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 from ..utils import ExtractorError
   8
   9
  10 class BRIE(InfoExtractor):
  11     IE_DESC = "Bayerischer Rundfunk Mediathek"
  12     _VALID_URL = r"^https?://(?:www\.)?br\.de/mediathek/video/(?:sendungen/)?(?:[a-z0-9\-/]+/)?(?P<id>[a-z0-9\-]+)\.html$"
  13     _BASE_URL = "http://www.br.de"
  14
  15     _TESTS = [
  16         {
  17             "url": "http://www.br.de/mediathek/video/anselm-gruen-114.html",
  18             "md5": "c4f83cf0f023ba5875aba0bf46860df2",
  19             "info_dict": {
  20                 "id": "2c8d81c5-6fb7-4a74-88d4-e768e5856532",
  21                 "ext": "mp4",
  22                 "title": "Feiern und Verzichten",
  23                 "description": "Anselm Grün: Feiern und Verzichten",
  24                 "uploader": "BR/Birgit Baier",
  25                 "upload_date": "20140301"
  26             }
  27         },
  28         {
  29             "url": "http://www.br.de/mediathek/video/sendungen/unter-unserem-himmel/unter-unserem-himmel-alpen-ueber-den-pass-100.html",
  30             "md5": "ab451b09d861dbed7d7cc9ab0be19ebe",
  31             "info_dict": {
  32                 "id": "2c060e69-3a27-4e13-b0f0-668fac17d812",
  33                 "ext": "mp4",
  34                 "title": "Über den Pass",
  35                 "description": "Die Eroberung der Alpen: Über den Pass",
  36                 "uploader": None,
  37                 "upload_date": None
  38             }
  39         }
  40     ]
  41
  42     def _real_extract(self, url):
  43         mobj = re.match(self._VALID_URL, url)
  44         display_id = mobj.group('id')
  45         page = self._download_webpage(url, display_id)
  46         xml_url = self._search_regex(
  47             r"return BRavFramework\.register\(BRavFramework\('avPlayer_(?:[a-f0-9-]{36})'\)\.setup\({dataURL:'(/mediathek/video/[a-z0-9/~_.-]+)'}\)\);", page, "XMLURL")
  48         xml = self._download_xml(self._BASE_URL + xml_url, None)
  49
  50         videos = []
  51         for xml_video in xml.findall("video"):
  52             video = {
  53                 "id": xml_video.get("externalId"),
  54                 "title": xml_video.find("title").text,
  55                 "formats": self._extract_formats(xml_video.find("assets")),
  56                 "thumbnails": self._extract_thumbnails(xml_video.find("teaserImage/variants")),
  57                 "description": " ".join(xml_video.find("shareTitle").text.splitlines()),
  58                 "webpage_url": xml_video.find("permalink").text
  59             }
  60             if xml_video.find("author").text:
  61                 video["uploader"] = xml_video.find("author").text
  62             if xml_video.find("broadcastDate").text:
  63                 video["upload_date"] =  "".join(reversed(xml_video.find("broadcastDate").text.split(".")))
  64             videos.append(video)
  65
  66         if len(videos) > 1:
  67             self._downloader.report_warning(
  68                 'found multiple videos; please '
  69                 'report this with the video URL to http://yt-dl.org/bug')
  70         if not videos:
  71             raise ExtractorError('No video entries found')
  72         return videos[0]
  73
  74     def _extract_formats(self, assets):
  75         formats = [{
  76             "url": asset.find("downloadUrl").text,
  77             "ext": asset.find("mediaType").text,
  78             "format_id": asset.get("type"),
  79             "width": int(asset.find("frameWidth").text),
  80             "height": int(asset.find("frameHeight").text),
  81             "tbr": int(asset.find("bitrateVideo").text),
  82             "abr": int(asset.find("bitrateAudio").text),
  83             "vcodec": asset.find("codecVideo").text,
  84             "container": asset.find("mediaType").text,
  85             "filesize": int(asset.find("size").text),
  86         } for asset in assets.findall("asset")
  87             if asset.find("downloadUrl") is not None]
  88
  89         self._sort_formats(formats)
  90         return formats
  91
  92     def _extract_thumbnails(self, variants):
  93         thumbnails = [{
  94             "url": self._BASE_URL + variant.find("url").text,
  95             "width": int(variant.find("width").text),
  96             "height": int(variant.find("height").text),
  97         } for variant in variants.findall("variant")]
  98         thumbnails.sort(key=lambda x: x["width"] * x["height"], reverse=True)
  99         return thumbnails