Raphaël G. Git Repositories - youtubedl/blob - youtube_dl/extractor/yahoo.py

   1 import datetime
   2 import itertools
   3 import json
   4 import re
   5
   6 from .common import InfoExtractor, SearchInfoExtractor
   7 from ..utils import (
   8     compat_urllib_parse,
   9
  10     ExtractorError,
  11 )
  12
  13 class YahooIE(InfoExtractor):
  14     IE_DESC = u'Yahoo screen'
  15     _VALID_URL = r'http://screen\.yahoo\.com/.*?-(?P<id>\d*?)\.html'
  16     _TEST = {
  17         u'url': u'http://screen.yahoo.com/julian-smith-travis-legg-watch-214727115.html',
  18         u'file': u'214727115.flv',
  19         u'md5': u'2e717f169c1be93d84d3794a00d4a325',
  20         u'info_dict': {
  21             u"title": u"Julian Smith & Travis Legg Watch Julian Smith"
  22         },
  23         u'skip': u'Requires rtmpdump'
  24     }
  25
  26     def _real_extract(self, url):
  27         mobj = re.match(self._VALID_URL, url)
  28         if mobj is None:
  29             raise ExtractorError(u'Invalid URL: %s' % url)
  30         video_id = mobj.group('id')
  31         webpage = self._download_webpage(url, video_id)
  32         m_id = re.search(r'YUI\.namespace\("Media"\)\.CONTENT_ID = "(?P<new_id>.+?)";', webpage)
  33
  34         if m_id is None:
  35             # TODO: Check which url parameters are required
  36             info_url = 'http://cosmos.bcst.yahoo.com/rest/v2/pops;lmsoverride=1;outputformat=mrss;cb=974419660;id=%s;rd=news.yahoo.com;datacontext=mdb;lg=KCa2IihxG3qE60vQ7HtyUy' % video_id
  37             webpage = self._download_webpage(info_url, video_id, u'Downloading info webpage')
  38             info_re = r'''<title><!\[CDATA\[(?P<title>.*?)\]\]></title>.*
  39                         <description><!\[CDATA\[(?P<description>.*?)\]\]></description>.*
  40                         <media:pubStart><!\[CDATA\[(?P<date>.*?)\ .*\]\]></media:pubStart>.*
  41                         <media:content\ medium="image"\ url="(?P<thumb>.*?)"\ name="LARGETHUMB"
  42                         '''
  43             self.report_extraction(video_id)
  44             m_info = re.search(info_re, webpage, re.VERBOSE|re.DOTALL)
  45             if m_info is None:
  46                 raise ExtractorError(u'Unable to extract video info')
  47             video_title = m_info.group('title')
  48             video_description = m_info.group('description')
  49             video_thumb = m_info.group('thumb')
  50             video_date = m_info.group('date')
  51             video_date = datetime.datetime.strptime(video_date, '%m/%d/%Y').strftime('%Y%m%d')
  52
  53             # TODO: Find a way to get mp4 videos
  54             rest_url = 'http://cosmos.bcst.yahoo.com/rest/v2/pops;element=stream;outputformat=mrss;id=%s;lmsoverride=1;bw=375;dynamicstream=1;cb=83521105;tech=flv,mp4;rd=news.yahoo.com;datacontext=mdb;lg=KCa2IihxG3qE60vQ7HtyUy' % video_id
  55             webpage = self._download_webpage(rest_url, video_id, u'Downloading video url webpage')
  56             m_rest = re.search(r'<media:content url="(?P<url>.*?)" path="(?P<path>.*?)"', webpage)
  57             video_url = m_rest.group('url')
  58             video_path = m_rest.group('path')
  59             if m_rest is None:
  60                 raise ExtractorError(u'Unable to extract video url')
  61
  62         else: # We have to use a different method if another id is defined
  63             long_id = m_id.group('new_id')
  64             info_url = 'http://video.query.yahoo.com/v1/public/yql?q=SELECT%20*%20FROM%20yahoo.media.video.streams%20WHERE%20id%3D%22' + long_id + '%22%20AND%20format%3D%22mp4%2Cflv%22%20AND%20protocol%3D%22rtmp%2Chttp%22%20AND%20plrs%3D%2286Gj0vCaSzV_Iuf6hNylf2%22%20AND%20acctid%3D%22389%22%20AND%20plidl%3D%22%22%20AND%20pspid%3D%22792700001%22%20AND%20offnetwork%3D%22false%22%20AND%20site%3D%22ivy%22%20AND%20lang%3D%22en-US%22%20AND%20region%3D%22US%22%20AND%20override%3D%22none%22%3B&env=prod&format=json&callback=YUI.Env.JSONP.yui_3_8_1_1_1368368376830_335'
  65             webpage = self._download_webpage(info_url, video_id, u'Downloading info json')
  66             json_str = re.search(r'YUI.Env.JSONP.yui.*?\((.*?)\);', webpage).group(1)
  67             info = json.loads(json_str)
  68             res = info[u'query'][u'results'][u'mediaObj'][0]
  69             stream = res[u'streams'][0]
  70             video_path = stream[u'path']
  71             video_url = stream[u'host']
  72             meta = res[u'meta']
  73             video_title = meta[u'title']
  74             video_description = meta[u'description']
  75             video_thumb = meta[u'thumbnail']
  76             video_date = None # I can't find it
  77
  78         info_dict = {
  79                      'id': video_id,
  80                      'url': video_url,
  81                      'play_path': video_path,
  82                      'title':video_title,
  83                      'description': video_description,
  84                      'thumbnail': video_thumb,
  85                      'upload_date': video_date,
  86                      'ext': 'flv',
  87                      }
  88         return info_dict
  89
  90 class YahooSearchIE(SearchInfoExtractor):
  91     IE_DESC = u'Yahoo screen search'
  92     _MAX_RESULTS = 1000
  93     IE_NAME = u'screen.yahoo:search'
  94     _SEARCH_KEY = 'yvsearch'
  95
  96     def _get_n_results(self, query, n):
  97         """Get a specified number of results for a query"""
  98
  99         res = {
 100             '_type': 'playlist',
 101             'id': query,
 102             'entries': []
 103         }
 104         for pagenum in itertools.count(0):
 105             result_url = u'http://video.search.yahoo.com/search/?p=%s&fr=screen&o=js&gs=0&b=%d' % (compat_urllib_parse.quote_plus(query), pagenum * 30)
 106             webpage = self._download_webpage(result_url, query,
 107                                              note='Downloading results page '+str(pagenum+1))
 108             info = json.loads(webpage)
 109             m = info[u'm']
 110             results = info[u'results']
 111
 112             for (i, r) in enumerate(results):
 113                 if (pagenum * 30) +i >= n:
 114                     break
 115                 mobj = re.search(r'(?P<url>screen\.yahoo\.com/.*?-\d*?\.html)"', r)
 116                 e = self.url_result('http://' + mobj.group('url'), 'Yahoo')
 117                 res['entries'].append(e)
 118             if (pagenum * 30 +i >= n) or (m[u'last'] >= (m[u'total'] -1 )):
 119                 break
 120
 121         return res