commit be8e99c7d4ac4d46069a358436a7036a3066d0a9
author Damien George <damien.p.george@gmail.com> Wed Nov 05 16:45:54 2014 +0000
committer Damien George <damien.p.george@gmail.com> Wed Nov 05 16:45:54 2014 +0000
tree fd9de43db5e599d0026dfcad7f28e77d4df5f51a
parent a65c03c6c07fa8b092f7cd8d02c81e9ef8cd4a50

py: Allow bytes object as argument to some str methods.

This turns failing assertions to type exceptions for things like
b"123".find(...).  We still don't support operations like this on bytes
objects (unlike CPython), but at least it no longer crashes.

py/objstr.c

diff --git a/py/objstr.c b/py/objstr.c
index d6cfa8b..5146f36 100644
--- a/py/objstr.c
+++ b/py/objstr.c
@@ -621,8 +621,12 @@
 STATIC mp_obj_t str_finder(mp_uint_t n_args, const mp_obj_t *args, mp_int_t direction, bool is_index) {
     const mp_obj_type_t *self_type = mp_obj_get_type(args[0]);
     assert(2 <= n_args && n_args <= 4);
-    assert(MP_OBJ_IS_STR(args[0]));
-    assert(MP_OBJ_IS_STR(args[1]));
+    assert(MP_OBJ_IS_STR_OR_BYTES(args[0]));
+
+    // check argument type
+    if (!MP_OBJ_IS_STR(args[1])) {
+        bad_implicit_conversion(args[1]);
+    }
 
     GET_STR_DATA_LEN(args[0], haystack, haystack_len);
     GET_STR_DATA_LEN(args[1], needle, needle_len);
@@ -828,7 +832,7 @@
 }
 
 mp_obj_t mp_obj_str_format(mp_uint_t n_args, const mp_obj_t *args) {
-    assert(MP_OBJ_IS_STR(args[0]));
+    assert(MP_OBJ_IS_STR_OR_BYTES(args[0]));
 
     GET_STR_DATA_LEN(args[0], str, len);
     int arg_i = 0;
@@ -1190,7 +1194,7 @@
 }
 
 STATIC mp_obj_t str_modulo_format(mp_obj_t pattern, mp_uint_t n_args, const mp_obj_t *args, mp_obj_t dict) {
-    assert(MP_OBJ_IS_STR(pattern));
+    assert(MP_OBJ_IS_STR_OR_BYTES(pattern));
 
     GET_STR_DATA_LEN(pattern, str, len);
     const byte *start_str = str;
@@ -1378,7 +1382,7 @@
 }
 
 STATIC mp_obj_t str_replace(mp_uint_t n_args, const mp_obj_t *args) {
-    assert(MP_OBJ_IS_STR(args[0]));
+    assert(MP_OBJ_IS_STR_OR_BYTES(args[0]));
 
     mp_int_t max_rep = -1;
     if (n_args == 4) {
@@ -1482,8 +1486,12 @@
 STATIC mp_obj_t str_count(mp_uint_t n_args, const mp_obj_t *args) {
     const mp_obj_type_t *self_type = mp_obj_get_type(args[0]);
     assert(2 <= n_args && n_args <= 4);
-    assert(MP_OBJ_IS_STR(args[0]));
-    assert(MP_OBJ_IS_STR(args[1]));
+    assert(MP_OBJ_IS_STR_OR_BYTES(args[0]));
+
+    // check argument type
+    if (!MP_OBJ_IS_STR(args[1])) {
+        bad_implicit_conversion(args[1]);
+    }
 
     GET_STR_DATA_LEN(args[0], haystack, haystack_len);
     GET_STR_DATA_LEN(args[1], needle, needle_len);
@@ -1872,7 +1880,7 @@
 
 mp_uint_t mp_obj_str_get_hash(mp_obj_t self_in) {
     // TODO: This has too big overhead for hash accessor
-    if (MP_OBJ_IS_STR(self_in) || MP_OBJ_IS_TYPE(self_in, &mp_type_bytes)) {
+    if (MP_OBJ_IS_STR_OR_BYTES(self_in)) {
         GET_STR_HASH(self_in, h);
         return h;
     } else {
@@ -1882,7 +1890,7 @@
 
 mp_uint_t mp_obj_str_get_len(mp_obj_t self_in) {
     // TODO This has a double check for the type, one in obj.c and one here
-    if (MP_OBJ_IS_STR(self_in) || MP_OBJ_IS_TYPE(self_in, &mp_type_bytes)) {
+    if (MP_OBJ_IS_STR_OR_BYTES(self_in)) {
         GET_STR_LEN(self_in, l);
         return l;
     } else {