kallithea Changeset - 0f69b5c35b2b

Changeset - 0f69b5c35b2b

Parent rev.

Child rev.

[Not reviewed]

default

0 2 0

Mads Kiilerich - 6 years ago 2019-12-27 23:30:56
mads@kiilerich.com

Grafted from: 57ed7fc231aa

lib: introduce string conversion functions for ASCII without further encoding concerns

Avoid the trial-and-error and vagueness of the "safe" functions.

This should replace some use of safe_unicode and safe_str. It will mostly be a
noop in py2 but will be crucial in py3.

2 files changed with 57 insertions and 1 deletions:

kallithea/lib/utils2.py

kallithea/lib/vcs/utils/__init__.py

0 comments (0 inline, 0 general)

kallithea/lib/utils2.py

➞

Show inline comments

 # -*- coding: utf-8 -*-
 # This program is free software: you can redistribute it and/or modify
 # it under the terms of the GNU General Public License as published by
 # the Free Software Foundation, either version 3 of the License, or
 # (at your option) any later version.
+#
 # This program is distributed in the hope that it will be useful,
 # but WITHOUT ANY WARRANTY; without even the implied warranty of
 # MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
 # GNU General Public License for more details.
+#
 # You should have received a copy of the GNU General Public License
 # along with this program.  If not, see <http://www.gnu.org/licenses/>.
 """
 kallithea.lib.utils2
 ~~~~~~~~~~~~~~~~~~~~
 Some simple helper functions.
 Note: all these functions should be independent of Kallithea classes, i.e.
 models, controllers, etc.  to prevent import cycles.
 This file was forked by the Kallithea project in July 2014.
 Original author and date, and relevant copyright and licensing information is below:
 :created_on: Jan 5, 2011
 :author: marcink
 :copyright: (c) 2013 RhodeCode GmbH, and others.
 :license: GPLv3, see LICENSE.md for more details.
 """
 from __future__ import print_function
 import binascii
 import datetime
 import os
 import pwd
 import re
 import time
 import urllib
 import urlobject
 from tg.i18n import ugettext as _
 from tg.i18n import ungettext
 from webhelpers2.text import collapse, remove_formatting, strip_tags
 from kallithea.lib.compat import json
 from kallithea.lib.vcs.utils import safe_bytes, safe_str, safe_unicode  # re-export
+from kallithea.lib.vcs.utils import ascii_bytes, ascii_str, safe_bytes, safe_str, safe_unicode  # re-export
 from kallithea.lib.vcs.utils.lazy import LazyProperty
 def str2bool(_str):
     """
     returns True/False value from given string, it tries to translate the
     string into boolean
     :param _str: string value to translate into boolean
     :rtype: boolean
     :returns: boolean from given string
     """
     if _str is None:
         return False
     if _str in (True, False):
         return _str
     _str = str(_str).strip().lower()
     return _str in ('t', 'true', 'y', 'yes', 'on', '1')
 def aslist(obj, sep=None, strip=True):
     """
     Returns given string separated by sep as list
     :param obj:
     :param sep:
     :param strip:
     """
     if isinstance(obj, (basestring)):
         lst = obj.split(sep)
         if strip:
             lst = [v.strip() for v in lst]
         return lst
     elif isinstance(obj, (list, tuple)):
         return obj
     elif obj is None:
         return []
     else:
         return [obj]
 def convert_line_endings(line, mode):
     """
     Converts a given line  "line end" according to given mode
     Available modes are::
 - Unix
 - Mac

kallithea/lib/vcs/utils/__init__.py

➞

Show inline comments

 # -*- coding: utf-8 -*-
 """
 This module provides some useful tools for ``vcs`` like annotate/diff html
 output. It also includes some internal helpers.
 """
 import datetime
 import re
 import time
 def makedate():
     lt = time.localtime()
     if lt[8] == 1 and time.daylight:
         tz = time.altzone
     else:
         tz = time.timezone
     return time.mktime(lt), tz
 def aslist(obj, sep=None, strip=True):
     """
     Returns given string separated by sep as list
     :param obj:
     :param sep:
     :param strip:
     """
     if isinstance(obj, basestring):
         lst = obj.split(sep)
         if strip:
             lst = [v.strip() for v in lst]
         return lst
     elif isinstance(obj, (list, tuple)):
         return obj
     elif obj is None:
         return []
     else:
         return [obj]
 def date_fromtimestamp(unixts, tzoffset=0):
     """
     Makes a local datetime object out of unix timestamp
     :param unixts:
     :param tzoffset:
     """
@@ @@ -76,90 +78,144 @@ def safe_unicode(s): @@
     if isinstance(s, unicode):
         return s
     if not isinstance(s, bytes):  # use __str__ / __unicode__ and don't expect UnicodeDecodeError
         return unicode(s)
     from kallithea.lib.vcs.conf import settings
     for enc in settings.DEFAULT_ENCODINGS:
         try:
             return unicode(s, enc)
         except UnicodeDecodeError:
             pass
     try:
         import chardet
         encoding = chardet.detect(s)['encoding']
         if encoding is not None:
             return s.decode(encoding)
     except (ImportError, UnicodeDecodeError):
         pass
     return unicode(s, settings.DEFAULT_ENCODINGS[0], 'replace')
 def safe_bytes(s):
     """
     Safe bytes function. Use a few tricks to turn s into bytes string:
     In case of UnicodeEncodeError with configured default encodings, fall back
     to first configured encoding with errors replaced.
     """
     if isinstance(s, bytes):
         return s
     assert isinstance(s, unicode), repr(s)  # bytes cannot coerse with __str__ or handle None or int
     from kallithea.lib.vcs.conf import settings
     for enc in settings.DEFAULT_ENCODINGS:
         try:
             return s.encode(enc)
         except UnicodeEncodeError:
             pass
     return s.encode(settings.DEFAULT_ENCODINGS[0], 'replace')
 safe_str = safe_bytes  # safe_str is deprecated - it will be redefined when changing to py3
 def ascii_bytes(s):
     """
     Simple conversion from unicode/str to bytes, *assuming* all codepoints are
 -bit and it thus is pure ASCII.
     Will fail badly with UnicodeError on invalid input.
     This should be used where enocding and "safe" ambiguity should be avoided.
     Where strings already have been encoded in other ways but still are unicode
     string - for example to hex, base64, json, urlencoding, or are known to be
     identifiers.
     >>> ascii_bytes('a')
     'a'
     >>> ascii_bytes(u'a')
     'a'
     >>> ascii_bytes('å')
     Traceback (most recent call last):
     UnicodeDecodeError: 'ascii' codec can't decode byte 0xc3 in position 0: ordinal not in range(128)
     >>> ascii_bytes(u'å')
     Traceback (most recent call last):
     UnicodeEncodeError: 'ascii' codec can't encode characters in position 0-1: ordinal not in range(128)
     """
     assert isinstance(s, (unicode, str)), repr(s)
     return s.encode('ascii')
 def ascii_str(s):
     r"""
     Simple conversion from bytes to str, *assuming* all codepoints are
 -bit and it thus is pure ASCII.
     Will fail badly with UnicodeError on invalid input.
     This should be used where enocding and "safe" ambiguity should be avoided.
     Where strings are encoded but also in other ways are known to be ASCII, and
     where a unicode string is wanted without caring about encoding. For example
     to hex, base64, urlencoding, or are known to be identifiers.
     >>> ascii_str('a')
     'a'
     >>> ascii_str(u'a')
     Traceback (most recent call last):
     AssertionError: u'a'
     >>> ascii_str('å')
     Traceback (most recent call last):
     UnicodeDecodeError: 'ascii' codec can't decode byte 0xc3 in position 0: ordinal not in range(128)
     >>> ascii_str(u'å')
     Traceback (most recent call last):
     AssertionError: u'\xc3\xa5'
     """
     assert isinstance(s, bytes), repr(s)
     # Note: we use "encode", even though we really *should* use "decode". But
     # we are in py2 and don't want py2, and encode is doing what we need for the
     # ascii subset.
     return s.encode('ascii')
 # Regex taken from http://www.regular-expressions.info/email.html
 email_re = re.compile(
     r"""[a-z0-9!#$%&'*+/=?^_`{|}~-]+(?:\.[a-z0-9!#$%&'*+/=?^_`{|}~-]+)*@"""
     r"""(?:[a-z0-9](?:[a-z0-9-]*[a-z0-9])?\.)+[a-z0-9](?:[a-z0-9-]*[a-z0-9])?""",
     re.IGNORECASE)
 def author_email(author):
     """
     Returns email address of given author string.
     If author contains <> brackets, only look inside that.
     If any RFC valid email address is found, return that.
     Else, return empty string.
     """
     if not author:
         return ''
     l = author.find('<') + 1
     if l != 0:
         r = author.find('>', l)
         if r != -1:
             author = author[l:r]
     m = email_re.search(author)
     if m is None:
         return ''
     return safe_str(m.group(0))
 def author_name(author):
     """
     get name of author, or else username.
     It'll try to find an email in the author string and just cut it off
     to get the username
     """
     if not author:
         return ''
     if '@' not in author:
         return author
     return author.replace(author_email(author), '').replace('<', '') \
         .replace('>', '').strip()

0 comments (0 inline, 0 general)