py/unicode: Fix check for valid utf8 being stricter about contn chars.

dpgeorge · dpgeorge · commit 7c85c7c210e3 · 2018-11-26T16:13:08.000+11:00
diff --git 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
@@ -47,3 +47,7 @@
     str(bytearray(b'ab\xc0a'), 'utf8')
 except UnicodeError:
     print('UnicodeError')
+try:
+    str(b'\xf0\xe0\xed\xe8', 'utf8')
+except UnicodeError:
+    print('UnicodeError')
