def test_strip_tags_in_scripts(input, expected):
r = Richtypo()
r.text = input
r.strip_tags()
expected = re.sub(r'~(\d+)~', '~([^~]+)~', expected) # it is a regex now
expected = re.compile(expected)
assert expected.match(r.text)
def test_unicode_with_ascii_only_characters():
"""
If this test does not fail, and the above test does,
then i've messed up with unicode
"""
r = Richtypo(ruleset='en-lite')
text = u'2 to 3'
assert r.richtypo(text) == u'2%sto%s3' % (NBSP, NBSP)
def test_punctuation(input, expected):
"""
Test the complex rule for spaces after punctuation marks:
the 'generic:punctuation_spaces' which should also be followed
by 'generic:spaces_eol'
"""
r = Richtypo()
assert r.richtypo(input) == expected
def test_build_rule_chain(rules):
r = Richtypo()
r.available_rules = {'b': Rule(pattern='b', replacement='d')}
with patch('richtypo.Richtypo._get_ruleset_rules') as rules:
rules.return_value = ['b', ABRule]
r.build_rule_chain('generic')
assert len(r.rules) == 2
def test_ruleset_input_param(get_ruleset):
get_ruleset.return_value = [
ABRule,
]
r = Richtypo(ruleset='generic') # this ruleset realy should exist
assert len(r.rules) == 1
assert ABRule in r.rules
def test_restore_tags(input):
r = Richtypo()
r.text = input
r.strip_tags()
r.restore_tags()
assert r.text == input
def test_loading_generic_ruledef(rules):
"""
If not specified, all rules should be taken from the generic ruledef (rules/generic.yaml)
"""
rules.return_value = ['one', 'two']
with patch('richtypo.Richtypo.build_rule_chain'
): # dont actualy build rule chain
with patch('richtypo.Richtypo._load_rules_from_file') as loader:
Richtypo()
assert loader.call_count == 1
loader.assert_called_with('generic')
def test_loading_ruledef_only_once(get_ruleset):
"""
Check if ruledefs are loaded only once
"""
get_ruleset.return_value = [
'ru:test', 'tst-ruledef:test', 'ru:test1', 'ru:test2'
]
with patch('richtypo.Richtypo.build_rule_chain'
): # dont actualy build rule chain
with patch('richtypo.Richtypo._load_rules_from_file') as loader:
Richtypo()
assert loader.call_count == 2 #
loader.assert_any_call('ru')
loader.assert_any_call('tst-ruledef')
def test_rule_applying():
r1 = Rule(pattern='b{2}', replacement='cc')
r2 = Rule(pattern='c{2}', replacement='dd')
r = Richtypo()
r.rules = [r1, r2]
r.text = 'abb'
r.apply_rule_chain()
assert r.text == 'add'
def test_loading_specified_ruledef(get_ruleset):
"""
Check for loading specified ruledef
ruledefs are specified when constructing a ruleset, like this:
ruledef: rulename
"""
get_ruleset.return_value = ['ru: test']
with patch('richtypo.Richtypo.build_rule_chain'
): # dont actualy build rule chain
with patch('richtypo.Richtypo._load_rules_from_file') as loader:
Richtypo()
assert loader.call_count == 1
loader.assert_called_with('ru')
def test_rule_order():
"""
Check if rules are applyied in order they are supplied
"""
r1 = Rule(pattern='b{2}', replacement='cc')
r2 = Rule(pattern='c{2}', replacement='dd')
r = Richtypo()
r.rules = [r2, r1] # r2 works only after r1
r.text = 'abb'
r.apply_rule_chain()
assert r.text == 'acc' # so the text should not be changed
def test_strip_tags_when_input_already_contains_richtypo_marks(input, expected):
r = Richtypo()
r.text = input
r.strip_tags()
r.restore_tags()
assert r.text == expected
def test_get_rule_from_predefined_rules():
r = Richtypo()
rule = r._get_rule(ABRule)
assert rule == ABRule
def test_get_rule_from_available():
r = Richtypo()
r.available_rules = {'b': Rule(pattern='b', replacement='d')}
rule = r._get_rule('b')
assert rule.pattern == 'b'
def test_rule_loading_by_default():
r = Richtypo()
assert len(r.available_rules.keys()) >= 1
def test_russian_simple():
r = Richtypo(ruleset='ru-lite')
text = u'Из-под топота копыт'
assert r.richtypo(text) == u'Из-под%sтопота копыт' % NBSP