py/lexer: Raise SyntaxError when str hex escape sequence is malformed.

Addresses issue #1390.

2 files changed, 20 insertions, 2 deletions

diff --git a/py/lexer.c b/py/lexer.c
index 3e7d589f61..17d711696b 100644
--- a/py/lexer.c
+++ b/py/lexer.c
@@ -454,8 +454,8 @@ STATIC void mp_lexer_next_token_into(mp_lexer_t *lex, bool first_token) {
                             {
                                 mp_uint_t num = 0;
                                 if (!get_hex(lex, (c == 'x' ? 2 : c == 'u' ? 4 : 8), &num)) {
-                                    // TODO error message
-                                    assert(0);
+                                    // not enough hex chars for escape sequence
+                                    lex->tok_kind = MP_TOKEN_INVALID;
                                 }
                                 c = num;
                                 break;
diff --git a/tests/basics/lexer.py b/tests/basics/lexer.py
index 1057c3961b..70a9cce993 100644
--- a/tests/basics/lexer.py
+++ b/tests/basics/lexer.py
@@ -38,3 +38,21 @@ def a(x):
                if x:
                 print(x)
 a(1)
+
+# badly formed hex escape sequences
+try:
+    exec(r"'\x0'")
+except SyntaxError:
+    print("SyntaxError")
+try:
+    exec(r"b'\x0'")
+except SyntaxError:
+    print("SyntaxError")
+try:
+    exec(r"'\u000'")
+except SyntaxError:
+    print("SyntaxError")
+try:
+    exec(r"'\U0000000'")
+except SyntaxError:
+    print("SyntaxError")