Cut text between characters, never inside one

The index description was cut at a byte count, which could split a
multibyte character and leave an invalid sequence on the page. Commit
subjects already backed up to the start of the character, so html_ntxt
now does that for every caller.

1 file changed, 6 insertions, 0 deletions

diff --git a/source/html.c b/source/html.cindex 1a6b7c5..b6e62a6 100644--- a/source/html.c+++ b/source/html.c@@ -247,6 +247,12 @@ ssize_t html_ntxt(const char *txt, size_t len) 		} 		p++; 	}+	// A cut inside a multibyte character would leave an invalid sequence+	// on the page, so it backs up to where that character starts. The+	// escaped bytes are all ASCII, so no character spans one.+	if (left < 0)+		while (p > txt && (*p & 0xC0) == 0x80)+			p--; 	if (p != txt) 		html_raw(txt, p - txt); 	return left;
diff --git a/source/html.c b/source/html.c
index 1a6b7c5..b6e62a6 100644
--- a/source/html.c
+++ b/source/html.c
@@ -247,6 +247,12 @@ ssize_t html_ntxt(const char *txt, size_t len)
 		}
 		p++;
 	}
+	// A cut inside a multibyte character would leave an invalid sequence
+	// on the page, so it backs up to where that character starts. The
+	// escaped bytes are all ASCII, so no character spans one.
+	if (left < 0)
+		while (p > txt && (*p & 0xC0) == 0x80)
+			p--;
 	if (p != txt)
 		html_raw(txt, p - txt);
 	return left;