From 7c85c7c210e3ad417f59038de95b71618783d76c Mon Sep 17 00:00:00 2001
From: Damien George <damien.p.george@gmail.com>
Date: Mon, 26 Nov 2018 16:13:08 +1100
Subject: [PATCH] py/unicode: Fix check for valid utf8 being stricter about
 contn chars.

---
 py/unicode.c             | 2 +-
 tests/unicode/unicode.py | 4 ++++
 2 files changed, 5 insertions(+), 1 deletion(-)

diff --git a/py/unicode.c b/py/unicode.c
index 935dc9012..d69b6f56f 100644
--- a/py/unicode.c
+++ b/py/unicode.c
@@ -180,7 +180,7 @@ bool utf8_check(const byte *p, size_t len) {
     for (; p < end; p++) {
         byte c = *p;
         if (need) {
-            if (c >= 0x80) {
+            if (UTF8_IS_CONT(c)) {
                 need--;
             } else {
                 // mismatch
diff --git a/tests/unicode/unicode.py b/tests/unicode/unicode.py
index 3a35ce894..b3d4b09ee 100644
--- a/tests/unicode/unicode.py
+++ b/tests/unicode/unicode.py
@@ -47,3 +47,7 @@ try:
     str(bytearray(b'ab\xc0a'), 'utf8')
 except UnicodeError:
     print('UnicodeError')
+try:
+    str(b'\xf0\xe0\xed\xe8', 'utf8')
+except UnicodeError:
+    print('UnicodeError')