Skip to content

[3.12] Move implementation specific RE tests to separate class (GH-106563) #106564

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 1 commit into from
Jul 9, 2023
Merged

[3.12] Move implementation specific RE tests to separate class (GH-106563) #106564

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
135 changes: 69 additions & 66 deletions Lib/test/test_re.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -1046,33 +1046,6 @@ def test_ignore_case_range(self):
def test_category(self):
self.assertEqual(re.match(r"(\s)", " ").group(1), " ")

@cpython_only
def test_case_helpers(self):
import _sre
for i in range(128):
c = chr(i)
lo = ord(c.lower())
self.assertEqual(_sre.ascii_tolower(i), lo)
self.assertEqual(_sre.unicode_tolower(i), lo)
iscased = c in string.ascii_letters
self.assertEqual(_sre.ascii_iscased(i), iscased)
self.assertEqual(_sre.unicode_iscased(i), iscased)

for i in list(range(128, 0x1000)) + [0x10400, 0x10428]:
c = chr(i)
self.assertEqual(_sre.ascii_tolower(i), i)
if i != 0x0130:
self.assertEqual(_sre.unicode_tolower(i), ord(c.lower()))
iscased = c != c.lower() or c != c.upper()
self.assertFalse(_sre.ascii_iscased(i))
self.assertEqual(_sre.unicode_iscased(i),
c != c.lower() or c != c.upper())

self.assertEqual(_sre.ascii_tolower(0x0130), 0x0130)
self.assertEqual(_sre.unicode_tolower(0x0130), ord('i'))
self.assertFalse(_sre.ascii_iscased(0x0130))
self.assertTrue(_sre.unicode_iscased(0x0130))

def test_not_literal(self):
self.assertEqual(re.search(r"\s([^a])", " b").group(1), "b")
self.assertEqual(re.search(r"\s([^a]*)", " bb").group(1), "bb")
Expand Down Expand Up @@ -1769,20 +1742,6 @@ def test_bug_6509(self):
pat = re.compile(b'..')
self.assertEqual(pat.sub(lambda m: b'bytes', b'a5'), b'bytes')

def test_dealloc(self):
# issue 3299: check for segfault in debug build
import _sre
# the overflow limit is different on wide and narrow builds and it
# depends on the definition of SRE_CODE (see sre.h).
# 2**128 should be big enough to overflow on both. For smaller values
# a RuntimeError is raised instead of OverflowError.
long_overflow = 2**128
self.assertRaises(TypeError, re.finditer, "a", {})
with self.assertRaises(OverflowError):
_sre.compile("abc", 0, [long_overflow], 0, {}, ())
with self.assertRaises(TypeError):
_sre.compile({}, 0, [], 0, [], [])

def test_search_dot_unicode(self):
self.assertTrue(re.search("123.*-", '123abc-'))
self.assertTrue(re.search("123.*-", '123\xe9-'))
Expand Down Expand Up @@ -1840,21 +1799,6 @@ def test_repeat_minmax_overflow(self):
self.assertRaises(OverflowError, re.compile, r".{%d,}?" % 2**128)
self.assertRaises(OverflowError, re.compile, r".{%d,%d}" % (2**129, 2**128))

@cpython_only
def test_repeat_minmax_overflow_maxrepeat(self):
try:
from _sre import MAXREPEAT
except ImportError:
self.skipTest('requires _sre.MAXREPEAT constant')
string = "x" * 100000
self.assertIsNone(re.match(r".{%d}" % (MAXREPEAT - 1), string))
self.assertEqual(re.match(r".{,%d}" % (MAXREPEAT - 1), string).span(),
(0, 100000))
self.assertIsNone(re.match(r".{%d,}?" % (MAXREPEAT - 1), string))
self.assertRaises(OverflowError, re.compile, r".{%d}" % MAXREPEAT)
self.assertRaises(OverflowError, re.compile, r".{,%d}" % MAXREPEAT)
self.assertRaises(OverflowError, re.compile, r".{%d,}?" % MAXREPEAT)

def test_backref_group_name_in_exception(self):
# Issue 17341: Poor error message when compiling invalid regex
self.checkPatternError('(?P=<foo>)',
Expand Down Expand Up @@ -2441,16 +2385,6 @@ def test_regression_gh94675(self):
p.terminate()
p.join()

def test_sre_template_invalid_group_index(self):
# see gh-106524
import _sre
with self.assertRaises(TypeError) as cm:
_sre.template("", ["", -1, ""])
self.assertIn("invalid template", str(cm.exception))
with self.assertRaises(TypeError) as cm:
_sre.template("", ["", (), ""])
self.assertIn("an integer is required", str(cm.exception))


def get_debug_out(pat):
with captured_stdout() as out:
Expand Down Expand Up @@ -2699,6 +2633,75 @@ def test_deprecated_modules(self):
self.assertTrue(hasattr(mod, attr))
del sys.modules[name]

@cpython_only
def test_case_helpers(self):
import _sre
for i in range(128):
c = chr(i)
lo = ord(c.lower())
self.assertEqual(_sre.ascii_tolower(i), lo)
self.assertEqual(_sre.unicode_tolower(i), lo)
iscased = c in string.ascii_letters
self.assertEqual(_sre.ascii_iscased(i), iscased)
self.assertEqual(_sre.unicode_iscased(i), iscased)

for i in list(range(128, 0x1000)) + [0x10400, 0x10428]:
c = chr(i)
self.assertEqual(_sre.ascii_tolower(i), i)
if i != 0x0130:
self.assertEqual(_sre.unicode_tolower(i), ord(c.lower()))
iscased = c != c.lower() or c != c.upper()
self.assertFalse(_sre.ascii_iscased(i))
self.assertEqual(_sre.unicode_iscased(i),
c != c.lower() or c != c.upper())

self.assertEqual(_sre.ascii_tolower(0x0130), 0x0130)
self.assertEqual(_sre.unicode_tolower(0x0130), ord('i'))
self.assertFalse(_sre.ascii_iscased(0x0130))
self.assertTrue(_sre.unicode_iscased(0x0130))

@cpython_only
def test_dealloc(self):
# issue 3299: check for segfault in debug build
import _sre
# the overflow limit is different on wide and narrow builds and it
# depends on the definition of SRE_CODE (see sre.h).
# 2**128 should be big enough to overflow on both. For smaller values
# a RuntimeError is raised instead of OverflowError.
long_overflow = 2**128
self.assertRaises(TypeError, re.finditer, "a", {})
with self.assertRaises(OverflowError):
_sre.compile("abc", 0, [long_overflow], 0, {}, ())
with self.assertRaises(TypeError):
_sre.compile({}, 0, [], 0, [], [])

@cpython_only
def test_repeat_minmax_overflow_maxrepeat(self):
try:
from _sre import MAXREPEAT
except ImportError:
self.skipTest('requires _sre.MAXREPEAT constant')
string = "x" * 100000
self.assertIsNone(re.match(r".{%d}" % (MAXREPEAT - 1), string))
self.assertEqual(re.match(r".{,%d}" % (MAXREPEAT - 1), string).span(),
(0, 100000))
self.assertIsNone(re.match(r".{%d,}?" % (MAXREPEAT - 1), string))
self.assertRaises(OverflowError, re.compile, r".{%d}" % MAXREPEAT)
self.assertRaises(OverflowError, re.compile, r".{,%d}" % MAXREPEAT)
self.assertRaises(OverflowError, re.compile, r".{%d,}?" % MAXREPEAT)

@cpython_only
def test_sre_template_invalid_group_index(self):
# see gh-106524
import _sre
with self.assertRaises(TypeError) as cm:
_sre.template("", ["", -1, ""])
self.assertIn("invalid template", str(cm.exception))
with self.assertRaises(TypeError) as cm:
_sre.template("", ["", (), ""])
self.assertIn("an integer is required", str(cm.exception))


class ExternalTests(unittest.TestCase):

def test_re_benchmarks(self):
Expand Down