[br] Allow '/' in URL, allow empty author + broadcastDate fields

* Allow URLs that have a 'subdirectory' before the actual program name, e.g.
  'xyz/xyz-episode-1'.
* The author and broadcastDate fields in the XML file may be empty.
* Add test case for the two problems above.
This commit is contained in:
David Triendl 2014-03-13 14:01:20 +01:00
parent 98ff9d82d4
commit c21215b421

View File

@ -9,10 +9,11 @@ from ..utils import ExtractorError
class BRIE(InfoExtractor): class BRIE(InfoExtractor):
IE_DESC = "Bayerischer Rundfunk Mediathek" IE_DESC = "Bayerischer Rundfunk Mediathek"
_VALID_URL = r"^https?://(?:www\.)?br\.de/mediathek/video/(?:sendungen/)?(?P<id>[a-z0-9\-]+)\.html$" _VALID_URL = r"^https?://(?:www\.)?br\.de/mediathek/video/(?:sendungen/)?(?:[a-z0-9\-/]+/)?(?P<id>[a-z0-9\-]+)\.html$"
_BASE_URL = "http://www.br.de" _BASE_URL = "http://www.br.de"
_TEST = { _TESTS = [
{
"url": "http://www.br.de/mediathek/video/anselm-gruen-114.html", "url": "http://www.br.de/mediathek/video/anselm-gruen-114.html",
"md5": "c4f83cf0f023ba5875aba0bf46860df2", "md5": "c4f83cf0f023ba5875aba0bf46860df2",
"info_dict": { "info_dict": {
@ -23,7 +24,20 @@ class BRIE(InfoExtractor):
"uploader": "BR/Birgit Baier", "uploader": "BR/Birgit Baier",
"upload_date": "20140301" "upload_date": "20140301"
} }
},
{
"url": "http://www.br.de/mediathek/video/sendungen/unter-unserem-himmel/unter-unserem-himmel-alpen-ueber-den-pass-100.html",
"md5": "ab451b09d861dbed7d7cc9ab0be19ebe",
"info_dict": {
"id": "2c060e69-3a27-4e13-b0f0-668fac17d812",
"ext": "mp4",
"title": "Über den Pass",
"description": "Die Eroberung der Alpen: Über den Pass",
"uploader": None,
"upload_date": None
} }
}
]
def _real_extract(self, url): def _real_extract(self, url):
mobj = re.match(self._VALID_URL, url) mobj = re.match(self._VALID_URL, url)
@ -33,16 +47,21 @@ class BRIE(InfoExtractor):
r"return BRavFramework\.register\(BRavFramework\('avPlayer_(?:[a-f0-9-]{36})'\)\.setup\({dataURL:'(/mediathek/video/[a-z0-9/~_.-]+)'}\)\);", page, "XMLURL") 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) xml = self._download_xml(self._BASE_URL + xml_url, None)
videos = [{ videos = []
for xml_video in xml.findall("video"):
video = {
"id": xml_video.get("externalId"), "id": xml_video.get("externalId"),
"title": xml_video.find("title").text, "title": xml_video.find("title").text,
"formats": self._extract_formats(xml_video.find("assets")), "formats": self._extract_formats(xml_video.find("assets")),
"thumbnails": self._extract_thumbnails(xml_video.find("teaserImage/variants")), "thumbnails": self._extract_thumbnails(xml_video.find("teaserImage/variants")),
"description": " ".join(xml_video.find("shareTitle").text.splitlines()), "description": " ".join(xml_video.find("shareTitle").text.splitlines()),
"uploader": xml_video.find("author").text, "webpage_url": xml_video.find("permalink").text
"upload_date": "".join(reversed(xml_video.find("broadcastDate").text.split("."))), }
"webpage_url": xml_video.find("permalink").text, if xml_video.find("author").text:
} for xml_video in xml.findall("video")] video["uploader"] = xml_video.find("author").text
if xml_video.find("broadcastDate").text:
video["upload_date"] = "".join(reversed(xml_video.find("broadcastDate").text.split(".")))
videos.append(video)
if len(videos) > 1: if len(videos) > 1:
self._downloader.report_warning( self._downloader.report_warning(