Merge "Fixes Error when contain Non-ascii in template"

This commit is contained in:
Jenkins 2013-11-18 05:19:11 +00:00 committed by Gerrit Code Review
commit 62252bd586
4 changed files with 227 additions and 1 deletions

View File

@ -17,6 +17,8 @@ import urllib
import urlparse
import collections
from heat.openstack.common import strutils
class HeatIdentifier(collections.Mapping):
@ -117,7 +119,8 @@ class HeatIdentifier(collections.Mapping):
'''
return 'stacks/%s/%s%s' % (urllib.quote(self.stack_name, ''),
urllib.quote(self.stack_id, ''),
urllib.quote(self.path))
urllib.quote(strutils.safe_encode(
self.path)))
def _path_components(self):
'''Return a list of the path components.'''

View File

@ -0,0 +1,218 @@
# vim: tabstop=4 shiftwidth=4 softtabstop=4
# Copyright 2011 OpenStack Foundation.
# All Rights Reserved.
#
# Licensed under the Apache License, Version 2.0 (the "License"); you may
# not use this file except in compliance with the License. You may obtain
# a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS, WITHOUT
# WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the
# License for the specific language governing permissions and limitations
# under the License.
"""
System-level utilities and helper functions.
"""
import re
import sys
import unicodedata
import six
from heat.openstack.common.gettextutils import _ # noqa
# Used for looking up extensions of text
# to their 'multiplied' byte amount
BYTE_MULTIPLIERS = {
'': 1,
't': 1024 ** 4,
'g': 1024 ** 3,
'm': 1024 ** 2,
'k': 1024,
}
BYTE_REGEX = re.compile(r'(^-?\d+)(\D*)')
TRUE_STRINGS = ('1', 't', 'true', 'on', 'y', 'yes')
FALSE_STRINGS = ('0', 'f', 'false', 'off', 'n', 'no')
SLUGIFY_STRIP_RE = re.compile(r"[^\w\s-]")
SLUGIFY_HYPHENATE_RE = re.compile(r"[-\s]+")
def int_from_bool_as_string(subject):
"""Interpret a string as a boolean and return either 1 or 0.
Any string value in:
('True', 'true', 'On', 'on', '1')
is interpreted as a boolean True.
Useful for JSON-decoded stuff and config file parsing
"""
return bool_from_string(subject) and 1 or 0
def bool_from_string(subject, strict=False):
"""Interpret a string as a boolean.
A case-insensitive match is performed such that strings matching 't',
'true', 'on', 'y', 'yes', or '1' are considered True and, when
`strict=False`, anything else is considered False.
Useful for JSON-decoded stuff and config file parsing.
If `strict=True`, unrecognized values, including None, will raise a
ValueError which is useful when parsing values passed in from an API call.
Strings yielding False are 'f', 'false', 'off', 'n', 'no', or '0'.
"""
if not isinstance(subject, six.string_types):
subject = str(subject)
lowered = subject.strip().lower()
if lowered in TRUE_STRINGS:
return True
elif lowered in FALSE_STRINGS:
return False
elif strict:
acceptable = ', '.join(
"'%s'" % s for s in sorted(TRUE_STRINGS + FALSE_STRINGS))
msg = _("Unrecognized value '%(val)s', acceptable values are:"
" %(acceptable)s") % {'val': subject,
'acceptable': acceptable}
raise ValueError(msg)
else:
return False
def safe_decode(text, incoming=None, errors='strict'):
"""Decodes incoming str using `incoming` if they're not already unicode.
:param incoming: Text's current encoding
:param errors: Errors handling policy. See here for valid
values http://docs.python.org/2/library/codecs.html
:returns: text or a unicode `incoming` encoded
representation of it.
:raises TypeError: If text is not an instance of str
"""
if not isinstance(text, six.string_types):
raise TypeError("%s can't be decoded" % type(text))
if isinstance(text, six.text_type):
return text
if not incoming:
incoming = (sys.stdin.encoding or
sys.getdefaultencoding())
try:
return text.decode(incoming, errors)
except UnicodeDecodeError:
# Note(flaper87) If we get here, it means that
# sys.stdin.encoding / sys.getdefaultencoding
# didn't return a suitable encoding to decode
# text. This happens mostly when global LANG
# var is not set correctly and there's no
# default encoding. In this case, most likely
# python will use ASCII or ANSI encoders as
# default encodings but they won't be capable
# of decoding non-ASCII characters.
#
# Also, UTF-8 is being used since it's an ASCII
# extension.
return text.decode('utf-8', errors)
def safe_encode(text, incoming=None,
encoding='utf-8', errors='strict'):
"""Encodes incoming str/unicode using `encoding`.
If incoming is not specified, text is expected to be encoded with
current python's default encoding. (`sys.getdefaultencoding`)
:param incoming: Text's current encoding
:param encoding: Expected encoding for text (Default UTF-8)
:param errors: Errors handling policy. See here for valid
values http://docs.python.org/2/library/codecs.html
:returns: text or a bytestring `encoding` encoded
representation of it.
:raises TypeError: If text is not an instance of str
"""
if not isinstance(text, six.string_types):
raise TypeError("%s can't be encoded" % type(text))
if not incoming:
incoming = (sys.stdin.encoding or
sys.getdefaultencoding())
if isinstance(text, six.text_type):
return text.encode(encoding, errors)
elif text and encoding != incoming:
# Decode text before encoding it with `encoding`
text = safe_decode(text, incoming, errors)
return text.encode(encoding, errors)
return text
def to_bytes(text, default=0):
"""Converts a string into an integer of bytes.
Looks at the last characters of the text to determine
what conversion is needed to turn the input text into a byte number.
Supports "B, K(B), M(B), G(B), and T(B)". (case insensitive)
:param text: String input for bytes size conversion.
:param default: Default return value when text is blank.
"""
match = BYTE_REGEX.search(text)
if match:
magnitude = int(match.group(1))
mult_key_org = match.group(2)
if not mult_key_org:
return magnitude
elif text:
msg = _('Invalid string format: %s') % text
raise TypeError(msg)
else:
return default
mult_key = mult_key_org.lower().replace('b', '', 1)
multiplier = BYTE_MULTIPLIERS.get(mult_key)
if multiplier is None:
msg = _('Unknown byte multiplier: %s') % mult_key_org
raise TypeError(msg)
return magnitude * multiplier
def to_slug(value, incoming=None, errors="strict"):
"""Normalize string.
Convert to lowercase, remove non-word characters, and convert spaces
to hyphens.
Inspired by Django's `slugify` filter.
:param value: Text to slugify
:param incoming: Text's current encoding
:param errors: Errors handling policy. See here for valid
values http://docs.python.org/2/library/codecs.html
:returns: slugified unicode representation of `value`
:raises TypeError: If text is not an instance of str
"""
value = safe_decode(value, incoming, errors)
# NOTE(aababilov): no need to use safe_(encode|decode) here:
# encodings are always "ascii", error handling is always "ignore"
# and types are always known (first: unicode; second: str)
value = unicodedata.normalize("NFKD", value).encode(
"ascii", "ignore").decode("ascii")
value = SLUGIFY_STRIP_RE.sub("", value).strip().lower()
return SLUGIFY_HYPHENATE_RE.sub("-", value)

View File

@ -227,6 +227,10 @@ class IdentifierTest(testtools.TestCase):
hi = identifier.HeatIdentifier('t', 's', 'i')
self.assertEqual(hi.url_path(), 't/stacks/s/i')
def test_url_path_with_unicode_path(self):
hi = identifier.HeatIdentifier('t', 's', 'i', u'\u5de5')
self.assertEqual(hi.url_path(), 't/stacks/s/i/%E5%B7%A5')
def test_tenant_escape(self):
hi = identifier.HeatIdentifier(':/', 's', 'i')
self.assertEqual(hi.tenant, ':/')

View File

@ -22,6 +22,7 @@ module=threadgroup
module=timeutils
module=uuidutils
module=config
module=strutils
# The base module to hold the copy of openstack.common
base=heat