def _remove_u_prefix(token: Token) -> Token:
prefix, rest = parse_string_literal(token.src)
if 'u' not in prefix.lower():
return token
else:
new_prefix = prefix.replace('u', '').replace('U', '')
return token._replace(src=new_prefix + rest)
def _fix_escape_sequences(token: Token) -> Token:
prefix, rest = parse_string_literal(token.src)
actual_prefix = prefix.lower()
if 'r' in actual_prefix or '\\' not in rest:
return token
is_bytestring = 'b' in actual_prefix
def _is_valid_escape(match: Match[str]) -> bool:
c = match.group()[1]
return (
c in ESCAPE_STARTS or
(not is_bytestring and c in 'uU') or
(
not is_bytestring and
c == 'N' and
bool(NAMED_ESCAPE_NAME.match(rest, match.end()))
)
)
has_valid_escapes = False
has_invalid_escapes = False
for match in ESCAPE_RE.finditer(rest):
if _is_valid_escape(match):
has_valid_escapes = True
else:
has_invalid_escapes = True
def cb(match: Match[str]) -> str:
matched = match.group()
if _is_valid_escape(match):
return matched
else:
return fr'\{matched}'
if has_invalid_escapes and (has_valid_escapes or 'u' in actual_prefix):
return token._replace(src=prefix + ESCAPE_RE.sub(cb, rest))
elif has_invalid_escapes and not has_valid_escapes:
return token._replace(src=prefix + 'r' + rest)
else:
return token
def _fix_ur_literals(token: Token) -> Token:
prefix, rest = parse_string_literal(token.src)
if prefix.lower() != 'ur':
return token
else:
def cb(match: Match[str]) -> str:
escape = match.group()
if escape[1].lower() == 'u':
return escape
else:
return '\\' + match.group()
rest = ESCAPE_RE.sub(cb, rest)
prefix = prefix.replace('r', '').replace('R', '')
return token._replace(src=prefix + rest)