_ 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             u"url": u"http://www.br.de/mediathek/video/anselm-gruen-114.html",
  14             u"file": u"2c8d81c5-6fb7-4a74-88d4-e768e5856532.mp4",
  15             u"md5": u"c4f83cf0f023ba5875aba0bf46860df2",
  16             u"info_dict": {
  17                 u"title": u"Feiern und Verzichten",
  18                 u"description": u"Anselm Grün: Feiern und Verzichten",
  19                 u"uploader": u"BR/Birgit Baier",
  20                 u"upload_date": u"20140301"
  21             }
  22         }
  23     ]
  24
  25     def _real_extract(self, url):
  26         page = self._download_webpage(url, None)
  27         xml_url = self._search_regex(r"return BRavFramework\.register\(BRavFramework\('avPlayer_(?:[a-f0-9-]{36})'\)\.setup\({dataURL:'(/mediathek/video/[a-z0-9/~_.-]+)'}\)\);", page, "XMLURL")
  28         xml = self._download_xml(self._BASE_URL + xml_url, None)
  29
  30         videos = []
  31         for xml_video in xml.findall("video"):
  32             video = {}
  33             video["id"] = xml_video.get("externalId")
  34             video["title"] = xml_video.find("title").text
  35             video["formats"] = self._extract_formats(xml_video.find("assets"))
  36             video["thumbnails"] = self._extract_thumbnails(xml_video.find("teaserImage/variants"))
  37             video["thumbnail"] = video["thumbnails"][0]["url"]
  38             video["description"] = " ".join(xml_video.find("shareTitle").text.splitlines())
  39             video["uploader"] = xml_video.find("author").text
  40             video["upload_date"] = "".join(reversed(xml_video.find("broadcastDate").text.split(".")))
  41             video["webpage_url"] = xml_video.find("permalink").text
  42             videos.append(video)
  43
  44         if len(videos) > 1:
  45             self._downloader.report_warning(u'found multiple videos; please'
  46                 u'report this with the video URL to http://yt-dl.org/bug')
  47         return videos[0]
  48
  49     def _extract_formats(self, assets):
  50         vformats = []
  51         for asset in assets.findall("asset"):
  52             if asset.find("downloadUrl") is None:
  53                 continue
  54             vformat = {}
  55             vformat["url"] = asset.find("downloadUrl").text
  56             vformat["ext"] = asset.find("mediaType").text
  57             vformat["format_id"] = asset.get("type")
  58             vformat["width"] = int(asset.find("frameWidth").text)
  59             vformat["height"] = int(asset.find("frameHeight").text)
  60             vformat["resolution"] = "%ix%i" % (vformat["width"], vformat["height"])
  61             vformat["tbr"] = int(asset.find("bitrateVideo").text)
  62             vformat["abr"] = int(asset.find("bitrateAudio").text)
  63             vformat["vcodec"] = asset.find("codecVideo").text
  64             vformat["container"] = vformat["ext"]
  65             vformat["filesize"] = int(asset.find("size").text)
  66             vformat["preference"] = vformat["quality"] = -1
  67             vformat["format"] = "%s container with %i Kbps %s" % (vformat["container"], vformat["tbr"], vformat["vcodec"])
  68             vformats.append(vformat)
  69         self._sort_formats(vformats)
  70         return vformats
  71
  72     def _extract_thumbnails(self, variants):
  73         thumbnails = []
  74         for variant in variants.findall("variant"):
  75             thumbnail = {}
  76             thumbnail["url"] = self._BASE_URL + variant.find("url").text
  77             thumbnail["width"] = int(variant.find("width").text)
  78             thumbnail["height"] = int(variant.find("height").text)
  79             thumbnail["resolution"] = "%ix%i" % (thumbnail["width"], thumbnail["height"])
  80             thumbnails.append(thumbnail)
  81         thumbnails.sort(key = lambda x: x["width"] * x["height"], reverse=True)
  82         return thumbnails