1 files changed, 64 insertions, 0 deletions
diff --git a/util.c b/util.c
index 39fb0b8..9c64542 100644
--- a/util.c
+++ b/util.c
@@ -94,3 +94,67 @@ utf8_substring(
 	*out_bytes = sub_end - sub_start;
 	return sub_length;
 }
+
+static const bool utf8_allow_surrogates = false;
+
+void
+utf8_decode_char_nocheck(const char **sp, unsigned int *uc)
+{
+	const unsigned char *s = (const unsigned char *) *sp;
+	unsigned char c = *s++;
+	unsigned int len;
+	unsigned char sf[4] = {0xFF, 0x1F, 0xF, 0x7};
+
+	if (c < 0x80)
+		len = 0;
+	else if (c < 0xE0)
+		len = 1;
+	else if (c < 0xF0)
+		len = 2;
+	else
+		len = 3;
+
+	*uc = c & sf[len];
+	while (len--)
+	{
+		*uc <<= 6;
+		*uc |= *s++ & 0x3F;
+	}
+
+	*sp = (const char *) s;
+}
+
+bool
+utf8_validate(const char *str, size_t length)
+{
+	const unsigned char *s = (const unsigned char *) str;
+	const unsigned char *e = s + length;
+	int len;
+	
+	while (s < e)
+	{
+		if (*s <= 0x7F)
+		{
+			s++;
+			continue;
+		}
+		
+		len = pg_utf_mblen(s);
+		if (s + len > e)
+			return false;
+		
+		if (!pg_utf8_islegal(s, len))
+			return false;
+		
+		s += len;
+	}
+	
+	return true;
+}
+
+int
+utf8_encode_char(char *out, unsigned int uc)
+{
+	unicode_to_utf8(uc, (unsigned char *) out);
+	return pg_utf_mblen((unsigned char *) out);
+}