diff options
Diffstat (limited to 'youtube_dl/extractor/br.py')
| -rw-r--r-- | youtube_dl/extractor/br.py | 123 |
1 files changed, 60 insertions, 63 deletions
diff --git a/youtube_dl/extractor/br.py b/youtube_dl/extractor/br.py index 41e16f6cb..2256f47bb 100644 --- a/youtube_dl/extractor/br.py +++ b/youtube_dl/extractor/br.py @@ -1,82 +1,79 @@ # coding: utf-8 +from __future__ import unicode_literals + +import re from .common import InfoExtractor -class BRIE(InfoExtractor): - IE_DESC = u"Bayerischer Rundfunk Mediathek" - _VALID_URL = r"^https?://(?:www\.)?br\.de/mediathek/video/(?:sendungen/)?(?:[a-z0-9\-]+\.html)$" - _BASE_URL = u"http://www.br.de" +class BRIE(InfoExtractor): + IE_DESC = "Bayerischer Rundfunk Mediathek" + _VALID_URL = r"^https?://(?:www\.)?br\.de/mediathek/video/(?:sendungen/)?(?P<id>[a-z0-9\-]+)\.html$" + _BASE_URL = "http://www.br.de" - _TESTS = [ - { - u"url": u"http://www.br.de/mediathek/video/anselm-gruen-114.html", - u"file": u"2c8d81c5-6fb7-4a74-88d4-e768e5856532.mp4", - u"md5": u"c4f83cf0f023ba5875aba0bf46860df2", - u"info_dict": { - u"title": u"Feiern und Verzichten", - u"description": u"Anselm Grün: Feiern und Verzichten", - u"uploader": u"BR/Birgit Baier", - u"upload_date": u"20140301" - } + _TEST = { + "url": "http://www.br.de/mediathek/video/anselm-gruen-114.html", + "md5": "c4f83cf0f023ba5875aba0bf46860df2", + "info_dict": { + "id": "2c8d81c5-6fb7-4a74-88d4-e768e5856532", + "ext": "mp4", + "title": "Feiern und Verzichten", + "description": "Anselm Grün: Feiern und Verzichten", + "uploader": "BR/Birgit Baier", + "upload_date": "20140301" } - ] + } def _real_extract(self, url): - page = self._download_webpage(url, None) - xml_url = self._search_regex(r"return BRavFramework\.register\(BRavFramework\('avPlayer_(?:[a-f0-9-]{36})'\)\.setup\({dataURL:'(/mediathek/video/[a-z0-9/~_.-]+)'}\)\);", page, "XMLURL") + mobj = re.match(self._VALID_URL, url) + display_id = mobj.group('id') + page = self._download_webpage(url, display_id) + xml_url = self._search_regex( + r"return BRavFramework\.register\(BRavFramework\('avPlayer_(?:[a-f0-9-]{36})'\)\.setup\({dataURL:'(/mediathek/video/[a-z0-9/~_.-]+)'}\)\);", page, "XMLURL") xml = self._download_xml(self._BASE_URL + xml_url, None) - videos = [] - for xml_video in xml.findall("video"): - video = {} - video["id"] = xml_video.get("externalId") - video["title"] = xml_video.find("title").text - video["formats"] = self._extract_formats(xml_video.find("assets")) - video["thumbnails"] = self._extract_thumbnails(xml_video.find("teaserImage/variants")) - video["thumbnail"] = video["thumbnails"][0]["url"] - video["description"] = " ".join(xml_video.find("shareTitle").text.splitlines()) - video["uploader"] = xml_video.find("author").text - video["upload_date"] = "".join(reversed(xml_video.find("broadcastDate").text.split("."))) - video["webpage_url"] = xml_video.find("permalink").text - videos.append(video) + videos = [{ + "id": xml_video.get("externalId"), + "title": xml_video.find("title").text, + "formats": self._extract_formats(xml_video.find("assets")), + "thumbnails": self._extract_thumbnails(xml_video.find("teaserImage/variants")), + "description": " ".join(xml_video.find("shareTitle").text.splitlines()), + "uploader": xml_video.find("author").text, + "upload_date": "".join(reversed(xml_video.find("broadcastDate").text.split("."))), + "webpage_url": xml_video.find("permalink").text, + } for xml_video in xml.findall("video")] if len(videos) > 1: - self._downloader.report_warning(u'found multiple videos; please' - u'report this with the video URL to http://yt-dl.org/bug') + self._downloader.report_warning( + 'found multiple videos; please ' + 'report this with the video URL to http://yt-dl.org/bug') + if not videos: + raise ExtractorError('No video entries found') return videos[0] def _extract_formats(self, assets): - vformats = [] - for asset in assets.findall("asset"): - if asset.find("downloadUrl") is None: - continue - vformat = {} - vformat["url"] = asset.find("downloadUrl").text - vformat["ext"] = asset.find("mediaType").text - vformat["format_id"] = asset.get("type") - vformat["width"] = int(asset.find("frameWidth").text) - vformat["height"] = int(asset.find("frameHeight").text) - vformat["resolution"] = "%ix%i" % (vformat["width"], vformat["height"]) - vformat["tbr"] = int(asset.find("bitrateVideo").text) - vformat["abr"] = int(asset.find("bitrateAudio").text) - vformat["vcodec"] = asset.find("codecVideo").text - vformat["container"] = vformat["ext"] - vformat["filesize"] = int(asset.find("size").text) - vformat["preference"] = vformat["quality"] = -1 - vformat["format"] = "%s container with %i Kbps %s" % (vformat["container"], vformat["tbr"], vformat["vcodec"]) - vformats.append(vformat) - self._sort_formats(vformats) - return vformats + formats = [{ + "url": asset.find("downloadUrl").text, + "ext": asset.find("mediaType").text, + "format_id": asset.get("type"), + "width": int(asset.find("frameWidth").text), + "height": int(asset.find("frameHeight").text), + "tbr": int(asset.find("bitrateVideo").text), + "abr": int(asset.find("bitrateAudio").text), + "vcodec": asset.find("codecVideo").text, + "container": asset.find("mediaType").text, + "filesize": int(asset.find("size").text), + } for asset in assets.findall("asset") + if asset.find("downloadUrl") is not None] + + self._sort_formats(formats) + return formats def _extract_thumbnails(self, variants): - thumbnails = [] - for variant in variants.findall("variant"): - thumbnail = {} - thumbnail["url"] = self._BASE_URL + variant.find("url").text - thumbnail["width"] = int(variant.find("width").text) - thumbnail["height"] = int(variant.find("height").text) - thumbnail["resolution"] = "%ix%i" % (thumbnail["width"], thumbnail["height"]) - thumbnails.append(thumbnail) - thumbnails.sort(key = lambda x: x["width"] * x["height"], reverse=True) + thumbnails = [{ + "url": self._BASE_URL + variant.find("url").text, + "width": int(variant.find("width").text), + "height": int(variant.find("height").text), + } for variant in variants.findall("variant")] + thumbnails.sort(key=lambda x: x["width"] * x["height"], reverse=True) return thumbnails |