third_party.pylibs.pylint.src/pylint/checkers/misc.py

100 lines
3.3 KiB
Python
Raw Normal View History

# Copyright (c) 2006, 2009-2013 LOGILAB S.A. (Paris, FRANCE) <contact@logilab.fr>
# Copyright (c) 2013-2014 Google, Inc.
# Copyright (c) 2014 Alexandru Coman <fcoman@bitdefender.com>
# Copyright (c) 2014-2016 Claudiu Popa <pcmanticore@gmail.com>
# Licensed under the GPL: https://www.gnu.org/licenses/old-licenses/gpl-2.0.html
# For details: https://github.com/PyCQA/pylint/blob/master/COPYING
2006-04-26 10:48:09 +00:00
"""Check source code is ascii only or has an encoding declaration (PEP 263)"""
# pylint: disable=W0511
2006-04-26 10:48:09 +00:00
2012-08-22 09:42:40 +00:00
import re
2006-04-26 10:48:09 +00:00
import six
2006-04-26 10:48:09 +00:00
from pylint.interfaces import IRawChecker
from pylint.checkers import BaseChecker
2010-04-16 15:52:38 +00:00
2006-04-26 10:48:09 +00:00
MSGS = {
'W0511': ('%s',
'fixme',
2006-04-26 10:48:09 +00:00
'Used when a warning note as FIXME or XXX is detected.'),
'W0512': ('Cannot decode using encoding "%s", unexpected byte at position %d',
'invalid-encoded-data',
'Used when a source line cannot be decoded using the specified '
'source file encoding.',
{'maxversion': (3, 0)}),
}
2006-04-26 10:48:09 +00:00
2006-04-26 10:48:09 +00:00
class EncodingChecker(BaseChecker):
2010-04-16 15:52:38 +00:00
"""checks for:
* warning notes in the code like FIXME, XXX
* encoding issues.
2006-04-26 10:48:09 +00:00
"""
__implements__ = IRawChecker
# configuration section name
name = 'miscellaneous'
msgs = MSGS
options = (('notes',
{'type': 'csv', 'metavar': '<comma separated values>',
'default': ('FIXME', 'XXX', 'TODO'),
'help': ('List of note tags to take in consideration, '
'separated by a comma.')}),)
2006-04-26 10:48:09 +00:00
def _check_note(self, notes, lineno, line):
# First, simply check if the notes are in the line at all. This is an
# optimisation to prevent using the regular expression on every line,
# but rather only on lines which may actually contain one of the notes.
# This prevents a pathological problem with lines that are hundreds
# of thousands of characters long.
for note in self.config.notes:
if note in line:
break
else:
return
2014-07-09 08:06:37 +00:00
match = notes.search(line)
if not match:
return
self.add_message('fixme', args=line[match.start(1):-1], line=lineno)
def _check_encoding(self, lineno, line, file_encoding):
try:
return six.text_type(line, file_encoding)
except UnicodeDecodeError as ex:
self.add_message('invalid-encoded-data', line=lineno,
2013-07-31 07:05:01 +00:00
args=(file_encoding, ex.args[2]))
def process_module(self, module):
"""inspect the source file to find encoding problem or fixmes like
2006-04-26 10:48:09 +00:00
notes
"""
if self.config.notes:
notes = re.compile(
r'.*?#\s*(%s)(:*\s*.*)' % "|".join(self.config.notes))
else:
notes = None
if module.file_encoding:
encoding = module.file_encoding
else:
encoding = 'ascii'
with module.stream() as stream:
for lineno, line in enumerate(stream):
line = self._check_encoding(lineno + 1, line, encoding)
if line is not None and notes:
self._check_note(notes, lineno + 1, line)
2010-04-16 15:52:38 +00:00
2006-04-26 10:48:09 +00:00
def register(linter):
"""required method to auto register this checker"""
linter.register_checker(EncodingChecker(linter))