py/objstr: Skip whitespace in bytes.fromhex().

Skip whitespace characters between pairs of hex numbers.
This makes `bytes.fromhex()` compatible with cpython.

Includes simple test in `tests/basic/builtin_str_hex.py`.

Signed-off-by: Glenn Moloney <glenn.moloney@gmail.com>

py/objstr.c

@@ -2014,27 +2014,21 @@ mp_obj_t mp_obj_bytes_fromhex(mp_obj_t type_in, mp_obj_t data) {
     mp_buffer_info_t bufinfo;
     mp_get_buffer_raise(data, &bufinfo, MP_BUFFER_READ);
 
-    if ((bufinfo.len & 1) != 0) {
-        mp_raise_ValueError(MP_ERROR_TEXT("odd-length string"));
-    }
     vstr_t vstr;
     vstr_init_len(&vstr, bufinfo.len / 2);
     byte *in = bufinfo.buf, *out = (byte *)vstr.buf;
-    byte hex_byte = 0;
-    for (mp_uint_t i = bufinfo.len; i--;) {
-        byte hex_ch = *in++;
-        if (unichar_isxdigit(hex_ch)) {
-            hex_byte += unichar_xdigit_value(hex_ch);
-        } else {
-            mp_raise_ValueError(MP_ERROR_TEXT("non-hex digit found"));
+    byte *in_end = in + bufinfo.len;
+    mp_uint_t ch1, ch2;
+    while (in < in_end) {
+        if (unichar_isspace(ch1 = *in++)) {
+            continue;  // Skip whitespace between hex digit pairs
         }
-        if (i & 1) {
-            hex_byte <<= 4;
-        } else {
-            *out++ = hex_byte;
-            hex_byte = 0;
+        if (in == in_end || !unichar_isxdigit(ch1) || !unichar_isxdigit(ch2 = *in++)) {
+            mp_raise_ValueError(MP_ERROR_TEXT("non-hex digit"));
         }
+        *out++ = (byte)((unichar_xdigit_value(ch1) << 4) | unichar_xdigit_value(ch2));
     }
+    vstr.len = out - (byte *)vstr.buf;  // Length may be shorter due to whitespace in input
     return mp_obj_new_str_type_from_vstr(MP_OBJ_TO_PTR(type_in), &vstr);
 }
 

tests/basics/builtin_str_hex.py

@@ -20,5 +20,20 @@ for x in (
     "08090a0b0c0d0e0f",
     "7f80ff",
     "313233344142434461626364",
+    "ab\tcd\n  ef ",
+    "ab cd ef",
+    "ab cd ef ",
+    " ab cd ef ",
+    # Invalid hex strings:
+    "abcde",  # Odd number of hex digits
+    "ab cd e",
+    "a b cd ef",  # Spaces between hex pairs
+    "ab cd e f ",
+    "abga",  # Invalid hex digits
+    "ab_cd",
+    "ab:cd",
 ):
+    try:
         print(bytes.fromhex(x))
+    except ValueError as e:
+        print("ValueError:", e)

tests/basics/builtin_str_hex.py.exp

@@ -26,3 +26,14 @@ b'\x00\x01\x02\x03\x04\x05\x06\x07'
 b'\x08\t\n\x0b\x0c\r\x0e\x0f'
 b'\x7f\x80\xff'
 b'1234ABCDabcd'
+b'\xab\xcd\xef'
+b'\xab\xcd\xef'
+b'\xab\xcd\xef'
+b'\xab\xcd\xef'
+ValueError: non-hex digit
+ValueError: non-hex digit
+ValueError: non-hex digit
+ValueError: non-hex digit
+ValueError: non-hex digit
+ValueError: non-hex digit
+ValueError: non-hex digit