#!/usr/bin/env python
# -*- coding: utf-8 -*-
from __future__ import absolute_import
import base64
import datetime
import itertools
import netrc
import os
import re
import socket
import time
import email.utils
import xml.etree.ElementTree
import random
import math
import operator
import hashlib
import binascii
import urllib
from .utils import *
from .extractor.common import InfoExtractor, SearchInfoExtractor
from .extractor.ard import ARDIE
from .extractor.arte import ArteTvIE
from .extractor.dailymotion import DailymotionIE
from .extractor.metacafe import MetacafeIE
from .extractor.statigram import StatigramIE
from .extractor.photobucket import PhotobucketIE
from .extractor.vimeo import VimeoIE
from .extractor.yahoo import YahooIE
from .extractor.youtube import YoutubeIE, YoutubePlaylistIE, YoutubeUserIE, YoutubeChannelIE
from .extractor.zdf import ZDFIE
class GenericIE(InfoExtractor):
"""Generic last-resort information extractor."""
_VALID_URL = r'.*'
IE_NAME = u'generic'
def report_download_webpage(self, video_id):
"""Report webpage download."""
if not self._downloader.params.get('test', False):
self._downloader.report_warning(u'Falling back on generic information extractor.')
super(GenericIE, self).report_download_webpage(video_id)
def report_following_redirect(self, new_url):
"""Report information extraction."""
self._downloader.to_screen(u'[redirect] Following redirect to %s' % new_url)
def _test_redirect(self, url):
"""Check if it is a redirect, like url shorteners, in case return the new url."""
class HeadRequest(compat_urllib_request.Request):
def get_method(self):
return "HEAD"
class HEADRedirectHandler(compat_urllib_request.HTTPRedirectHandler):
"""
Subclass the HTTPRedirectHandler to make it use our
HeadRequest also on the redirected URL
"""
def redirect_request(self, req, fp, code, msg, headers, newurl):
if code in (301, 302, 303, 307):
newurl = newurl.replace(' ', '%20')
newheaders = dict((k,v) for k,v in req.headers.items()
if k.lower() not in ("content-length", "content-type"))
return HeadRequest(newurl,
headers=newheaders,
origin_req_host=req.get_origin_req_host(),
unverifiable=True)
else:
raise compat_urllib_error.HTTPError(req.get_full_url(), code, msg, headers, fp)
class HTTPMethodFallback(compat_urllib_request.BaseHandler):
"""
Fallback to GET if HEAD is not allowed (405 HTTP error)
"""
def http_error_405(self, req, fp, code, msg, headers):
fp.read()
fp.close()
newheaders = dict((k,v) for k,v in req.headers.items()
if k.lower() not in ("content-length", "content-type"))
return self.parent.open(compat_urllib_request.Request(req.get_full_url(),
headers=newheaders,
origin_req_host=req.get_origin_req_host(),
unverifiable=True))
# Build our opener
opener = compat_urllib_request.OpenerDirector()
for handler in [compat_urllib_request.HTTPHandler, compat_urllib_request.HTTPDefaultErrorHandler,
HTTPMethodFallback, HEADRedirectHandler,
compat_urllib_request.HTTPErrorProcessor, compat_urllib_request.HTTPSHandler]:
opener.add_handler(handler())
response = opener.open(HeadRequest(url))
if response is None:
raise ExtractorError(u'Invalid URL protocol')
new_url = response.geturl()
if url == new_url:
return False
self.report_following_redirect(new_url)
return new_url
def _real_extract(self, url):
new_url = self._test_redirect(url)
if new_url: return [self.url_result(new_url)]
video_id = url.split('/')[-1]
try:
webpage = self._download_webpage(url, video_id)
except ValueError as err:
# since this is the last-resort InfoExtractor, if
# this error is thrown, it'll be thrown here
raise ExtractorError(u'Invalid URL: %s' % url)
self.report_extraction(video_id)
# Start with something easy: JW Player in SWFObject
mobj = re.search(r'flashvars: [\'"](?:.*&)?file=(http[^\'"&]*)', webpage)
if mobj is None:
# Broaden the search a little bit
mobj = re.search(r'[^A-Za-z0-9]?(?:file|source)=(http[^\'"&]*)', webpage)
if mobj is None:
# Broaden the search a little bit: JWPlayer JS loader
mobj = re.search(r'[^A-Za-z0-9]?file:\s*["\'](http[^\'"&]*)', webpage)
if mobj is None:
# Try to find twitter cards info
mobj = re.search(r'(.*)',
webpage, u'video title')
# video uploader is domain name
video_uploader = self._search_regex(r'(?:https?://)?([^/]*)/.*',
url, u'video uploader')
return [{
'id': video_id,
'url': video_url,
'uploader': video_uploader,
'upload_date': None,
'title': video_title,
'ext': video_extension,
}]
class YoutubeSearchIE(SearchInfoExtractor):
"""Information Extractor for YouTube search queries."""
_API_URL = 'https://gdata.youtube.com/feeds/api/videos?q=%s&start-index=%i&max-results=50&v=2&alt=jsonc'
_MAX_RESULTS = 1000
IE_NAME = u'youtube:search'
_SEARCH_KEY = 'ytsearch'
def report_download_page(self, query, pagenum):
"""Report attempt to download search page with given number."""
self._downloader.to_screen(u'[youtube] query "%s": Downloading page %s' % (query, pagenum))
def _get_n_results(self, query, n):
"""Get a specified number of results for a query"""
video_ids = []
pagenum = 0
limit = n
while (50 * pagenum) < limit:
self.report_download_page(query, pagenum+1)
result_url = self._API_URL % (compat_urllib_parse.quote_plus(query), (50*pagenum)+1)
request = compat_urllib_request.Request(result_url)
try:
data = compat_urllib_request.urlopen(request).read().decode('utf-8')
except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
raise ExtractorError(u'Unable to download API page: %s' % compat_str(err))
api_response = json.loads(data)['data']
if not 'items' in api_response:
raise ExtractorError(u'[youtube] No video results')
new_ids = list(video['id'] for video in api_response['items'])
video_ids += new_ids
limit = min(n, api_response['totalItems'])
pagenum += 1
if len(video_ids) > n:
video_ids = video_ids[:n]
videos = [self.url_result('http://www.youtube.com/watch?v=%s' % id, 'Youtube') for id in video_ids]
return self.playlist_result(videos, query)
class GoogleSearchIE(SearchInfoExtractor):
"""Information Extractor for Google Video search queries."""
_MORE_PAGES_INDICATOR = r'id="pnnext" class="pn"'
_MAX_RESULTS = 1000
IE_NAME = u'video.google:search'
_SEARCH_KEY = 'gvsearch'
def _get_n_results(self, query, n):
"""Get a specified number of results for a query"""
res = {
'_type': 'playlist',
'id': query,
'entries': []
}
for pagenum in itertools.count(1):
result_url = u'http://www.google.com/search?tbm=vid&q=%s&start=%s&hl=en' % (compat_urllib_parse.quote_plus(query), pagenum*10)
webpage = self._download_webpage(result_url, u'gvsearch:' + query,
note='Downloading result page ' + str(pagenum))
for mobj in re.finditer(r'