def __init__(self, *args, **kwargs) -> None:
"""Initializes new visitor and saves all docstrings."""
super().__init__(*args, **kwargs)
self._docstrings = get_docstring_tokens(self.file_tokens)
def get_quotes_errors(self, file_contents):
tokens = [
Token(t) for t in tokenize.generate_tokens(
lambda L=iter(file_contents): next(L))
]
docstring_tokens = get_docstring_tokens(tokens)
for token in tokens:
if token.type != tokenize.STRING:
# ignore non strings
continue
# Remove any prefixes in strings like `u` from `u"foo"`
# DEV: `last_quote_char` is 1 character, even for multiline strings
# `"foo"` -> `"foo"`
# `b"foo"` -> `"foo"`
# `br"foo"` -> `"foo"`
# `b"""foo"""` -> `"""foo"""`
last_quote_char = token.string[-1]
first_quote_index = token.string.index(last_quote_char)
prefix = token.string[:first_quote_index].lower()
unprefixed_string = token.string[first_quote_index:]
# Determine if our string is multiline-based
# "foo"[0] * 3 = " * 3 = """
# "foo"[0:3] = "fo
# """foo"""[0:3] = """
is_docstring = token in docstring_tokens
is_multiline_string = unprefixed_string[
0] * 3 == unprefixed_string[0:3]
start_row, start_col = token.start
# If our string is a docstring
# DEV: Docstring quotes must come before multiline quotes as it can as a multiline quote
if is_docstring:
if self.config['good_docstring'] in unprefixed_string:
continue
yield {
'message': 'Q002 Remove bad quotes from docstring',
'line': start_row,
'col': start_col,
}
# Otherwise if our string is multiline
elif is_multiline_string:
# If our string is or containing a known good string, then ignore it
# (""")foo""" -> good (continue)
# '''foo(""")''' -> good (continue)
# (''')foo''' -> possibly bad
if self.config['good_multiline'] in unprefixed_string:
continue
# If our string ends with a known good ending, then ignore it
# '''foo("''') -> good (continue)
# Opposite, """foo"""", would break our parser (cannot handle """" ending)
if unprefixed_string.endswith(
self.config['good_multiline_ending']):
continue
# Output our error
yield {
'message': 'Q001 Remove bad quotes from multiline string',
'line': start_row,
'col': start_col,
}
# Otherwise (string is inline quote)
else:
# 'This is a string' -> Good
# 'This is a "string"' -> Good
# 'This is a \"string\"' -> Good
# 'This is a \'string\'' -> Bad (Q003) Escaped inner quotes
# '"This" is a \'string\'' -> Good Changing outer quotes would not avoid escaping
# "This is a string" -> Bad (Q000)
# "This is a 'string'" -> Good Avoids escaped inner quotes
# "This is a \"string\"" -> Bad (Q000)
# "\"This\" is a 'string'" -> Good
string_contents = unprefixed_string[1:-1]
# If string preferred type, check for escapes
if last_quote_char == self.config['good_single']:
if not self.config['avoid_escape'] or 'r' in prefix:
continue
if (self.config['good_single'] in string_contents and
not self.config['bad_single'] in string_contents):
yield {
'message':
'Q003 Change outer quotes to avoid escaping inner quotes',
'line': start_row,
'col': start_col,
}
if len(string_contents) < 1:
yield {
'message': 'Q000 Remove bad quotes',
'line': start_row,
'col': start_col,
}
continue
# If not preferred type, only allow use to avoid escapes.
if not self.config['good_single'] in string_contents:
if not (len(string_contents) <= 1 and re.match(
r"^'(?:.|\n|\t)?'$", unprefixed_string)):
yield {
'message': 'Q000 Remove bad quotes',
'line': start_row,
'col': start_col,
}
def get_quotes_errors(self, file_contents):
tokens = [
Token(t) for t in tokenize.generate_tokens(
lambda L=iter(file_contents): next(L))
]
docstring_tokens = get_docstring_tokens(tokens)
for token in tokens:
if token.type != tokenize.STRING:
# ignore non strings
continue
# Remove any prefixes in strings like `u` from `u"foo"`
# DEV: `last_quote_char` is 1 character, even for multiline strings
# `"foo"` -> `"foo"`
# `b"foo"` -> `"foo"`
# `br"foo"` -> `"foo"`
# `b"""foo"""` -> `"""foo"""`
last_quote_char = token.string[-1]
first_quote_index = token.string.index(last_quote_char)
unprefixed_string = token.string[first_quote_index:]
# Determine if our string is multiline-based
# "foo"[0] * 3 = " * 3 = """
# "foo"[0:3] = "fo
# """foo"""[0:3] = """
is_docstring = token in docstring_tokens
is_multiline_string = unprefixed_string[
0] * 3 == unprefixed_string[0:3]
start_row, start_col = token.start
# If our string is a docstring
# DEV: Docstring quotes must come before multiline quotes as it can as a multiline quote
if is_docstring:
if self.config['good_docstring'] in unprefixed_string:
continue
yield {
'message': 'Q002 Remove bad quotes from docstring',
'line': start_row,
'col': start_col,
}
# Otherwise if our string is multiline
elif is_multiline_string:
# If our string is or containing a known good string, then ignore it
# (""")foo""" -> good (continue)
# '''foo(""")''' -> good (continue)
# (''')foo''' -> possibly bad
if self.config['good_multiline'] in unprefixed_string:
continue
# Output our error
yield {
'message': 'Q001 Remove bad quotes from multiline string',
'line': start_row,
'col': start_col,
}
# Otherwise (string is inline quote)
else:
# If our string is a known good string, then ignore it
# (')foo' -> good (continue)
# "it(')s" -> good (continue)
# (")foo" -> possibly bad
if self.config['good_single'] in unprefixed_string:
continue
# Output our error
yield {
'message': 'Q000 Remove bad quotes',
'line': start_row,
'col': start_col,
}
def __init__(self, *args, **kwargs) -> None:
"""Creates line tracking for tokens."""
super().__init__(*args, **kwargs)
self._lines: TokenLines = defaultdict(list)
self._docstrings = get_docstring_tokens(self.file_tokens)
def __init__(self, logical_line: str, previous_logical: str, tokens: List[tokenize.TokenInfo]) -> None:
self.line = logical_line
self.tokens = tokens
prev_tokens = tokenize.tokenize(lambda L=iter([previous_logical.encode('utf-8')]): next(L))
self.docstring_tokens = get_docstring_tokens(prev_tokens, self.tokens)