mirror of
https://github.com/python/cpython.git
synced 2025-07-08 03:45:36 +00:00

Provide consistent and informative error messages in function struct.pack when its integral arguments are not in range.
955 lines
38 KiB
Python
955 lines
38 KiB
Python
from collections import abc
|
|
import array
|
|
import gc
|
|
import math
|
|
import operator
|
|
import unittest
|
|
import struct
|
|
import sys
|
|
import weakref
|
|
|
|
from test import support
|
|
from test.support import import_helper
|
|
from test.support.script_helper import assert_python_ok
|
|
|
|
ISBIGENDIAN = sys.byteorder == "big"
|
|
|
|
integer_codes = 'b', 'B', 'h', 'H', 'i', 'I', 'l', 'L', 'q', 'Q', 'n', 'N'
|
|
byteorders = '', '@', '=', '<', '>', '!'
|
|
|
|
def iter_integer_formats(byteorders=byteorders):
|
|
for code in integer_codes:
|
|
for byteorder in byteorders:
|
|
if (byteorder not in ('', '@') and code in ('n', 'N')):
|
|
continue
|
|
yield code, byteorder
|
|
|
|
def string_reverse(s):
|
|
return s[::-1]
|
|
|
|
def bigendian_to_native(value):
|
|
if ISBIGENDIAN:
|
|
return value
|
|
else:
|
|
return string_reverse(value)
|
|
|
|
class StructTest(unittest.TestCase):
|
|
def test_isbigendian(self):
|
|
self.assertEqual((struct.pack('=i', 1)[0] == 0), ISBIGENDIAN)
|
|
|
|
def test_consistence(self):
|
|
self.assertRaises(struct.error, struct.calcsize, 'Z')
|
|
|
|
sz = struct.calcsize('i')
|
|
self.assertEqual(sz * 3, struct.calcsize('iii'))
|
|
|
|
fmt = 'cbxxxxxxhhhhiillffd?'
|
|
fmt3 = '3c3b18x12h6i6l6f3d3?'
|
|
sz = struct.calcsize(fmt)
|
|
sz3 = struct.calcsize(fmt3)
|
|
self.assertEqual(sz * 3, sz3)
|
|
|
|
self.assertRaises(struct.error, struct.pack, 'iii', 3)
|
|
self.assertRaises(struct.error, struct.pack, 'i', 3, 3, 3)
|
|
self.assertRaises((TypeError, struct.error), struct.pack, 'i', 'foo')
|
|
self.assertRaises((TypeError, struct.error), struct.pack, 'P', 'foo')
|
|
self.assertRaises(struct.error, struct.unpack, 'd', b'flap')
|
|
s = struct.pack('ii', 1, 2)
|
|
self.assertRaises(struct.error, struct.unpack, 'iii', s)
|
|
self.assertRaises(struct.error, struct.unpack, 'i', s)
|
|
|
|
def test_transitiveness(self):
|
|
c = b'a'
|
|
b = 1
|
|
h = 255
|
|
i = 65535
|
|
l = 65536
|
|
f = 3.1415
|
|
d = 3.1415
|
|
t = True
|
|
|
|
for prefix in ('', '@', '<', '>', '=', '!'):
|
|
for format in ('xcbhilfd?', 'xcBHILfd?'):
|
|
format = prefix + format
|
|
s = struct.pack(format, c, b, h, i, l, f, d, t)
|
|
cp, bp, hp, ip, lp, fp, dp, tp = struct.unpack(format, s)
|
|
self.assertEqual(cp, c)
|
|
self.assertEqual(bp, b)
|
|
self.assertEqual(hp, h)
|
|
self.assertEqual(ip, i)
|
|
self.assertEqual(lp, l)
|
|
self.assertEqual(int(100 * fp), int(100 * f))
|
|
self.assertEqual(int(100 * dp), int(100 * d))
|
|
self.assertEqual(tp, t)
|
|
|
|
def test_new_features(self):
|
|
# Test some of the new features in detail
|
|
# (format, argument, big-endian result, little-endian result, asymmetric)
|
|
tests = [
|
|
('c', b'a', b'a', b'a', 0),
|
|
('xc', b'a', b'\0a', b'\0a', 0),
|
|
('cx', b'a', b'a\0', b'a\0', 0),
|
|
('s', b'a', b'a', b'a', 0),
|
|
('0s', b'helloworld', b'', b'', 1),
|
|
('1s', b'helloworld', b'h', b'h', 1),
|
|
('9s', b'helloworld', b'helloworl', b'helloworl', 1),
|
|
('10s', b'helloworld', b'helloworld', b'helloworld', 0),
|
|
('11s', b'helloworld', b'helloworld\0', b'helloworld\0', 1),
|
|
('20s', b'helloworld', b'helloworld'+10*b'\0', b'helloworld'+10*b'\0', 1),
|
|
('b', 7, b'\7', b'\7', 0),
|
|
('b', -7, b'\371', b'\371', 0),
|
|
('B', 7, b'\7', b'\7', 0),
|
|
('B', 249, b'\371', b'\371', 0),
|
|
('h', 700, b'\002\274', b'\274\002', 0),
|
|
('h', -700, b'\375D', b'D\375', 0),
|
|
('H', 700, b'\002\274', b'\274\002', 0),
|
|
('H', 0x10000-700, b'\375D', b'D\375', 0),
|
|
('i', 70000000, b'\004,\035\200', b'\200\035,\004', 0),
|
|
('i', -70000000, b'\373\323\342\200', b'\200\342\323\373', 0),
|
|
('I', 70000000, b'\004,\035\200', b'\200\035,\004', 0),
|
|
('I', 0x100000000-70000000, b'\373\323\342\200', b'\200\342\323\373', 0),
|
|
('l', 70000000, b'\004,\035\200', b'\200\035,\004', 0),
|
|
('l', -70000000, b'\373\323\342\200', b'\200\342\323\373', 0),
|
|
('L', 70000000, b'\004,\035\200', b'\200\035,\004', 0),
|
|
('L', 0x100000000-70000000, b'\373\323\342\200', b'\200\342\323\373', 0),
|
|
('f', 2.0, b'@\000\000\000', b'\000\000\000@', 0),
|
|
('d', 2.0, b'@\000\000\000\000\000\000\000',
|
|
b'\000\000\000\000\000\000\000@', 0),
|
|
('f', -2.0, b'\300\000\000\000', b'\000\000\000\300', 0),
|
|
('d', -2.0, b'\300\000\000\000\000\000\000\000',
|
|
b'\000\000\000\000\000\000\000\300', 0),
|
|
('?', 0, b'\0', b'\0', 0),
|
|
('?', 3, b'\1', b'\1', 1),
|
|
('?', True, b'\1', b'\1', 0),
|
|
('?', [], b'\0', b'\0', 1),
|
|
('?', (1,), b'\1', b'\1', 1),
|
|
]
|
|
|
|
for fmt, arg, big, lil, asy in tests:
|
|
for (xfmt, exp) in [('>'+fmt, big), ('!'+fmt, big), ('<'+fmt, lil),
|
|
('='+fmt, ISBIGENDIAN and big or lil)]:
|
|
res = struct.pack(xfmt, arg)
|
|
self.assertEqual(res, exp)
|
|
self.assertEqual(struct.calcsize(xfmt), len(res))
|
|
rev = struct.unpack(xfmt, res)[0]
|
|
if rev != arg:
|
|
self.assertTrue(asy)
|
|
|
|
def test_calcsize(self):
|
|
expected_size = {
|
|
'b': 1, 'B': 1,
|
|
'h': 2, 'H': 2,
|
|
'i': 4, 'I': 4,
|
|
'l': 4, 'L': 4,
|
|
'q': 8, 'Q': 8,
|
|
}
|
|
|
|
# standard integer sizes
|
|
for code, byteorder in iter_integer_formats(('=', '<', '>', '!')):
|
|
format = byteorder+code
|
|
size = struct.calcsize(format)
|
|
self.assertEqual(size, expected_size[code])
|
|
|
|
# native integer sizes
|
|
native_pairs = 'bB', 'hH', 'iI', 'lL', 'nN', 'qQ'
|
|
for format_pair in native_pairs:
|
|
for byteorder in '', '@':
|
|
signed_size = struct.calcsize(byteorder + format_pair[0])
|
|
unsigned_size = struct.calcsize(byteorder + format_pair[1])
|
|
self.assertEqual(signed_size, unsigned_size)
|
|
|
|
# bounds for native integer sizes
|
|
self.assertEqual(struct.calcsize('b'), 1)
|
|
self.assertLessEqual(2, struct.calcsize('h'))
|
|
self.assertLessEqual(4, struct.calcsize('l'))
|
|
self.assertLessEqual(struct.calcsize('h'), struct.calcsize('i'))
|
|
self.assertLessEqual(struct.calcsize('i'), struct.calcsize('l'))
|
|
self.assertLessEqual(8, struct.calcsize('q'))
|
|
self.assertLessEqual(struct.calcsize('l'), struct.calcsize('q'))
|
|
self.assertGreaterEqual(struct.calcsize('n'), struct.calcsize('i'))
|
|
self.assertGreaterEqual(struct.calcsize('n'), struct.calcsize('P'))
|
|
|
|
def test_integers(self):
|
|
# Integer tests (bBhHiIlLqQnN).
|
|
import binascii
|
|
|
|
class IntTester(unittest.TestCase):
|
|
def __init__(self, format):
|
|
super(IntTester, self).__init__(methodName='test_one')
|
|
self.format = format
|
|
self.code = format[-1]
|
|
self.byteorder = format[:-1]
|
|
if not self.byteorder in byteorders:
|
|
raise ValueError("unrecognized packing byteorder: %s" %
|
|
self.byteorder)
|
|
self.bytesize = struct.calcsize(format)
|
|
self.bitsize = self.bytesize * 8
|
|
if self.code in tuple('bhilqn'):
|
|
self.signed = True
|
|
self.min_value = -(2**(self.bitsize-1))
|
|
self.max_value = 2**(self.bitsize-1) - 1
|
|
elif self.code in tuple('BHILQN'):
|
|
self.signed = False
|
|
self.min_value = 0
|
|
self.max_value = 2**self.bitsize - 1
|
|
else:
|
|
raise ValueError("unrecognized format code: %s" %
|
|
self.code)
|
|
|
|
def test_one(self, x, pack=struct.pack,
|
|
unpack=struct.unpack,
|
|
unhexlify=binascii.unhexlify):
|
|
|
|
format = self.format
|
|
if self.min_value <= x <= self.max_value:
|
|
expected = x
|
|
if self.signed and x < 0:
|
|
expected += 1 << self.bitsize
|
|
self.assertGreaterEqual(expected, 0)
|
|
expected = '%x' % expected
|
|
if len(expected) & 1:
|
|
expected = "0" + expected
|
|
expected = expected.encode('ascii')
|
|
expected = unhexlify(expected)
|
|
expected = (b"\x00" * (self.bytesize - len(expected)) +
|
|
expected)
|
|
if (self.byteorder == '<' or
|
|
self.byteorder in ('', '@', '=') and not ISBIGENDIAN):
|
|
expected = string_reverse(expected)
|
|
self.assertEqual(len(expected), self.bytesize)
|
|
|
|
# Pack work?
|
|
got = pack(format, x)
|
|
self.assertEqual(got, expected)
|
|
|
|
# Unpack work?
|
|
retrieved = unpack(format, got)[0]
|
|
self.assertEqual(x, retrieved)
|
|
|
|
# Adding any byte should cause a "too big" error.
|
|
self.assertRaises((struct.error, TypeError), unpack, format,
|
|
b'\x01' + got)
|
|
else:
|
|
# x is out of range -- verify pack realizes that.
|
|
self.assertRaises((OverflowError, ValueError, struct.error),
|
|
pack, format, x)
|
|
|
|
def run(self):
|
|
from random import randrange
|
|
|
|
# Create all interesting powers of 2.
|
|
values = []
|
|
for exp in range(self.bitsize + 3):
|
|
values.append(1 << exp)
|
|
|
|
# Add some random values.
|
|
for i in range(self.bitsize):
|
|
val = 0
|
|
for j in range(self.bytesize):
|
|
val = (val << 8) | randrange(256)
|
|
values.append(val)
|
|
|
|
# Values absorbed from other tests
|
|
values.extend([300, 700000, sys.maxsize*4])
|
|
|
|
# Try all those, and their negations, and +-1 from
|
|
# them. Note that this tests all power-of-2
|
|
# boundaries in range, and a few out of range, plus
|
|
# +-(2**n +- 1).
|
|
for base in values:
|
|
for val in -base, base:
|
|
for incr in -1, 0, 1:
|
|
x = val + incr
|
|
self.test_one(x)
|
|
|
|
# Some error cases.
|
|
class NotAnInt:
|
|
def __int__(self):
|
|
return 42
|
|
|
|
# Objects with an '__index__' method should be allowed
|
|
# to pack as integers. That is assuming the implemented
|
|
# '__index__' method returns an 'int'.
|
|
class Indexable(object):
|
|
def __init__(self, value):
|
|
self._value = value
|
|
|
|
def __index__(self):
|
|
return self._value
|
|
|
|
# If the '__index__' method raises a type error, then
|
|
# '__int__' should be used with a deprecation warning.
|
|
class BadIndex(object):
|
|
def __index__(self):
|
|
raise TypeError
|
|
|
|
def __int__(self):
|
|
return 42
|
|
|
|
self.assertRaises((TypeError, struct.error),
|
|
struct.pack, self.format,
|
|
"a string")
|
|
self.assertRaises((TypeError, struct.error),
|
|
struct.pack, self.format,
|
|
randrange)
|
|
self.assertRaises((TypeError, struct.error),
|
|
struct.pack, self.format,
|
|
3+42j)
|
|
self.assertRaises((TypeError, struct.error),
|
|
struct.pack, self.format,
|
|
NotAnInt())
|
|
self.assertRaises((TypeError, struct.error),
|
|
struct.pack, self.format,
|
|
BadIndex())
|
|
|
|
# Check for legitimate values from '__index__'.
|
|
for obj in (Indexable(0), Indexable(10), Indexable(17),
|
|
Indexable(42), Indexable(100), Indexable(127)):
|
|
try:
|
|
struct.pack(format, obj)
|
|
except:
|
|
self.fail("integer code pack failed on object "
|
|
"with '__index__' method")
|
|
|
|
# Check for bogus values from '__index__'.
|
|
for obj in (Indexable(b'a'), Indexable('b'), Indexable(None),
|
|
Indexable({'a': 1}), Indexable([1, 2, 3])):
|
|
self.assertRaises((TypeError, struct.error),
|
|
struct.pack, self.format,
|
|
obj)
|
|
|
|
for code, byteorder in iter_integer_formats():
|
|
format = byteorder+code
|
|
t = IntTester(format)
|
|
t.run()
|
|
|
|
def test_nN_code(self):
|
|
# n and N don't exist in standard sizes
|
|
def assertStructError(func, *args, **kwargs):
|
|
with self.assertRaises(struct.error) as cm:
|
|
func(*args, **kwargs)
|
|
self.assertIn("bad char in struct format", str(cm.exception))
|
|
for code in 'nN':
|
|
for byteorder in ('=', '<', '>', '!'):
|
|
format = byteorder+code
|
|
assertStructError(struct.calcsize, format)
|
|
assertStructError(struct.pack, format, 0)
|
|
assertStructError(struct.unpack, format, b"")
|
|
|
|
def test_p_code(self):
|
|
# Test p ("Pascal string") code.
|
|
for code, input, expected, expectedback in [
|
|
('p', b'abc', b'\x00', b''),
|
|
('1p', b'abc', b'\x00', b''),
|
|
('2p', b'abc', b'\x01a', b'a'),
|
|
('3p', b'abc', b'\x02ab', b'ab'),
|
|
('4p', b'abc', b'\x03abc', b'abc'),
|
|
('5p', b'abc', b'\x03abc\x00', b'abc'),
|
|
('6p', b'abc', b'\x03abc\x00\x00', b'abc'),
|
|
('1000p', b'x'*1000, b'\xff' + b'x'*999, b'x'*255)]:
|
|
got = struct.pack(code, input)
|
|
self.assertEqual(got, expected)
|
|
(got,) = struct.unpack(code, got)
|
|
self.assertEqual(got, expectedback)
|
|
|
|
def test_705836(self):
|
|
# SF bug 705836. "<f" and ">f" had a severe rounding bug, where a carry
|
|
# from the low-order discarded bits could propagate into the exponent
|
|
# field, causing the result to be wrong by a factor of 2.
|
|
for base in range(1, 33):
|
|
# smaller <- largest representable float less than base.
|
|
delta = 0.5
|
|
while base - delta / 2.0 != base:
|
|
delta /= 2.0
|
|
smaller = base - delta
|
|
# Packing this rounds away a solid string of trailing 1 bits.
|
|
packed = struct.pack("<f", smaller)
|
|
unpacked = struct.unpack("<f", packed)[0]
|
|
# This failed at base = 2, 4, and 32, with unpacked = 1, 2, and
|
|
# 16, respectively.
|
|
self.assertEqual(base, unpacked)
|
|
bigpacked = struct.pack(">f", smaller)
|
|
self.assertEqual(bigpacked, string_reverse(packed))
|
|
unpacked = struct.unpack(">f", bigpacked)[0]
|
|
self.assertEqual(base, unpacked)
|
|
|
|
# Largest finite IEEE single.
|
|
big = (1 << 24) - 1
|
|
big = math.ldexp(big, 127 - 23)
|
|
packed = struct.pack(">f", big)
|
|
unpacked = struct.unpack(">f", packed)[0]
|
|
self.assertEqual(big, unpacked)
|
|
|
|
# The same, but tack on a 1 bit so it rounds up to infinity.
|
|
big = (1 << 25) - 1
|
|
big = math.ldexp(big, 127 - 24)
|
|
self.assertRaises(OverflowError, struct.pack, ">f", big)
|
|
|
|
def test_1530559(self):
|
|
for code, byteorder in iter_integer_formats():
|
|
format = byteorder + code
|
|
self.assertRaises(struct.error, struct.pack, format, 1.0)
|
|
self.assertRaises(struct.error, struct.pack, format, 1.5)
|
|
self.assertRaises(struct.error, struct.pack, 'P', 1.0)
|
|
self.assertRaises(struct.error, struct.pack, 'P', 1.5)
|
|
|
|
def test_unpack_from(self):
|
|
test_string = b'abcd01234'
|
|
fmt = '4s'
|
|
s = struct.Struct(fmt)
|
|
for cls in (bytes, bytearray):
|
|
data = cls(test_string)
|
|
self.assertEqual(s.unpack_from(data), (b'abcd',))
|
|
self.assertEqual(s.unpack_from(data, 2), (b'cd01',))
|
|
self.assertEqual(s.unpack_from(data, 4), (b'0123',))
|
|
for i in range(6):
|
|
self.assertEqual(s.unpack_from(data, i), (data[i:i+4],))
|
|
for i in range(6, len(test_string) + 1):
|
|
self.assertRaises(struct.error, s.unpack_from, data, i)
|
|
for cls in (bytes, bytearray):
|
|
data = cls(test_string)
|
|
self.assertEqual(struct.unpack_from(fmt, data), (b'abcd',))
|
|
self.assertEqual(struct.unpack_from(fmt, data, 2), (b'cd01',))
|
|
self.assertEqual(struct.unpack_from(fmt, data, 4), (b'0123',))
|
|
for i in range(6):
|
|
self.assertEqual(struct.unpack_from(fmt, data, i), (data[i:i+4],))
|
|
for i in range(6, len(test_string) + 1):
|
|
self.assertRaises(struct.error, struct.unpack_from, fmt, data, i)
|
|
|
|
# keyword arguments
|
|
self.assertEqual(s.unpack_from(buffer=test_string, offset=2),
|
|
(b'cd01',))
|
|
|
|
def test_pack_into(self):
|
|
test_string = b'Reykjavik rocks, eow!'
|
|
writable_buf = array.array('b', b' '*100)
|
|
fmt = '21s'
|
|
s = struct.Struct(fmt)
|
|
|
|
# Test without offset
|
|
s.pack_into(writable_buf, 0, test_string)
|
|
from_buf = writable_buf.tobytes()[:len(test_string)]
|
|
self.assertEqual(from_buf, test_string)
|
|
|
|
# Test with offset.
|
|
s.pack_into(writable_buf, 10, test_string)
|
|
from_buf = writable_buf.tobytes()[:len(test_string)+10]
|
|
self.assertEqual(from_buf, test_string[:10] + test_string)
|
|
|
|
# Go beyond boundaries.
|
|
small_buf = array.array('b', b' '*10)
|
|
self.assertRaises((ValueError, struct.error), s.pack_into, small_buf, 0,
|
|
test_string)
|
|
self.assertRaises((ValueError, struct.error), s.pack_into, small_buf, 2,
|
|
test_string)
|
|
|
|
# Test bogus offset (issue 3694)
|
|
sb = small_buf
|
|
self.assertRaises((TypeError, struct.error), struct.pack_into, b'', sb,
|
|
None)
|
|
|
|
def test_pack_into_fn(self):
|
|
test_string = b'Reykjavik rocks, eow!'
|
|
writable_buf = array.array('b', b' '*100)
|
|
fmt = '21s'
|
|
pack_into = lambda *args: struct.pack_into(fmt, *args)
|
|
|
|
# Test without offset.
|
|
pack_into(writable_buf, 0, test_string)
|
|
from_buf = writable_buf.tobytes()[:len(test_string)]
|
|
self.assertEqual(from_buf, test_string)
|
|
|
|
# Test with offset.
|
|
pack_into(writable_buf, 10, test_string)
|
|
from_buf = writable_buf.tobytes()[:len(test_string)+10]
|
|
self.assertEqual(from_buf, test_string[:10] + test_string)
|
|
|
|
# Go beyond boundaries.
|
|
small_buf = array.array('b', b' '*10)
|
|
self.assertRaises((ValueError, struct.error), pack_into, small_buf, 0,
|
|
test_string)
|
|
self.assertRaises((ValueError, struct.error), pack_into, small_buf, 2,
|
|
test_string)
|
|
|
|
def test_unpack_with_buffer(self):
|
|
# SF bug 1563759: struct.unpack doesn't support buffer protocol objects
|
|
data1 = array.array('B', b'\x12\x34\x56\x78')
|
|
data2 = memoryview(b'\x12\x34\x56\x78') # XXX b'......XXXX......', 6, 4
|
|
for data in [data1, data2]:
|
|
value, = struct.unpack('>I', data)
|
|
self.assertEqual(value, 0x12345678)
|
|
|
|
def test_bool(self):
|
|
class ExplodingBool(object):
|
|
def __bool__(self):
|
|
raise OSError
|
|
for prefix in tuple("<>!=")+('',):
|
|
false = (), [], [], '', 0
|
|
true = [1], 'test', 5, -1, 0xffffffff+1, 0xffffffff/2
|
|
|
|
falseFormat = prefix + '?' * len(false)
|
|
packedFalse = struct.pack(falseFormat, *false)
|
|
unpackedFalse = struct.unpack(falseFormat, packedFalse)
|
|
|
|
trueFormat = prefix + '?' * len(true)
|
|
packedTrue = struct.pack(trueFormat, *true)
|
|
unpackedTrue = struct.unpack(trueFormat, packedTrue)
|
|
|
|
self.assertEqual(len(true), len(unpackedTrue))
|
|
self.assertEqual(len(false), len(unpackedFalse))
|
|
|
|
for t in unpackedFalse:
|
|
self.assertFalse(t)
|
|
for t in unpackedTrue:
|
|
self.assertTrue(t)
|
|
|
|
packed = struct.pack(prefix+'?', 1)
|
|
|
|
self.assertEqual(len(packed), struct.calcsize(prefix+'?'))
|
|
|
|
if len(packed) != 1:
|
|
self.assertFalse(prefix, msg='encoded bool is not one byte: %r'
|
|
%packed)
|
|
|
|
try:
|
|
struct.pack(prefix + '?', ExplodingBool())
|
|
except OSError:
|
|
pass
|
|
else:
|
|
self.fail("Expected OSError: struct.pack(%r, "
|
|
"ExplodingBool())" % (prefix + '?'))
|
|
|
|
for c in [b'\x01', b'\x7f', b'\xff', b'\x0f', b'\xf0']:
|
|
self.assertTrue(struct.unpack('>?', c)[0])
|
|
|
|
def test_count_overflow(self):
|
|
hugecount = '{}b'.format(sys.maxsize+1)
|
|
self.assertRaises(struct.error, struct.calcsize, hugecount)
|
|
|
|
hugecount2 = '{}b{}H'.format(sys.maxsize//2, sys.maxsize//2)
|
|
self.assertRaises(struct.error, struct.calcsize, hugecount2)
|
|
|
|
def test_trailing_counter(self):
|
|
store = array.array('b', b' '*100)
|
|
|
|
# format lists containing only count spec should result in an error
|
|
self.assertRaises(struct.error, struct.pack, '12345')
|
|
self.assertRaises(struct.error, struct.unpack, '12345', b'')
|
|
self.assertRaises(struct.error, struct.pack_into, '12345', store, 0)
|
|
self.assertRaises(struct.error, struct.unpack_from, '12345', store, 0)
|
|
|
|
# Format lists with trailing count spec should result in an error
|
|
self.assertRaises(struct.error, struct.pack, 'c12345', 'x')
|
|
self.assertRaises(struct.error, struct.unpack, 'c12345', b'x')
|
|
self.assertRaises(struct.error, struct.pack_into, 'c12345', store, 0,
|
|
'x')
|
|
self.assertRaises(struct.error, struct.unpack_from, 'c12345', store,
|
|
0)
|
|
|
|
# Mixed format tests
|
|
self.assertRaises(struct.error, struct.pack, '14s42', 'spam and eggs')
|
|
self.assertRaises(struct.error, struct.unpack, '14s42',
|
|
b'spam and eggs')
|
|
self.assertRaises(struct.error, struct.pack_into, '14s42', store, 0,
|
|
'spam and eggs')
|
|
self.assertRaises(struct.error, struct.unpack_from, '14s42', store, 0)
|
|
|
|
def test_Struct_reinitialization(self):
|
|
# Issue 9422: there was a memory leak when reinitializing a
|
|
# Struct instance. This test can be used to detect the leak
|
|
# when running with regrtest -L.
|
|
s = struct.Struct('i')
|
|
s.__init__('ii')
|
|
|
|
def check_sizeof(self, format_str, number_of_codes):
|
|
# The size of 'PyStructObject'
|
|
totalsize = support.calcobjsize('2n3P')
|
|
# The size taken up by the 'formatcode' dynamic array
|
|
totalsize += struct.calcsize('P3n0P') * (number_of_codes + 1)
|
|
support.check_sizeof(self, struct.Struct(format_str), totalsize)
|
|
|
|
@support.cpython_only
|
|
def test__sizeof__(self):
|
|
for code in integer_codes:
|
|
self.check_sizeof(code, 1)
|
|
self.check_sizeof('BHILfdspP', 9)
|
|
self.check_sizeof('B' * 1234, 1234)
|
|
self.check_sizeof('fd', 2)
|
|
self.check_sizeof('xxxxxxxxxxxxxx', 0)
|
|
self.check_sizeof('100H', 1)
|
|
self.check_sizeof('187s', 1)
|
|
self.check_sizeof('20p', 1)
|
|
self.check_sizeof('0s', 1)
|
|
self.check_sizeof('0c', 0)
|
|
|
|
def test_boundary_error_message(self):
|
|
regex1 = (
|
|
r'pack_into requires a buffer of at least 6 '
|
|
r'bytes for packing 1 bytes at offset 5 '
|
|
r'\(actual buffer size is 1\)'
|
|
)
|
|
with self.assertRaisesRegex(struct.error, regex1):
|
|
struct.pack_into('b', bytearray(1), 5, 1)
|
|
|
|
regex2 = (
|
|
r'unpack_from requires a buffer of at least 6 '
|
|
r'bytes for unpacking 1 bytes at offset 5 '
|
|
r'\(actual buffer size is 1\)'
|
|
)
|
|
with self.assertRaisesRegex(struct.error, regex2):
|
|
struct.unpack_from('b', bytearray(1), 5)
|
|
|
|
def test_boundary_error_message_with_negative_offset(self):
|
|
byte_list = bytearray(10)
|
|
with self.assertRaisesRegex(
|
|
struct.error,
|
|
r'no space to pack 4 bytes at offset -2'):
|
|
struct.pack_into('<I', byte_list, -2, 123)
|
|
|
|
with self.assertRaisesRegex(
|
|
struct.error,
|
|
'offset -11 out of range for 10-byte buffer'):
|
|
struct.pack_into('<B', byte_list, -11, 123)
|
|
|
|
with self.assertRaisesRegex(
|
|
struct.error,
|
|
r'not enough data to unpack 4 bytes at offset -2'):
|
|
struct.unpack_from('<I', byte_list, -2)
|
|
|
|
with self.assertRaisesRegex(
|
|
struct.error,
|
|
"offset -11 out of range for 10-byte buffer"):
|
|
struct.unpack_from('<B', byte_list, -11)
|
|
|
|
def test_boundary_error_message_with_large_offset(self):
|
|
# Test overflows cause by large offset and value size (issue 30245)
|
|
regex1 = (
|
|
r'pack_into requires a buffer of at least ' + str(sys.maxsize + 4) +
|
|
r' bytes for packing 4 bytes at offset ' + str(sys.maxsize) +
|
|
r' \(actual buffer size is 10\)'
|
|
)
|
|
with self.assertRaisesRegex(struct.error, regex1):
|
|
struct.pack_into('<I', bytearray(10), sys.maxsize, 1)
|
|
|
|
regex2 = (
|
|
r'unpack_from requires a buffer of at least ' + str(sys.maxsize + 4) +
|
|
r' bytes for unpacking 4 bytes at offset ' + str(sys.maxsize) +
|
|
r' \(actual buffer size is 10\)'
|
|
)
|
|
with self.assertRaisesRegex(struct.error, regex2):
|
|
struct.unpack_from('<I', bytearray(10), sys.maxsize)
|
|
|
|
def test_issue29802(self):
|
|
# When the second argument of struct.unpack() was of wrong type
|
|
# the Struct object was decrefed twice and the reference to
|
|
# deallocated object was left in a cache.
|
|
with self.assertRaises(TypeError):
|
|
struct.unpack('b', 0)
|
|
# Shouldn't crash.
|
|
self.assertEqual(struct.unpack('b', b'a'), (b'a'[0],))
|
|
|
|
def test_format_attr(self):
|
|
s = struct.Struct('=i2H')
|
|
self.assertEqual(s.format, '=i2H')
|
|
|
|
# use a bytes string
|
|
s2 = struct.Struct(s.format.encode())
|
|
self.assertEqual(s2.format, s.format)
|
|
|
|
def test_struct_cleans_up_at_runtime_shutdown(self):
|
|
code = """if 1:
|
|
import struct
|
|
|
|
class C:
|
|
def __init__(self):
|
|
self.pack = struct.pack
|
|
def __del__(self):
|
|
self.pack('I', -42)
|
|
|
|
struct.x = C()
|
|
"""
|
|
rc, stdout, stderr = assert_python_ok("-c", code)
|
|
self.assertEqual(rc, 0)
|
|
self.assertEqual(stdout.rstrip(), b"")
|
|
self.assertIn(b"Exception ignored in:", stderr)
|
|
self.assertIn(b"C.__del__", stderr)
|
|
|
|
def test__struct_reference_cycle_cleaned_up(self):
|
|
# Regression test for python/cpython#94207.
|
|
|
|
# When we create a new struct module, trigger use of its cache,
|
|
# and then delete it ...
|
|
_struct_module = import_helper.import_fresh_module("_struct")
|
|
module_ref = weakref.ref(_struct_module)
|
|
_struct_module.calcsize("b")
|
|
del _struct_module
|
|
|
|
# Then the module should have been garbage collected.
|
|
gc.collect()
|
|
self.assertIsNone(
|
|
module_ref(), "_struct module was not garbage collected")
|
|
|
|
@support.cpython_only
|
|
def test__struct_types_immutable(self):
|
|
# See https://github.com/python/cpython/issues/94254
|
|
|
|
Struct = struct.Struct
|
|
unpack_iterator = type(struct.iter_unpack("b", b'x'))
|
|
for cls in (Struct, unpack_iterator):
|
|
with self.subTest(cls=cls):
|
|
with self.assertRaises(TypeError):
|
|
cls.x = 1
|
|
|
|
@support.cpython_only
|
|
def test__struct_Struct__new__initialized(self):
|
|
# See https://github.com/python/cpython/issues/78724
|
|
|
|
s = struct.Struct.__new__(struct.Struct, "b")
|
|
s.unpack_from(b"abcd")
|
|
|
|
@support.cpython_only
|
|
def test__struct_Struct_subclassing(self):
|
|
class Bob(struct.Struct):
|
|
pass
|
|
|
|
s = Bob("b")
|
|
s.unpack_from(b"abcd")
|
|
|
|
def test_issue35714(self):
|
|
# Embedded null characters should not be allowed in format strings.
|
|
for s in '\0', '2\0i', b'\0':
|
|
with self.assertRaisesRegex(struct.error,
|
|
'embedded null character'):
|
|
struct.calcsize(s)
|
|
|
|
@support.cpython_only
|
|
def test_issue98248(self):
|
|
def test_error_msg(prefix, int_type, is_unsigned):
|
|
fmt_str = prefix + int_type
|
|
size = struct.calcsize(fmt_str)
|
|
if is_unsigned:
|
|
max_ = 2 ** (size * 8) - 1
|
|
min_ = 0
|
|
else:
|
|
max_ = 2 ** (size * 8 - 1) - 1
|
|
min_ = -2 ** (size * 8 - 1)
|
|
error_msg = f"'{int_type}' format requires {min_} <= number <= {max_}"
|
|
for number in [int(-1e50), min_ - 1, max_ + 1, int(1e50)]:
|
|
with self.subTest(format_str=fmt_str, number=number):
|
|
with self.assertRaisesRegex(struct.error, error_msg):
|
|
struct.pack(fmt_str, number)
|
|
error_msg = "required argument is not an integer"
|
|
not_number = ""
|
|
with self.subTest(format_str=fmt_str, number=not_number):
|
|
with self.assertRaisesRegex(struct.error, error_msg):
|
|
struct.pack(fmt_str, not_number)
|
|
|
|
for prefix in '@=<>':
|
|
for int_type in 'BHILQ':
|
|
test_error_msg(prefix, int_type, True)
|
|
for int_type in 'bhilq':
|
|
test_error_msg(prefix, int_type, False)
|
|
|
|
int_type = 'N'
|
|
test_error_msg('@', int_type, True)
|
|
|
|
int_type = 'n'
|
|
test_error_msg('@', int_type, False)
|
|
|
|
@support.cpython_only
|
|
def test_issue98248_error_propagation(self):
|
|
class Div0:
|
|
def __index__(self):
|
|
1 / 0
|
|
|
|
def test_error_propagation(fmt_str):
|
|
with self.subTest(format_str=fmt_str, exception="ZeroDivisionError"):
|
|
with self.assertRaises(ZeroDivisionError):
|
|
struct.pack(fmt_str, Div0())
|
|
|
|
for prefix in '@=<>':
|
|
for int_type in 'BHILQbhilq':
|
|
test_error_propagation(prefix + int_type)
|
|
|
|
test_error_propagation('N')
|
|
test_error_propagation('n')
|
|
|
|
class UnpackIteratorTest(unittest.TestCase):
|
|
"""
|
|
Tests for iterative unpacking (struct.Struct.iter_unpack).
|
|
"""
|
|
|
|
def test_construct(self):
|
|
def _check_iterator(it):
|
|
self.assertIsInstance(it, abc.Iterator)
|
|
self.assertIsInstance(it, abc.Iterable)
|
|
s = struct.Struct('>ibcp')
|
|
it = s.iter_unpack(b"")
|
|
_check_iterator(it)
|
|
it = s.iter_unpack(b"1234567")
|
|
_check_iterator(it)
|
|
# Wrong bytes length
|
|
with self.assertRaises(struct.error):
|
|
s.iter_unpack(b"123456")
|
|
with self.assertRaises(struct.error):
|
|
s.iter_unpack(b"12345678")
|
|
# Zero-length struct
|
|
s = struct.Struct('>')
|
|
with self.assertRaises(struct.error):
|
|
s.iter_unpack(b"")
|
|
with self.assertRaises(struct.error):
|
|
s.iter_unpack(b"12")
|
|
|
|
def test_uninstantiable(self):
|
|
iter_unpack_type = type(struct.Struct(">ibcp").iter_unpack(b""))
|
|
self.assertRaises(TypeError, iter_unpack_type)
|
|
|
|
def test_iterate(self):
|
|
s = struct.Struct('>IB')
|
|
b = bytes(range(1, 16))
|
|
it = s.iter_unpack(b)
|
|
self.assertEqual(next(it), (0x01020304, 5))
|
|
self.assertEqual(next(it), (0x06070809, 10))
|
|
self.assertEqual(next(it), (0x0b0c0d0e, 15))
|
|
self.assertRaises(StopIteration, next, it)
|
|
self.assertRaises(StopIteration, next, it)
|
|
|
|
def test_arbitrary_buffer(self):
|
|
s = struct.Struct('>IB')
|
|
b = bytes(range(1, 11))
|
|
it = s.iter_unpack(memoryview(b))
|
|
self.assertEqual(next(it), (0x01020304, 5))
|
|
self.assertEqual(next(it), (0x06070809, 10))
|
|
self.assertRaises(StopIteration, next, it)
|
|
self.assertRaises(StopIteration, next, it)
|
|
|
|
def test_length_hint(self):
|
|
lh = operator.length_hint
|
|
s = struct.Struct('>IB')
|
|
b = bytes(range(1, 16))
|
|
it = s.iter_unpack(b)
|
|
self.assertEqual(lh(it), 3)
|
|
next(it)
|
|
self.assertEqual(lh(it), 2)
|
|
next(it)
|
|
self.assertEqual(lh(it), 1)
|
|
next(it)
|
|
self.assertEqual(lh(it), 0)
|
|
self.assertRaises(StopIteration, next, it)
|
|
self.assertEqual(lh(it), 0)
|
|
|
|
def test_module_func(self):
|
|
# Sanity check for the global struct.iter_unpack()
|
|
it = struct.iter_unpack('>IB', bytes(range(1, 11)))
|
|
self.assertEqual(next(it), (0x01020304, 5))
|
|
self.assertEqual(next(it), (0x06070809, 10))
|
|
self.assertRaises(StopIteration, next, it)
|
|
self.assertRaises(StopIteration, next, it)
|
|
|
|
def test_half_float(self):
|
|
# Little-endian examples from:
|
|
# http://en.wikipedia.org/wiki/Half_precision_floating-point_format
|
|
format_bits_float__cleanRoundtrip_list = [
|
|
(b'\x00\x3c', 1.0),
|
|
(b'\x00\xc0', -2.0),
|
|
(b'\xff\x7b', 65504.0), # (max half precision)
|
|
(b'\x00\x04', 2**-14), # ~= 6.10352 * 10**-5 (min pos normal)
|
|
(b'\x01\x00', 2**-24), # ~= 5.96046 * 10**-8 (min pos subnormal)
|
|
(b'\x00\x00', 0.0),
|
|
(b'\x00\x80', -0.0),
|
|
(b'\x00\x7c', float('+inf')),
|
|
(b'\x00\xfc', float('-inf')),
|
|
(b'\x55\x35', 0.333251953125), # ~= 1/3
|
|
]
|
|
|
|
for le_bits, f in format_bits_float__cleanRoundtrip_list:
|
|
be_bits = le_bits[::-1]
|
|
self.assertEqual(f, struct.unpack('<e', le_bits)[0])
|
|
self.assertEqual(le_bits, struct.pack('<e', f))
|
|
self.assertEqual(f, struct.unpack('>e', be_bits)[0])
|
|
self.assertEqual(be_bits, struct.pack('>e', f))
|
|
if sys.byteorder == 'little':
|
|
self.assertEqual(f, struct.unpack('e', le_bits)[0])
|
|
self.assertEqual(le_bits, struct.pack('e', f))
|
|
else:
|
|
self.assertEqual(f, struct.unpack('e', be_bits)[0])
|
|
self.assertEqual(be_bits, struct.pack('e', f))
|
|
|
|
# Check for NaN handling:
|
|
format_bits__nan_list = [
|
|
('<e', b'\x01\xfc'),
|
|
('<e', b'\x00\xfe'),
|
|
('<e', b'\xff\xff'),
|
|
('<e', b'\x01\x7c'),
|
|
('<e', b'\x00\x7e'),
|
|
('<e', b'\xff\x7f'),
|
|
]
|
|
|
|
for formatcode, bits in format_bits__nan_list:
|
|
self.assertTrue(math.isnan(struct.unpack('<e', bits)[0]))
|
|
self.assertTrue(math.isnan(struct.unpack('>e', bits[::-1])[0]))
|
|
|
|
# Check that packing produces a bit pattern representing a quiet NaN:
|
|
# all exponent bits and the msb of the fraction should all be 1.
|
|
packed = struct.pack('<e', math.nan)
|
|
self.assertEqual(packed[1] & 0x7e, 0x7e)
|
|
packed = struct.pack('<e', -math.nan)
|
|
self.assertEqual(packed[1] & 0x7e, 0x7e)
|
|
|
|
# Checks for round-to-even behavior
|
|
format_bits_float__rounding_list = [
|
|
('>e', b'\x00\x01', 2.0**-25 + 2.0**-35), # Rounds to minimum subnormal
|
|
('>e', b'\x00\x00', 2.0**-25), # Underflows to zero (nearest even mode)
|
|
('>e', b'\x00\x00', 2.0**-26), # Underflows to zero
|
|
('>e', b'\x03\xff', 2.0**-14 - 2.0**-24), # Largest subnormal.
|
|
('>e', b'\x03\xff', 2.0**-14 - 2.0**-25 - 2.0**-65),
|
|
('>e', b'\x04\x00', 2.0**-14 - 2.0**-25),
|
|
('>e', b'\x04\x00', 2.0**-14), # Smallest normal.
|
|
('>e', b'\x3c\x01', 1.0+2.0**-11 + 2.0**-16), # rounds to 1.0+2**(-10)
|
|
('>e', b'\x3c\x00', 1.0+2.0**-11), # rounds to 1.0 (nearest even mode)
|
|
('>e', b'\x3c\x00', 1.0+2.0**-12), # rounds to 1.0
|
|
('>e', b'\x7b\xff', 65504), # largest normal
|
|
('>e', b'\x7b\xff', 65519), # rounds to 65504
|
|
('>e', b'\x80\x01', -2.0**-25 - 2.0**-35), # Rounds to minimum subnormal
|
|
('>e', b'\x80\x00', -2.0**-25), # Underflows to zero (nearest even mode)
|
|
('>e', b'\x80\x00', -2.0**-26), # Underflows to zero
|
|
('>e', b'\xbc\x01', -1.0-2.0**-11 - 2.0**-16), # rounds to 1.0+2**(-10)
|
|
('>e', b'\xbc\x00', -1.0-2.0**-11), # rounds to 1.0 (nearest even mode)
|
|
('>e', b'\xbc\x00', -1.0-2.0**-12), # rounds to 1.0
|
|
('>e', b'\xfb\xff', -65519), # rounds to 65504
|
|
]
|
|
|
|
for formatcode, bits, f in format_bits_float__rounding_list:
|
|
self.assertEqual(bits, struct.pack(formatcode, f))
|
|
|
|
# This overflows, and so raises an error
|
|
format_bits_float__roundingError_list = [
|
|
# Values that round to infinity.
|
|
('>e', 65520.0),
|
|
('>e', 65536.0),
|
|
('>e', 1e300),
|
|
('>e', -65520.0),
|
|
('>e', -65536.0),
|
|
('>e', -1e300),
|
|
('<e', 65520.0),
|
|
('<e', 65536.0),
|
|
('<e', 1e300),
|
|
('<e', -65520.0),
|
|
('<e', -65536.0),
|
|
('<e', -1e300),
|
|
]
|
|
|
|
for formatcode, f in format_bits_float__roundingError_list:
|
|
self.assertRaises(OverflowError, struct.pack, formatcode, f)
|
|
|
|
# Double rounding
|
|
format_bits_float__doubleRoundingError_list = [
|
|
('>e', b'\x67\xff', 0x1ffdffffff * 2**-26), # should be 2047, if double-rounded 64>32>16, becomes 2048
|
|
]
|
|
|
|
for formatcode, bits, f in format_bits_float__doubleRoundingError_list:
|
|
self.assertEqual(bits, struct.pack(formatcode, f))
|
|
|
|
|
|
if __name__ == '__main__':
|
|
unittest.main()
|