mirrors/cpython
1
0
Fork 0
mirror of https://github.com/python/cpython.git synced 2025-09-27 02:39:58 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions 13

gh-93575: Use correct way to calculate PyUnicode struct sizes (GH-93602)

Browse source 
* gh-93575: Use correct way to calculate PyUnicode struct sizes

* Add comment to keep test_sys and test_unicode in sync

* Fix case code < 256
This commit is contained in:
Christian Heimes 2022-06-08 20:18:08 +02:00 • committed by GitHub
parent 2dece90808
commit 5442561c1a
No known key found for this signature in database
GPG key ID: 4AEE18F83AFDEB23
3 changed files with 23 additions and 5 deletions
Download patch file Download diff file Expand all files Collapse all files

1
Lib/test/test_sys.py
View file

@ -1539,6 +1539,7 @@ class SizeofTest(unittest.TestCase):
samples = ['1'*100, '\xff'*50, samples = ['1'*100, '\xff'*50,
'\u0100'*40, '\uffff'*100, '\u0100'*40, '\uffff'*100,
'\U00010000'*30, '\U0010ffff'*100] '\U00010000'*30, '\U0010ffff'*100]
# also update field definitions in test_unicode.test_raiseMemError
asciifields = "nnb" asciifields = "nnb"
compactfields = asciifields + "nP" compactfields = asciifields + "nP"
unicodefields = compactfields + "P" unicodefields = compactfields + "P"

23
Lib/test/test_unicode.py
View file

@ -2370,15 +2370,19 @@ class UnicodeTest(string_tests.CommonTest,
self.assertIs(s.expandtabs(), s) self.assertIs(s.expandtabs(), s)
def test_raiseMemError(self): def test_raiseMemError(self):
null_byte = 1 asciifields = "nnb"
ascii_struct_size = sys.getsizeof("a") - len("a") - null_byte compactfields = asciifields + "nP"
compact_struct_size = sys.getsizeof("\xff") - len("\xff") - null_byte ascii_struct_size = support.calcobjsize(asciifields)
compact_struct_size = support.calcobjsize(compactfields)
for char in ('a', '\xe9', '\u20ac', '\U0010ffff'): for char in ('a', '\xe9', '\u20ac', '\U0010ffff'):
code = ord(char) code = ord(char)
if code < 0x100: if code < 0x80:
char_size = 1 # sizeof(Py_UCS1) char_size = 1 # sizeof(Py_UCS1)
struct_size = ascii_struct_size struct_size = ascii_struct_size
elif code < 0x100:
char_size = 1 # sizeof(Py_UCS1)
struct_size = compact_struct_size
elif code < 0x10000: elif code < 0x10000:
char_size = 2 # sizeof(Py_UCS2) char_size = 2 # sizeof(Py_UCS2)
struct_size = compact_struct_size struct_size = compact_struct_size
@ -2390,7 +2394,16 @@ class UnicodeTest(string_tests.CommonTest,
# be allocatable, given enough memory. # be allocatable, given enough memory.
maxlen = ((sys.maxsize - struct_size) // char_size) maxlen = ((sys.maxsize - struct_size) // char_size)
alloc = lambda: char * maxlen alloc = lambda: char * maxlen
with self.subTest(char=char): with self.subTest(
char=char,
struct_size=struct_size,
char_size=char_size
):
# self-check
self.assertEqual(
sys.getsizeof(char * 42),
struct_size + (char_size * (42 + 1))
)
self.assertRaises(MemoryError, alloc) self.assertRaises(MemoryError, alloc)
self.assertRaises(MemoryError, alloc) self.assertRaises(MemoryError, alloc)

4
Misc/NEWS.d/next/Tests/2022-06-08-14-17-59.gh-issue-93575.Xb2LNB.rst Normal file
View file

@ -0,0 +1,4 @@
Fix issue with test_unicode test_raiseMemError. The test case now use
``test.support.calcobjsize`` to calculate size of PyUnicode structs.
:func:`sys.getsizeof` may return different size when string has UTF-8
memory.