dinislage/CouchPotatoServer
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Tornado update

Browse Source
This commit is contained in:
Ruud
2013-02-22 23:20:16 +01:00
parent 3eed34c710
commit a7b78d4131
36 changed files with 1784 additions and 1293 deletions
Download Patch File Download Diff File Expand all files Collapse all files
libs/tornado/escape.py
+41 -54
View File
@@ -20,49 +20,34 @@ Also includes a few other miscellaneous string manipulation functions that
have crept in over time.
"""
from __future__ import absolute_import, division, with_statement
from __future__ import absolute_import, division, print_function, with_statement
import htmlentitydefs
import re
import sys
import urllib
# Python3 compatibility: On python2.5, introduce the bytes alias from 2.6
try:
bytes
except Exception:
bytes = str
from tornado.util import bytes_type, unicode_type, basestring_type, u
try:
from urlparse import parse_qs # Python 2.6+
from urllib.parse import parse_qs # py3
except ImportError:
from cgi import parse_qs
from urlparse import parse_qs # Python 2.6+
# json module is in the standard library as of python 2.6; fall back to
# simplejson if present for older versions.
try:
import json
assert hasattr(json, "loads") and hasattr(json, "dumps")
_json_decode = json.loads
_json_encode = json.dumps
except Exception:
try:
import simplejson
_json_decode = lambda s: simplejson.loads(_unicode(s))
_json_encode = lambda v: simplejson.dumps(v)
except ImportError:
try:
# For Google AppEngine
from django.utils import simplejson
_json_decode = lambda s: simplejson.loads(_unicode(s))
_json_encode = lambda v: simplejson.dumps(v)
except ImportError:
def _json_decode(s):
raise NotImplementedError(
"A JSON parser is required, e.g., simplejson at "
"http://pypi.python.org/pypi/simplejson/")
_json_encode = _json_decode
import htmlentitydefs # py2
except ImportError:
import html.entities as htmlentitydefs # py3
try:
import urllib.parse as urllib_parse # py3
except ImportError:
import urllib as urllib_parse # py2
import json
try:
unichr
except NameError:
unichr = chr
_XHTML_ESCAPE_RE = re.compile('[&<>"]')
_XHTML_ESCAPE_DICT = {'&': '&amp;', '<': '&lt;', '>': '&gt;', '"': '&quot;'}
@@ -87,12 +72,12 @@ def json_encode(value):
# the javscript. Some json libraries do this escaping by default,
# although python's standard library does not, so we do it here.
# http://stackoverflow.com/questions/1580647/json-why-are-forward-slashes-escaped
return _json_encode(recursive_unicode(value)).replace("</", "<\\/")
return json.dumps(recursive_unicode(value)).replace("</", "<\\/")
def json_decode(value):
"""Returns Python objects for the given JSON string."""
return _json_decode(to_basestring(value))
return json.loads(to_basestring(value))
def squeeze(value):
@@ -102,7 +87,7 @@ def squeeze(value):
def url_escape(value):
"""Returns a valid URL-encoded version of the given value."""
return urllib.quote_plus(utf8(value))
return urllib_parse.quote_plus(utf8(value))
# python 3 changed things around enough that we need two separate
# implementations of url_unescape. We also need our own implementation
@@ -117,9 +102,9 @@ if sys.version_info[0] < 3:
the result is a unicode string in the specified encoding.
"""
if encoding is None:
return urllib.unquote_plus(utf8(value))
return urllib_parse.unquote_plus(utf8(value))
else:
return unicode(urllib.unquote_plus(utf8(value)), encoding)
return unicode_type(urllib_parse.unquote_plus(utf8(value)), encoding)
parse_qs_bytes = parse_qs
else:
@@ -132,9 +117,9 @@ else:
the result is a unicode string in the specified encoding.
"""
if encoding is None:
return urllib.parse.unquote_to_bytes(value)
return urllib_parse.unquote_to_bytes(value)
else:
return urllib.unquote_plus(to_basestring(value), encoding=encoding)
return urllib_parse.unquote_plus(to_basestring(value), encoding=encoding)
def parse_qs_bytes(qs, keep_blank_values=False, strict_parsing=False):
"""Parses a query string like urlparse.parse_qs, but returns the
@@ -149,12 +134,12 @@ else:
result = parse_qs(qs, keep_blank_values, strict_parsing,
encoding='latin1', errors='strict')
encoded = {}
for k, v in result.iteritems():
for k, v in result.items():
encoded[k] = [i.encode('latin1') for i in v]
return encoded
_UTF8_TYPES = (bytes, type(None))
_UTF8_TYPES = (bytes_type, type(None))
def utf8(value):
@@ -165,10 +150,10 @@ def utf8(value):
"""
if isinstance(value, _UTF8_TYPES):
return value
assert isinstance(value, unicode)
assert isinstance(value, unicode_type)
return value.encode("utf-8")
_TO_UNICODE_TYPES = (unicode, type(None))
_TO_UNICODE_TYPES = (unicode_type, type(None))
def to_unicode(value):
@@ -179,7 +164,7 @@ def to_unicode(value):
"""
if isinstance(value, _TO_UNICODE_TYPES):
return value
assert isinstance(value, bytes)
assert isinstance(value, bytes_type)
return value.decode("utf-8")
# to_unicode was previously named _unicode not because it was private,
@@ -188,12 +173,12 @@ _unicode = to_unicode
# When dealing with the standard library across python 2 and 3 it is
# sometimes useful to have a direct conversion to the native string type
if str is unicode:
if str is unicode_type:
native_str = to_unicode
else:
native_str = utf8
_BASESTRING_TYPES = (basestring, type(None))
_BASESTRING_TYPES = (basestring_type, type(None))
def to_basestring(value):
@@ -207,7 +192,7 @@ def to_basestring(value):
"""
if isinstance(value, _BASESTRING_TYPES):
return value
assert isinstance(value, bytes)
assert isinstance(value, bytes_type)
return value.decode("utf-8")
@@ -217,12 +202,12 @@ def recursive_unicode(obj):
Supports lists, tuples, and dictionaries.
"""
if isinstance(obj, dict):
return dict((recursive_unicode(k), recursive_unicode(v)) for (k, v) in obj.iteritems())
return dict((recursive_unicode(k), recursive_unicode(v)) for (k, v) in obj.items())
elif isinstance(obj, list):
return list(recursive_unicode(i) for i in obj)
elif isinstance(obj, tuple):
return tuple(recursive_unicode(i) for i in obj)
elif isinstance(obj, bytes):
elif isinstance(obj, bytes_type):
return to_unicode(obj)
else:
return obj
@@ -232,7 +217,9 @@ def recursive_unicode(obj):
# but it gets all exponential on certain patterns (such as too many trailing
# dots), causing the regex matcher to never return.
# This regex should avoid those problems.
_URL_RE = re.compile(ur"""\b((?:([\w-]+):(/{1,3})|www[.])(?:(?:(?:[^\s&()]|&amp;|&quot;)*(?:[^!"#$%&'()*+,.:;<=>?@\[\]^`{|}~\s]))|(?:\((?:[^\s&()]|&amp;|&quot;)*\)))+)""")
# Use to_unicode instead of tornado.util.u - we don't want backslashes getting
# processed as escapes.
_URL_RE = re.compile(to_unicode(r"""\b((?:([\w-]+):(/{1,3})|www[.])(?:(?:(?:[^\s&()]|&amp;|&quot;)*(?:[^!"#$%&'()*+,.:;<=>?@\[\]^`{|}~\s]))|(?:\((?:[^\s&()]|&amp;|&quot;)*\)))+)"""))
def linkify(text, shorten=False, extra_params="",
@@ -302,7 +289,7 @@ def linkify(text, shorten=False, extra_params="",
# (no more slug, etc), so it really just provides a little
# extra indication of shortening.
url = url[:proto_len] + parts[0] + "/" + \
parts[1][:8].split('?')[0].split('.')[0]
parts[1][:8].split('?')[0].split('.')[0]
if len(url) > max_len * 1.5: # still too long
url = url[:max_len]
@@ -321,7 +308,7 @@ def linkify(text, shorten=False, extra_params="",
# have a status bar, such as Safari by default)
params += ' title="%s"' % href
return u'<a href="%s"%s>%s</a>' % (href, params, url)
return u('<a href="%s"%s>%s</a>') % (href, params, url)
# First HTML-escape so that our strings are all safe.
# The regex is modified to avoid character entites other than &amp; so
@@ -344,7 +331,7 @@ def _convert_entity(m):
def _build_unicode_map():
unicode_map = {}
for name, value in htmlentitydefs.name2codepoint.iteritems():
for name, value in htmlentitydefs.name2codepoint.items():
unicode_map[name] = unichr(value)
return unicode_map