Merge 7568b3b105 into 7b050b366b

py/objint.c

@@ -55,7 +55,7 @@ static mp_obj_t mp_obj_int_make_new(const mp_obj_type_t *type_in, size_t n_args,
                 return o;
             } else if (mp_get_buffer(args[0], &bufinfo, MP_BUFFER_READ)) {
                 // a textual representation, parse it
-                return mp_parse_num_integer(bufinfo.buf, bufinfo.len, 0, NULL);
+                return mp_parse_num_integer(bufinfo.buf, bufinfo.len, 10, NULL);
             #if MICROPY_PY_BUILTINS_FLOAT
             } else if (mp_obj_is_float(args[0])) {
                 return mp_obj_new_int_from_float(mp_obj_float_get(args[0]));

py/parsenum.c

@@ -78,6 +78,9 @@ mp_obj_t mp_parse_num_integer(const char *restrict str_, size_t len, int base, m
     // string should be an integer number
     mp_int_t int_val = 0;
     const byte *restrict str_val_start = str;
+    if (base == 0) {
+        goto value_error;
+    }
     for (; str < top; str++) {
         // get next digit as a value
         mp_uint_t dig = *str;

py/parsenumbase.c

@@ -27,9 +27,20 @@
 #include "py/mpconfig.h"
 #include "py/misc.h"
 #include "py/parsenumbase.h"
+#include "py/runtime.h"
+
+static bool is_all_zeros(const char *str, size_t len) {
+    while (len-- && unichar_isdigit(*str)) {
+        if (*str++ != '0') {
+            return false;
+        }
+    }
+    return true;
+}
 
 // find real radix base, and strip preceding '0x', '0o' and '0b'
 // puts base in *base, and returns number of bytes to skip the prefix
+// puts 0 in *base to indicate an invalid C-style octal literal
 size_t mp_parse_num_base(const char *str, size_t len, int *base) {
     const byte *p = (const byte *)str;
     if (len <= 1) {
@@ -49,6 +60,9 @@ size_t mp_parse_num_base(const char *str, size_t len, int *base) {
                 *base = 10;
             }
             p -= 2;
+            if (!is_all_zeros(str, len)) {
+                *base = 0;
+            }
         }
     } else if (*base == 8 && c == '0') {
         c = *(p++);

tests/basics/int1.py

@@ -13,6 +13,7 @@ print(int('1'))
 print(int('+1'))
 print(int('-1'))
 print(int('01'))
+print(int('00'))
 print(int('9'))
 print(int('10'))
 print(int('+10'))
@@ -31,6 +32,7 @@ print(int(' -3 '))
 print(int('0', 10))
 print(int('1', 10))
 print(int(' \t 1 \t ', 10))
+print(int(' \t 00 \t ', 10))
 print(int('11', 10))
 print(int('11', 16))
 print(int('11', 8))
@@ -79,6 +81,7 @@ test('0o8', 8)
 test('0xg', 16)
 test('1 1', 16)
 test('123', 37)
+test('01', 0)
 
 # check that we don't parse this as a floating point number
 print(0x1e+1)

tests/basics/lexer.py

@@ -83,3 +83,11 @@ try:
     exec(r"'\U0000000'")
 except SyntaxError:
     print("SyntaxError")
+
+# Properly formed integer literals
+print(eval("00"))
+# badly formed integer literals
+try:
+    eval("01")
+except SyntaxError:
+    print("SyntaxError")

tests/basics/nanbox_smallint.py

@@ -23,17 +23,17 @@ if float("1e100") == float("inf"):
     raise SystemExit
 
 micropython.heap_lock()
-print(int("0x80000000"))
+print(int("0x80000000", 16))
 micropython.heap_unlock()
 
 # This is the most positive small integer.
 micropython.heap_lock()
-print(int("0x3fffffffffff"))
+print(int("0x3fffffffffff", 16))
 micropython.heap_unlock()
 
 # This is the most negative small integer.
 micropython.heap_lock()
-print(int("-0x3fffffffffff") - 1)
+print(int("-0x3fffffffffff", 16) - 1)
 micropython.heap_unlock()
 
 x = 1