def test_leaves_unmatched_punctuation_alone(self):
f = normalizers.StripMatchedPunctuationNormalizer()
# quotes
# " character
self.assertEqual(f('"foo'), '"foo')
self.assertEqual(f('foo"'), 'foo"')
self.assertEqual(f('"foo bar'), '"foo bar')
self.assertEqual(f('foo bar"'), 'foo bar"')
# ' character
self.assertEqual(f("'foo"), "'foo")
self.assertEqual(f("foo'"), "foo'")
self.assertEqual(f("'foo bar"), "'foo bar")
self.assertEqual(f("foo bar'"), "foo bar'")
# brackets
# () characters
self.assertEqual(f('(foo'), '(foo')
self.assertEqual(f('foo)'), 'foo)')
self.assertEqual(f('(foo bar'), '(foo bar')
self.assertEqual(f('foo bar)'), 'foo bar)')
# [] characters
self.assertEqual(f('[foo'), '[foo')
self.assertEqual(f('foo]'), 'foo]')
self.assertEqual(f('[foo bar'), '[foo bar')
self.assertEqual(f('foo bar]'), 'foo bar]')
# {} characters
self.assertEqual(f('{foo'), '{foo')
self.assertEqual(f('foo}'), 'foo}')
self.assertEqual(f('{foo bar'), '{foo bar')
self.assertEqual(f('foo bar}'), 'foo bar}')
def test_leaves_leading_and_trailing_punctuation_alone(self):
f = normalizers.StripMatchedPunctuationNormalizer()
# leading
self.assertEqual(f(':foo'), ':foo')
self.assertEqual(f(';foo bar'), ';foo bar')
# trailing
self.assertEqual(f('foo.'), 'foo.')
self.assertEqual(f('foo bar!'), 'foo bar!')
# leading and trailing
self.assertEqual(f('.foo,'), '.foo,')
self.assertEqual(f('- foo bar!'), '- foo bar!')
def test_leaves_matched_punctuation_not_wrapping_text_alone(self):
f = normalizers.StripMatchedPunctuationNormalizer()
# quotes
# " character
self.assertEqual(f('"foo" bar'), '"foo" bar')
self.assertEqual(f('foo "bar"'), 'foo "bar"')
# ' character
self.assertEqual(f("'foo' bar"), "'foo' bar")
self.assertEqual(f("foo 'bar'"), "foo 'bar'")
# brackets
# () characters
self.assertEqual(f('(foo) bar'), '(foo) bar')
self.assertEqual(f('foo (bar)'), 'foo (bar)')
# [] characters
self.assertEqual(f('[foo] bar'), '[foo] bar')
self.assertEqual(f('foo [bar]'), 'foo [bar]')
# {} characters
self.assertEqual(f('{foo} bar'), '{foo} bar')
self.assertEqual(f('foo {bar}'), 'foo {bar}')
def test_leaves_normal_text_alone(self):
f = normalizers.StripMatchedPunctuationNormalizer()
self.assertEqual(f('foo'), 'foo')
self.assertEqual(f('foo bar'), 'foo bar')