# -*- coding: utf-8 -*-
import gzip
+import io
import locale
import os
import re
import email.utils
import json
-try:
- import cStringIO as StringIO
-except ImportError:
- import StringIO
-
try:
import urllib.request as compat_urllib_request
except ImportError: # Python 2
try:
import html.entities as compat_html_entities
-except NameError: # Python 2
+except ImportError: # Python 2
import htmlentitydefs as compat_html_entities
try:
import html.parser as compat_html_parser
-except NameError: # Python 2
+except ImportError: # Python 2
import HTMLParser as compat_html_parser
+try:
+ import http.client as compat_http_client
+except ImportError: # Python 2
+ import httplib as compat_http_client
+
+try:
+ from urllib.parse import parse_qs as compat_parse_qs
+except ImportError: # Python 2
+ from urlparse import parse_qs as compat_parse_qs
+
try:
compat_str = unicode # Python 2
except NameError:
except NameError:
compat_chr = chr
-
std_headers = {
'User-Agent': 'Mozilla/5.0 (X11; Linux x86_64; rv:10.0) Gecko/20100101 Firefox/10.0',
'Accept-Charset': 'ISO-8859-1,utf-8;q=0.7,*;q=0.7',
return pref
+if sys.version_info < (3,0):
+ def compat_print(s):
+ print(s.encode(preferredencoding(), 'xmlcharrefreplace'))
+else:
+ def compat_print(s):
+ assert type(s) == type(u'')
+ print(s)
def htmlentity_transform(matchobj):
"""Transforms an HTML entity to a character.
old_resp = resp
# gzip
if resp.headers.get('Content-encoding', '') == 'gzip':
- gz = gzip.GzipFile(fileobj=StringIO.StringIO(resp.read()), mode='r')
+ gz = gzip.GzipFile(fileobj=io.BytesIO(resp.read()), mode='r')
resp = self.addinfourl_wrapper(gz, old_resp.headers, old_resp.url, old_resp.code)
resp.msg = old_resp.msg
# deflate
if resp.headers.get('Content-encoding', '') == 'deflate':
- gz = StringIO.StringIO(self.deflate(resp.read()))
+ gz = io.BytesIO(self.deflate(resp.read()))
resp = self.addinfourl_wrapper(gz, old_resp.headers, old_resp.url, old_resp.code)
resp.msg = old_resp.msg
return resp