py/objint: Fix int.to_bytes() buffer size checks.

Fixes and improvements to `int.to_bytes()` are:
- No longer overflows if byte size is 0 (closes #13041).
- Raises OverflowError in any case where number won't fit into byte length
  (now matches CPython, previously MicroPython would return a truncated
  bytes object).
- Document that `micropython int.to_bytes()` doesn't implement the optional
  signed kwarg, but will behave as if `signed=True` when the integer is
  negative (this is the current behaviour).  Add tests for this also.

Requires changes for small ints, MPZ large ints, and "long long" large
ints.

Adds a new set of unit tests for ints between 32 and 64 bits to increase
coverage of "long long" large ints, which are otherwise untested.

Tested on unix port (64 bit small ints, MPZ long ints) and Zephyr STM32WB
board (32 bit small ints, long long large ints).

This work was funded through GitHub Sponsors.

Signed-off-by: Angus Gratton <angus@redyak.com.au>
This commit is contained in:
Angus Gratton
2023-11-29 11:23:16 +11:00
committed by Damien George
parent d933210d96
commit 908ab1ceca
12 changed files with 302 additions and 29 deletions

View File

@@ -421,29 +421,50 @@ static MP_DEFINE_CONST_FUN_OBJ_VAR_BETWEEN(int_from_bytes_fun_obj, 3, 4, int_fro
static MP_DEFINE_CONST_CLASSMETHOD_OBJ(int_from_bytes_obj, MP_ROM_PTR(&int_from_bytes_fun_obj));
static mp_obj_t int_to_bytes(size_t n_args, const mp_obj_t *args) {
// TODO: Support signed param (assumes signed=False)
// TODO: Support signed (currently behaves as if signed=(val < 0))
(void)n_args;
bool overflow;
mp_int_t len = mp_obj_get_int(args[1]);
if (len < 0) {
mp_int_t dlen = mp_obj_get_int(args[1]);
if (dlen < 0) {
mp_raise_ValueError(NULL);
}
bool big_endian = args[2] != MP_OBJ_NEW_QSTR(MP_QSTR_little);
vstr_t vstr;
vstr_init_len(&vstr, len);
vstr_init_len(&vstr, dlen);
byte *data = (byte *)vstr.buf;
memset(data, 0, len);
#if MICROPY_LONGINT_IMPL != MICROPY_LONGINT_IMPL_NONE
if (!mp_obj_is_small_int(args[0])) {
mp_obj_int_to_bytes_impl(args[0], big_endian, len, data);
overflow = !mp_obj_int_to_bytes_impl(args[0], big_endian, dlen, data);
} else
#endif
{
mp_int_t val = MP_OBJ_SMALL_INT_VALUE(args[0]);
size_t l = MIN((size_t)len, sizeof(val));
mp_binary_set_int(l, big_endian, data + (big_endian ? (len - l) : 0), val);
int slen = 0; // Number of bytes to represent val
// This logic has a twin in objint_longlong.c
if (val > 0) {
slen = (sizeof(mp_int_t) * 8 - mp_clz_mpi(val) + 7) / 8;
} else if (val < -1) {
slen = (sizeof(mp_int_t) * 8 - mp_clz_mpi(~val) + 8) / 8;
} else {
// clz of 0 is defined, so 0 and -1 map to 0 and 1
slen = -val;
}
if (slen <= dlen) {
memset(data, val < 0 ? 0xFF : 0x00, dlen);
mp_binary_set_int(slen, big_endian, data + (big_endian ? (dlen - slen) : 0), val);
overflow = false;
} else {
overflow = true;
}
}
if (overflow) {
mp_raise_msg(&mp_type_OverflowError, MP_ERROR_TEXT("buffer too small"));
}
return mp_obj_new_bytes_from_vstr(&vstr);