youtube_dl/extractor/br.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import re
   5
   6 from .common import InfoExtractor
   7
   8
   9 class BRIE(InfoExtractor):
  10     IE_DESC = "Bayerischer Rundfunk Mediathek"
  11     _VALID_URL = r"^https?://(?:www\.)?br\.de/mediathek/video/(?:sendungen/)?(?P<id>[a-z0-9\-]+)\.html$"
  12     _BASE_URL = "http://www.br.de"
  13
  14     _TEST = {
  15         "url": "http://www.br.de/mediathek/video/anselm-gruen-114.html",
  16         "md5": "c4f83cf0f023ba5875aba0bf46860df2",
  17         "info_dict": {
  18             "id": "2c8d81c5-6fb7-4a74-88d4-e768e5856532",
  19             "ext": "mp4",
  20             "title": "Feiern und Verzichten",
  21             "description": "Anselm Grün: Feiern und Verzichten",
  22             "uploader": "BR/Birgit Baier",
  23             "upload_date": "20140301"
  24         }
  25     }
  26
  27     def _real_extract(self, url):
  28         mobj = re.match(self._VALID_URL, url)
  29         display_id = mobj.group('id')
  30         page = self._download_webpage(url, display_id)
  31         xml_url = self._search_regex(
  32             r"return BRavFramework\.register\(BRavFramework\('avPlayer_(?:[a-f0-9-]{36})'\)\.setup\({dataURL:'(/mediathek/video/[a-z0-9/~_.-]+)'}\)\);", page, "XMLURL")
  33         xml = self._download_xml(self._BASE_URL + xml_url, None)
  34
  35         videos = [{
  36             "id": xml_video.get("externalId"),
  37             "title": xml_video.find("title").text,
  38             "formats": self._extract_formats(xml_video.find("assets")),
  39             "thumbnails": self._extract_thumbnails(xml_video.find("teaserImage/variants")),
  40             "description": " ".join(xml_video.find("shareTitle").text.splitlines()),
  41             "uploader": xml_video.find("author").text,
  42             "upload_date": "".join(reversed(xml_video.find("broadcastDate").text.split("."))),
  43             "webpage_url": xml_video.find("permalink").text,
  44         } for xml_video in xml.findall("video")]
  45
  46         if len(videos) > 1:
  47             self._downloader.report_warning(
  48                 'found multiple videos; please '
  49                 'report this with the video URL to http://yt-dl.org/bug')
  50         if not videos:
  51             raise ExtractorError('No video entries found')
  52         return videos[0]
  53
  54     def _extract_formats(self, assets):
  55         formats = [{
  56             "url": asset.find("downloadUrl").text,
  57             "ext": asset.find("mediaType").text,
  58             "format_id": asset.get("type"),
  59             "width": int(asset.find("frameWidth").text),
  60             "height": int(asset.find("frameHeight").text),
  61             "tbr": int(asset.find("bitrateVideo").text),
  62             "abr": int(asset.find("bitrateAudio").text),
  63             "vcodec": asset.find("codecVideo").text,
  64             "container": asset.find("mediaType").text,
  65             "filesize": int(asset.find("size").text),
  66         } for asset in assets.findall("asset")
  67             if asset.find("downloadUrl") is not None]
  68
  69         self._sort_formats(formats)
  70         return formats
  71
  72     def _extract_thumbnails(self, variants):
  73         thumbnails = [{
  74             "url": self._BASE_URL + variant.find("url").text,
  75             "width": int(variant.find("width").text),
  76             "height": int(variant.find("height").text),
  77         } for variant in variants.findall("variant")]
  78         thumbnails.sort(key=lambda x: x["width"] * x["height"], reverse=True)
  79         return thumbnails