header(const char *p)
 }
 
 /*
- * A git built without iconv still offers reencode_string, where it always
- * fails, so the field is left in whatever encoding it arrived in.
+ * Converts a field into the page encoding, which is UTF-8. A git built
+ * without iconv still offers reencode_string, where it always fails, and a
+ * field can be labelled UTF-8 without being it, so whatever is left that is
+ * not UTF-8 has its stray bytes replaced, and a page declared UTF-8 never
+ * carries any.
  */
 static const char *reencode(char **text, const char *from, const char *to)
 {
@@ -102,14 +105,14 @@ static const char *reencode(char **text, const char *from, const char *to)
 	if (!*text || !from || !to)
 		return *text;
 
-	if (!strcasecmp(from, to))
-		return *text;
-
-	converted = reencode_string(*text, to, from);
-	if (converted) {
-		free(*text);
-		*text = converted;
+	if (strcasecmp(from, to)) {
+		converted = reencode_string(*text, to, from);
+		if (converted) {
+			free(*text);
+			*text = converted;
+		}
 	}
+	cgit_utf8_sanitize(*text);
 	return *text;
 }
 
@@ -256,6 +259,12 @@ struct taginfo *cgit_parse_tag(struct tag *tag)
 	if (p && *p)
 		info->msg = xstrdup(p);
 
+	// A tag carries no encoding header, so its text is taken as UTF-8 and
+	// cleaned the way a commit's is.
+	cgit_utf8_sanitize(info->tagger);
+	cgit_utf8_sanitize(info->tagger_email);
+	cgit_utf8_sanitize(info->msg);
+
 cleanup:
 	free(data);
 	return info;