from __future__ import unicode_literals
+import binascii
import collections
+import email
import getpass
+import io
import optparse
import os
import re
+import shlex
import shutil
import socket
+import struct
import subprocess
import sys
import itertools
+import xml.etree.ElementTree
try:
except ImportError: # Python 2
import urlparse as compat_urlparse
+try:
+ import urllib.response as compat_urllib_response
+except ImportError: # Python 2
+ import urllib as compat_urllib_response
+
try:
import http.cookiejar as compat_cookiejar
except ImportError: # Python 2
import cookielib as compat_cookiejar
+try:
+ import http.cookies as compat_cookies
+except ImportError: # Python 2
+ import Cookie as compat_cookies
+
try:
import html.entities as compat_html_entities
except ImportError: # Python 2
except ImportError: # Python 2
from urllib import urlretrieve as compat_urlretrieve
+try:
+ from html.parser import HTMLParser as compat_HTMLParser
+except ImportError: # Python 2
+ from HTMLParser import HTMLParser as compat_HTMLParser
+
try:
from subprocess import DEVNULL
except ImportError:
import BaseHTTPServer as compat_http_server
+try:
+ compat_str = unicode # Python 2
+except NameError:
+ compat_str = str
+
try:
from urllib.parse import unquote_to_bytes as compat_urllib_parse_unquote_to_bytes
from urllib.parse import unquote as compat_urllib_parse_unquote
from urllib.parse import unquote_plus as compat_urllib_parse_unquote_plus
except ImportError: # Python 2
- _asciire = re.compile('([\x00-\x7f]+)') if sys.version_info < (2, 7) else compat_urllib_parse._asciire
+ _asciire = (compat_urllib_parse._asciire if hasattr(compat_urllib_parse, '_asciire')
+ else re.compile('([\x00-\x7f]+)'))
# HACK: The following are the correct unquote_to_bytes, unquote and unquote_plus
# implementations from cpython 3.4.3's stdlib. Python 2's version
# Is it a string-like object?
string.split
return b''
- if isinstance(string, unicode):
+ if isinstance(string, compat_str):
string = string.encode('utf-8')
bits = string.split(b'%')
if len(bits) == 1:
return compat_urllib_parse_unquote(string, encoding, errors)
try:
- compat_str = unicode # Python 2
-except NameError:
- compat_str = str
+ from urllib.parse import urlencode as compat_urllib_parse_urlencode
+except ImportError: # Python 2
+ # Python 2 will choke in urlencode on mixture of byte and unicode strings.
+ # Possible solutions are to either port it from python 3 with all
+ # the friends or manually ensure input query contains only byte strings.
+ # We will stick with latter thus recursively encoding the whole query.
+ def compat_urllib_parse_urlencode(query, doseq=0, encoding='utf-8'):
+ def encode_elem(e):
+ if isinstance(e, dict):
+ e = encode_dict(e)
+ elif isinstance(e, (list, tuple,)):
+ list_e = encode_list(e)
+ e = tuple(list_e) if isinstance(e, tuple) else list_e
+ elif isinstance(e, compat_str):
+ e = e.encode(encoding)
+ return e
+
+ def encode_dict(d):
+ return dict((encode_elem(k), encode_elem(v)) for k, v in d.items())
+
+ def encode_list(l):
+ return [encode_elem(e) for e in l]
+
+ return compat_urllib_parse.urlencode(encode_elem(query), doseq=doseq)
+
+try:
+ from urllib.request import DataHandler as compat_urllib_request_DataHandler
+except ImportError: # Python < 3.4
+ # Ported from CPython 98774:1733b3bd46db, Lib/urllib/request.py
+ class compat_urllib_request_DataHandler(compat_urllib_request.BaseHandler):
+ def data_open(self, req):
+ # data URLs as specified in RFC 2397.
+ #
+ # ignores POSTed data
+ #
+ # syntax:
+ # dataurl := "data:" [ mediatype ] [ ";base64" ] "," data
+ # mediatype := [ type "/" subtype ] *( ";" parameter )
+ # data := *urlchar
+ # parameter := attribute "=" value
+ url = req.get_full_url()
+
+ scheme, data = url.split(':', 1)
+ mediatype, data = data.split(',', 1)
+
+ # even base64 encoded data URLs might be quoted so unquote in any case:
+ data = compat_urllib_parse_unquote_to_bytes(data)
+ if mediatype.endswith(';base64'):
+ data = binascii.a2b_base64(data)
+ mediatype = mediatype[:-7]
+
+ if not mediatype:
+ mediatype = 'text/plain;charset=US-ASCII'
+
+ headers = email.message_from_string(
+ 'Content-type: %s\nContent-length: %d\n' % (mediatype, len(data)))
+
+ return compat_urllib_response.addinfourl(io.BytesIO(data), headers, url)
try:
compat_basestring = basestring # Python 2
except ImportError: # Python 2.6
from xml.parsers.expat import ExpatError as compat_xml_parse_error
+if sys.version_info[0] >= 3:
+ compat_etree_fromstring = xml.etree.ElementTree.fromstring
+else:
+ # python 2.x tries to encode unicode strings with ascii (see the
+ # XMLParser._fixtext method)
+ etree = xml.etree.ElementTree
+
+ try:
+ _etree_iter = etree.Element.iter
+ except AttributeError: # Python <=2.6
+ def _etree_iter(root):
+ for el in root.findall('*'):
+ yield el
+ for sub in _etree_iter(el):
+ yield sub
+
+ # on 2.6 XML doesn't have a parser argument, function copied from CPython
+ # 2.7 source
+ def _XML(text, parser=None):
+ if not parser:
+ parser = etree.XMLParser(target=etree.TreeBuilder())
+ parser.feed(text)
+ return parser.close()
+
+ def _element_factory(*args, **kwargs):
+ el = etree.Element(*args, **kwargs)
+ for k, v in el.items():
+ if isinstance(v, bytes):
+ el.set(k, v.decode('utf-8'))
+ return el
+
+ def compat_etree_fromstring(text):
+ doc = _XML(text, parser=etree.XMLParser(target=etree.TreeBuilder(element_factory=_element_factory)))
+ for el in _etree_iter(doc):
+ if el.text is not None and isinstance(el.text, bytes):
+ el.text = el.text.decode('utf-8')
+ return doc
+
+if sys.version_info < (2, 7):
+ # Here comes the crazy part: In 2.6, if the xpath is a unicode,
+ # .//node does not match if a node is a direct child of . !
+ def compat_xpath(xpath):
+ if isinstance(xpath, compat_str):
+ xpath = xpath.encode('ascii')
+ return xpath
+else:
+ compat_xpath = lambda xpath: xpath
try:
from urllib.parse import parse_qs as compat_parse_qs
nv = name_value.split('=', 1)
if len(nv) != 2:
if strict_parsing:
- raise ValueError("bad query field: %r" % (name_value,))
+ raise ValueError('bad query field: %r' % (name_value,))
# Handle case of a control-name with no equal sign
if keep_blank_values:
nv.append('')
return parsed_result
try:
- from shlex import quote as shlex_quote
+ from shlex import quote as compat_shlex_quote
except ImportError: # Python < 3.3
- def shlex_quote(s):
+ def compat_shlex_quote(s):
if re.match(r'^[-_\w./]+$', s):
return s
else:
return "'" + s.replace("'", "'\"'\"'") + "'"
+if sys.version_info >= (2, 7, 3):
+ compat_shlex_split = shlex.split
+else:
+ # Working around shlex issue with unicode strings on some python 2
+ # versions (see http://bugs.python.org/issue1548891)
+ def compat_shlex_split(s, comments=False, posix=True):
+ if isinstance(s, compat_str):
+ s = s.encode('utf-8')
+ return shlex.split(s, comments, posix)
+
+
def compat_ord(c):
if type(c) is int:
return c
return ord(c)
+compat_os_name = os._name if os.name == 'java' else os.name
+
+
if sys.version_info >= (3, 0):
compat_getenv = os.getenv
compat_expanduser = os.path.expanduser
+
+ def compat_setenv(key, value, env=os.environ):
+ env[key] = value
else:
# Environment variables should be decoded with filesystem encoding.
# Otherwise it will fail if any non-ASCII characters present (see #3854 #3217 #2918)
env = env.decode(get_filesystem_encoding())
return env
+ def compat_setenv(key, value, env=os.environ):
+ def encode(v):
+ from .utils import get_filesystem_encoding
+ return v.encode(get_filesystem_encoding()) if isinstance(v, compat_str) else v
+ env[encode(key)] = encode(value)
+
# HACK: The default implementations of os.path.expanduser from cpython do not decode
# environment variables with filesystem encoding. We will work around this by
# providing adjusted implementations.
# The following are os.path.expanduser implementations from cpython 2.7.8 stdlib
# for different platforms with correct environment variables decoding.
- if os.name == 'posix':
+ if compat_os_name == 'posix':
def compat_expanduser(path):
"""Expand ~ and ~user constructions. If user or $HOME is unknown,
do nothing."""
userhome = pwent.pw_dir
userhome = userhome.rstrip('/')
return (userhome + path[i:]) or '/'
- elif os.name == 'nt' or os.name == 'ce':
+ elif compat_os_name == 'nt' or compat_os_name == 'ce':
def compat_expanduser(path):
"""Expand ~ and ~user constructs.
print(s)
-try:
- subprocess_check_output = subprocess.check_output
-except AttributeError:
- def subprocess_check_output(*args, **kwargs):
- assert 'input' not in kwargs
- p = subprocess.Popen(*args, stdout=subprocess.PIPE, **kwargs)
- output, _ = p.communicate()
- ret = p.poll()
- if ret:
- raise subprocess.CalledProcessError(ret, p.args, output=output)
- return output
-
if sys.version_info < (3, 0) and sys.platform == 'win32':
def compat_getpass(prompt, *args, **kwargs):
if isinstance(prompt, compat_str):
else:
compat_getpass = getpass.getpass
-# Old 2.6 and 2.7 releases require kwargs to be bytes
+# Python < 2.6.5 require kwargs to be bytes
try:
def _testfunc(x):
pass
if err is not None:
raise err
else:
- raise socket.error("getaddrinfo returns an empty list")
+ raise socket.error('getaddrinfo returns an empty list')
else:
compat_socket_create_connection = socket.create_connection
else:
_terminal_size = collections.namedtuple('terminal_size', ['columns', 'lines'])
- def compat_get_terminal_size():
- columns = compat_getenv('COLUMNS', None)
+ def compat_get_terminal_size(fallback=(80, 24)):
+ columns = compat_getenv('COLUMNS')
if columns:
columns = int(columns)
else:
columns = None
- lines = compat_getenv('LINES', None)
+ lines = compat_getenv('LINES')
if lines:
lines = int(lines)
else:
lines = None
- try:
- sp = subprocess.Popen(
- ['stty', 'size'],
- stdout=subprocess.PIPE, stderr=subprocess.PIPE)
- out, err = sp.communicate()
- lines, columns = map(int, out.split())
- except Exception:
- pass
+ if columns is None or lines is None or columns <= 0 or lines <= 0:
+ try:
+ sp = subprocess.Popen(
+ ['stty', 'size'],
+ stdout=subprocess.PIPE, stderr=subprocess.PIPE)
+ out, err = sp.communicate()
+ _lines, _columns = map(int, out.split())
+ except Exception:
+ _columns, _lines = _terminal_size(*fallback)
+
+ if columns is None or columns <= 0:
+ columns = _columns
+ if lines is None or lines <= 0:
+ lines = _lines
return _terminal_size(columns, lines)
try:
yield n
n += step
+if sys.version_info >= (3, 0):
+ from tokenize import tokenize as compat_tokenize_tokenize
+else:
+ from tokenize import generate_tokens as compat_tokenize_tokenize
+
+
+try:
+ struct.pack('!I', 0)
+except TypeError:
+ # In Python 2.6 and 2.7.x < 2.7.7, struct requires a bytes argument
+ # See https://bugs.python.org/issue19099
+ def compat_struct_pack(spec, *args):
+ if isinstance(spec, compat_str):
+ spec = spec.encode('ascii')
+ return struct.pack(spec, *args)
+
+ def compat_struct_unpack(spec, *args):
+ if isinstance(spec, compat_str):
+ spec = spec.encode('ascii')
+ return struct.unpack(spec, *args)
+else:
+ compat_struct_pack = struct.pack
+ compat_struct_unpack = struct.unpack
+
+
__all__ = [
+ 'compat_HTMLParser',
'compat_HTTPError',
'compat_basestring',
'compat_chr',
'compat_cookiejar',
+ 'compat_cookies',
+ 'compat_etree_fromstring',
'compat_expanduser',
'compat_get_terminal_size',
'compat_getenv',
'compat_itertools_count',
'compat_kwargs',
'compat_ord',
+ 'compat_os_name',
'compat_parse_qs',
'compat_print',
+ 'compat_setenv',
+ 'compat_shlex_quote',
+ 'compat_shlex_split',
'compat_socket_create_connection',
'compat_str',
+ 'compat_struct_pack',
+ 'compat_struct_unpack',
'compat_subprocess_get_DEVNULL',
+ 'compat_tokenize_tokenize',
'compat_urllib_error',
'compat_urllib_parse',
'compat_urllib_parse_unquote',
'compat_urllib_parse_unquote_plus',
'compat_urllib_parse_unquote_to_bytes',
+ 'compat_urllib_parse_urlencode',
'compat_urllib_parse_urlparse',
'compat_urllib_request',
+ 'compat_urllib_request_DataHandler',
+ 'compat_urllib_response',
'compat_urlparse',
'compat_urlretrieve',
'compat_xml_parse_error',
- 'shlex_quote',
- 'subprocess_check_output',
+ 'compat_xpath',
'workaround_optparse_bug9161',
]
projects / youtube-dl / blobdiff