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

   1 import re
   2 import socket
   3
   4 from .common import InfoExtractor
   5 from ..utils import (
   6     compat_http_client,
   7     compat_str,
   8     compat_urllib_error,
   9     compat_urllib_parse,
  10     compat_urllib_request,
  11
  12     ExtractorError,
  13     unified_strdate,
  14 )
  15
  16 class ArteTvIE(InfoExtractor):
  17     """arte.tv information extractor."""
  18
  19     _VALID_URL = r'(?:http://)?videos\.arte\.tv/(?:fr|de)/videos/.*'
  20     _LIVE_URL = r'index-[0-9]+\.html$'
  21
  22     IE_NAME = u'arte.tv'
  23
  24     def fetch_webpage(self, url):
  25         request = compat_urllib_request.Request(url)
  26         try:
  27             self.report_download_webpage(url)
  28             webpage = compat_urllib_request.urlopen(request).read()
  29         except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
  30             raise ExtractorError(u'Unable to retrieve video webpage: %s' % compat_str(err))
  31         except ValueError as err:
  32             raise ExtractorError(u'Invalid URL: %s' % url)
  33         return webpage
  34
  35     def grep_webpage(self, url, regex, regexFlags, matchTuples):
  36         page = self.fetch_webpage(url)
  37         mobj = re.search(regex, page, regexFlags)
  38         info = {}
  39
  40         if mobj is None:
  41             raise ExtractorError(u'Invalid URL: %s' % url)
  42
  43         for (i, key, err) in matchTuples:
  44             if mobj.group(i) is None:
  45                 raise ExtractorError(err)
  46             else:
  47                 info[key] = mobj.group(i)
  48
  49         return info
  50
  51     # TODO implement Live Stream
  52     # def extractLiveStream(self, url):
  53     #     video_lang = url.split('/')[-4]
  54     #     info = self.grep_webpage(
  55     #         url,
  56     #         r'src="(.*?/videothek_js.*?\.js)',
  57     #         0,
  58     #         [
  59     #             (1, 'url', u'Invalid URL: %s' % url)
  60     #         ]
  61     #     )
  62     #     http_host = url.split('/')[2]
  63     #     next_url = 'http://%s%s' % (http_host, compat_urllib_parse.unquote(info.get('url')))
  64     #     info = self.grep_webpage(
  65     #         next_url,
  66     #         r'(s_artestras_scst_geoFRDE_' + video_lang + '.*?)\'.*?' +
  67     #             '(http://.*?\.swf).*?' +
  68     #             '(rtmp://.*?)\'',
  69     #         re.DOTALL,
  70     #         [
  71     #             (1, 'path',   u'could not extract video path: %s' % url),
  72     #             (2, 'player', u'could not extract video player: %s' % url),
  73     #             (3, 'url',    u'could not extract video url: %s' % url)
  74     #         ]
  75     #     )
  76     #     video_url = u'%s/%s' % (info.get('url'), info.get('path'))
  77
  78     def extractPlus7Stream(self, url):
  79         video_lang = url.split('/')[-3]
  80         info = self.grep_webpage(
  81             url,
  82             r'param name="movie".*?videorefFileUrl=(http[^\'"&]*)',
  83             0,
  84             [
  85                 (1, 'url', u'Invalid URL: %s' % url)
  86             ]
  87         )
  88         next_url = compat_urllib_parse.unquote(info.get('url'))
  89         info = self.grep_webpage(
  90             next_url,
  91             r'<video lang="%s" ref="(http[^\'"&]*)' % video_lang,
  92             0,
  93             [
  94                 (1, 'url', u'Could not find <video> tag: %s' % url)
  95             ]
  96         )
  97         next_url = compat_urllib_parse.unquote(info.get('url'))
  98
  99         info = self.grep_webpage(
 100             next_url,
 101             r'<video id="(.*?)".*?>.*?' +
 102                 '<name>(.*?)</name>.*?' +
 103                 '<dateVideo>(.*?)</dateVideo>.*?' +
 104                 '<url quality="hd">(.*?)</url>',
 105             re.DOTALL,
 106             [
 107                 (1, 'id',    u'could not extract video id: %s' % url),
 108                 (2, 'title', u'could not extract video title: %s' % url),
 109                 (3, 'date',  u'could not extract video date: %s' % url),
 110                 (4, 'url',   u'could not extract video url: %s' % url)
 111             ]
 112         )
 113
 114         return {
 115             'id':           info.get('id'),
 116             'url':          compat_urllib_parse.unquote(info.get('url')),
 117             'uploader':     u'arte.tv',
 118             'upload_date':  unified_strdate(info.get('date')),
 119             'title':        info.get('title').decode('utf-8'),
 120             'ext':          u'mp4',
 121             'format':       u'NA',
 122             'player_url':   None,
 123         }
 124
 125     def _real_extract(self, url):
 126         video_id = url.split('/')[-1]
 127         self.report_extraction(video_id)
 128
 129         if re.search(self._LIVE_URL, video_id) is not None:
 130             raise ExtractorError(u'Arte live streams are not yet supported, sorry')
 131             # self.extractLiveStream(url)
 132             # return
 133         else:
 134             info = self.extractPlus7Stream(url)
 135
 136         return [info]