Skip to content

[Unicode] support startswith with args, start and end. #8557

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
sklam merged 11 commits into from
Nov 7, 2022
Merged

[Unicode] support startswith with args, start and end. #8557

Show file tree
Hide file tree
Changes from 1 commit
Commits
Show all changes
11 commits
Select commit Hold shift + click to select a range
5f2cc16
support startswith with args, start and end.
Nov 1, 2022
5e65597
Update numba/cpython/unicode.py
Nov 2, 2022
7fa462e
Update numba/cpython/unicode.py
Nov 2, 2022
4e5960d
add testcases to raise exceptions.
Nov 2, 2022
76b58c2
address comments: replace argument name and move out the exception te…
Nov 3, 2022
9742576
address reviews: rename, make error msg more precise, check tuple pre…
Nov 4, 2022
33aaa1a
Update numba/cpython/unicode.py
Nov 7, 2022
ac5ba28
Update numba/cpython/unicode.py
Nov 7, 2022
ae74063
Update numba/cpython/unicode.py
Nov 7, 2022
2216a7e
Update numba/cpython/unicode.py
Nov 7, 2022
696dd8d
address reviews: clear handling with `prefix`.
Nov 7, 2022
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
62 changes: 50 additions & 12 deletions numba/cpython/unicode.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -813,18 +813,6 @@ def impl(data, sep):
return impl


@overload_method(types.UnicodeType, 'startswith')
def unicode_startswith(a, b):
if isinstance(b, types.UnicodeType):
def startswith_impl(a, b):
return _cmp_region(a, 0, b, 0, len(b)) == 0
return startswith_impl
if isinstance(b, types.UnicodeCharSeq):
def startswith_impl(a, b):
return a.startswith(str(b))
return startswith_impl


# https://github.com/python/cpython/blob/201c8f79450628241574fba940e08107178dc3a5/Objects/unicodeobject.c#L9342-L9354 # noqa: E501
@register_jitable
def _adjust_indices(length, start, end):
Expand All @@ -842,6 +830,56 @@ def _adjust_indices(length, start, end):
return start, end


@overload_method(types.UnicodeType, 'startswith')
def unicode_startswith(s, substr, start=None, end=None):
Comment thread
dlee992 marked this conversation as resolved.
Resolved
if not (start is None or isinstance(start, (types.Omitted,
Comment thread
dlee992 marked this conversation as resolved.
Resolved
types.Integer,
types.NoneType))):
raise TypingError('The arg must be a Integer or None')
Comment thread
dlee992 marked this conversation as resolved.
Resolved

if not (end is None or isinstance(end, (types.Omitted,
types.Integer,
types.NoneType))):
raise TypingError('The arg must be a Integer or None')
Comment thread
dlee992 marked this conversation as resolved.
Resolved

if isinstance(substr, (types.Tuple, types.UniTuple)):
def startswith_tuple_impl(s, substr, start=None, end=None):
for item in substr:
if s.startswith(item, start, end):
return True
return False

return startswith_tuple_impl

if isinstance(substr, types.UnicodeCharSeq):
def startswith_char_seq_impl(s, substr, start=None, end=None):
return s.startswith(str(substr), start, end)

return startswith_char_seq_impl

if isinstance(substr, types.UnicodeType):
def startswith_unicode_impl(s, substr, start=None, end=None):
length = len(s)
sub_length = len(substr)
if start is None:
start = 0
if end is None:
end = length

start, end = _adjust_indices(length, start, end)
if end - start < sub_length:
return False

if sub_length == 0:
return True

s = s[start:end]

return _cmp_region(s, 0, substr, 0, sub_length) == 0

return startswith_unicode_impl

Comment thread
dlee992 marked this conversation as resolved.
Resolved

@overload_method(types.UnicodeType, 'endswith')
def unicode_endswith(s, substr, start=None, end=None):
if not (start is None or isinstance(start, (types.Omitted,
Expand Down
102 changes: 89 additions & 13 deletions numba/tests/test_unicode.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -177,6 +177,14 @@ def expandtabs_with_tabsize_kwarg_usecase(s, tabsize):
return s.expandtabs(tabsize=tabsize)


def startswith_with_start_only_usecase(x, y, start):
return x.startswith(y, start)


def startswith_with_start_end_usecase(x, y, start, end):
return x.startswith(y, start, end)


def endswith_with_start_only_usecase(x, y, start):
return x.endswith(y, start)

Expand Down Expand Up @@ -551,23 +559,91 @@ def test_expandtabs_exception_noninteger_tabsize(self):
msg = '"tabsize" must be {}, not float'.format(accepted_types)
self.assertIn(msg, str(raises.exception))

def test_startswith(self, flags=no_pyobj_flags):
def test_startswith_default(self):
pyfunc = startswith_usecase
cfunc = njit(pyfunc)
for a in UNICODE_EXAMPLES:
for b in ['', 'x', a[:-2], a[3:], a, a + a]:
self.assertEqual(pyfunc(a, b),
cfunc(a, b),
'%s, %s' % (a, b))

def test_endswith(self, flags=no_pyobj_flags):
pyfunc = endswith_usecase
cpython_str = ['hello', 'helloworld', '']
cpython_subs = [
'he', 'hello', 'helloworld', 'ello',
'', 'lowo', 'lo', 'he', 'lo', 'o',
]
extra_subs = ['hellohellohello', ' ']
for s in cpython_str + UNICODE_EXAMPLES:
default_subs = ['', 'x', s[:-2], s[3:], s, s + s]
for sub_str in cpython_subs + default_subs + extra_subs:
self.assertEqual(pyfunc(s, sub_str), cfunc(s, sub_str))

def test_startswith_with_start(self):
pyfunc = startswith_with_start_only_usecase
cfunc = njit(pyfunc)
for a in UNICODE_EXAMPLES:
for b in ['', 'x', a[:-2], a[3:], a, a + a]:
self.assertEqual(pyfunc(a, b),
cfunc(a, b),
'%s, %s' % (a, b))

cpython_str = ['hello', 'helloworld', '']
cpython_subs = [
'he', 'hello', 'helloworld', 'ello',
'', 'lowo', 'lo', 'he', 'lo', 'o',
]
extra_subs = ['hellohellohello', ' ']
for s in cpython_str + UNICODE_EXAMPLES:
default_subs = ['', 'x', s[:-2], s[3:], s, s + s]
for sub_str in cpython_subs + default_subs + extra_subs:
for start in list(range(-20, 20)) + [None]:
self.assertEqual(pyfunc(s, sub_str, start),
cfunc(s, sub_str, start))

def test_startswith_with_start_end(self):
pyfunc = startswith_with_start_end_usecase
cfunc = njit(pyfunc)

cpython_str = ['hello', 'helloworld', '']
cpython_subs = [
'he', 'hello', 'helloworld', 'ello',
'', 'lowo', 'lo', 'he', 'lo', 'o',
]
extra_subs = ['hellohellohello', ' ']
for s in cpython_str + UNICODE_EXAMPLES:
default_subs = ['', 'x', s[:-2], s[3:], s, s + s]
for sub_str in cpython_subs + default_subs + extra_subs:
for start in list(range(-20, 20)) + [None]:
for end in list(range(-20, 20)) + [None]:
self.assertEqual(pyfunc(s, sub_str, start, end),
cfunc(s, sub_str, start, end))

def test_startswith_tuple(self):
pyfunc = startswith_usecase
cfunc = njit(pyfunc)

cpython_str = ['hello', 'helloworld', '']
cpython_subs = [
'he', 'hello', 'helloworld', 'ello',
'', 'lowo', 'lo', 'he', 'lo', 'o',
]
extra_subs = ['hellohellohello', ' ']
for s in cpython_str + UNICODE_EXAMPLES:
default_subs = ['', 'x', s[:-2], s[3:], s, s + s]
for sub_str in cpython_subs + default_subs + extra_subs:
tuple_subs = (sub_str, 'lo')
self.assertEqual(pyfunc(s, tuple_subs),
cfunc(s, tuple_subs))

def test_startswith_tuple_args(self):
pyfunc = startswith_with_start_end_usecase
cfunc = njit(pyfunc)

cpython_str = ['hello', 'helloworld', '']
cpython_subs = [
'he', 'hello', 'helloworld', 'ello',
'', 'lowo', 'lo', 'he', 'lo', 'o',
]
extra_subs = ['hellohellohello', ' ']
for s in cpython_str + UNICODE_EXAMPLES:
default_subs = ['', 'x', s[:-2], s[3:], s, s + s]
for sub_str in cpython_subs + default_subs + extra_subs:
for start in list(range(-20, 20)) + [None]:
for end in list(range(-20, 20)) + [None]:
tuple_subs = (sub_str, 'lo')
self.assertEqual(pyfunc(s, tuple_subs, start, end),
cfunc(s, tuple_subs, start, end))

def test_endswith_default(self):
pyfunc = endswith_usecase
Expand Down