_ Git - youtube-dl/blob - youtube_dl/extractor/br.py

   1 # coding: utf-8
   2
   3 from .common import InfoExtractor
   4
   5 class BRIE(InfoExtractor):
   6
   7     IE_DESC = u"Bayerischer Rundfunk Mediathek"
   8     _VALID_URL = r"^https?://(?:www\.)?br\.de/mediathek/video/(?:sendungen/)?(?:[a-z0-9\-]+\.html)$"
   9     _BASE_URL = u"http://www.br.de"
  10
  11     _TESTS = []
  12
  13     def _real_extract(self, url):
  14         page = self._download_webpage(url, None)
  15         xml_url = self._search_regex(r"return BRavFramework\.register\(BRavFramework\('avPlayer_(?:[a-f0-9-]{36})'\)\.setup\({dataURL:'(/mediathek/video/[a-z0-9/~_.-]+)'}\)\);", page, "XMLURL")
  16         xml = self._download_xml(self._BASE_URL + xml_url, None)
  17
  18         videos = []
  19         for xml_video in xml.findall("video"):
  20             video = {}
  21             video["id"] = xml_video.get("externalId")
  22             video["title"] = xml_video.find("title").text
  23             video["formats"] = self._extract_formats(xml_video.find("assets"))
  24             video["thumbnails"] = self._extract_thumbnails(xml_video.find("teaserImage/variants"))
  25             video["thumbnail"] = video["thumbnails"][0]["url"]
  26             video["description"] = " ".join(xml_video.find("shareTitle").text.splitlines())
  27             video["uploader"] = xml_video.find("author").text
  28             video["upload_date"] = "".join(reversed(xml_video.find("broadcastDate").text.split(".")))
  29             video["webpage_url"] = xml_video.find("permalink").text
  30             videos.append(video)
  31
  32         if len(videos) > 1:
  33             self._downloader.report_warning(u'found multiple videos; please'
  34                 u'report this with the video URL to http://yt-dl.org/bug')
  35         return videos[0]
  36
  37     def _extract_formats(self, assets):
  38         vformats = []
  39         for asset in assets.findall("asset"):
  40             if asset.find("downloadUrl") is None:
  41                 continue
  42             vformat = {}
  43             vformat["url"] = asset.find("downloadUrl").text
  44             vformat["ext"] = asset.find("mediaType").text
  45             vformat["format_id"] = asset.get("type")
  46             vformat["width"] = int(asset.find("frameWidth").text)
  47             vformat["height"] = int(asset.find("frameHeight").text)
  48             vformat["resolution"] = "%ix%i" % (vformat["width"], vformat["height"])
  49             vformat["tbr"] = int(asset.find("bitrateVideo").text)
  50             vformat["abr"] = int(asset.find("bitrateAudio").text)
  51             vformat["vcodec"] = asset.find("codecVideo").text
  52             vformat["container"] = vformat["ext"]
  53             vformat["filesize"] = int(asset.find("size").text)
  54             vformat["preference"] = vformat["quality"] = -1
  55             vformat["format"] = "%s container with %i Kbps %s" % (vformat["container"], vformat["tbr"], vformat["vcodec"])
  56             vformats.append(vformat)
  57         self._sort_formats(vformats)
  58         return vformats
  59
  60     def _extract_thumbnails(self, variants):
  61         thumbnails = []
  62         for variant in variants.findall("variant"):
  63             thumbnail = {}
  64             thumbnail["url"] = self._BASE_URL + variant.find("url").text
  65             thumbnail["width"] = int(variant.find("width").text)
  66             thumbnail["height"] = int(variant.find("height").text)
  67             thumbnail["resolution"] = "%ix%i" % (thumbnail["width"], thumbnail["height"])
  68             thumbnails.append(thumbnail)
  69         thumbnails.sort(key = lambda x: x["width"] * x["height"], reverse=True)
  70         return thumbnails