[zdf] Add support for hash URLs and simplify (#1518)

This commit is contained in:
Philipp Hagemeister 2013-09-28 15:43:34 +02:00
parent c53f9d30c8
commit 0b7c2485b6

View File

@ -2,16 +2,14 @@
from .common import InfoExtractor from .common import InfoExtractor
from ..utils import ( from ..utils import (
determine_ext,
ExtractorError, ExtractorError,
unescapeHTML,
) )
class ZDFIE(InfoExtractor): class ZDFIE(InfoExtractor):
_VALID_URL = r'^http://www\.zdf\.de\/ZDFmediathek\/(.*beitrag\/video\/)(?P<video_id>[^/\?]+)(?:\?.*)?' _VALID_URL = r'^http://www\.zdf\.de\/ZDFmediathek(?P<hash>#)?\/(.*beitrag\/video\/)(?P<video_id>[^/\?]+)(?:\?.*)?'
_TITLE = r'<h1(?: class="beitragHeadline")?>(?P<title>.*)</h1>'
_MEDIA_STREAM = r'<a href="(?P<video_url>.+(?P<media_type>.streaming).+/zdf/(?P<quality>[^\/]+)/[^"]*)".+class="play".+>' _MEDIA_STREAM = r'<a href="(?P<video_url>.+(?P<media_type>.streaming).+/zdf/(?P<quality>[^\/]+)/[^"]*)".+class="play".+>'
_MMS_STREAM = r'href="(?P<video_url>mms://[^"]*)"'
_RTSP_STREAM = r'(?P<video_url>rtsp://[^"]*.mp4)'
def _real_extract(self, url): def _real_extract(self, url):
mobj = re.match(self._VALID_URL, url) mobj = re.match(self._VALID_URL, url)
@ -19,6 +17,9 @@ def _real_extract(self, url):
raise ExtractorError(u'Invalid URL: %s' % url) raise ExtractorError(u'Invalid URL: %s' % url)
video_id = mobj.group('video_id') video_id = mobj.group('video_id')
if mobj.group('hash'):
url = url.replace(u'#', u'', 1)
html = self._download_webpage(url, video_id) html = self._download_webpage(url, video_id)
streams = [m.groupdict() for m in re.finditer(self._MEDIA_STREAM, html)] streams = [m.groupdict() for m in re.finditer(self._MEDIA_STREAM, html)]
if streams is None: if streams is None:
@ -27,39 +28,48 @@ def _real_extract(self, url):
# s['media_type'] == 'wstreaming' -> use 'Windows Media Player' and mms url # s['media_type'] == 'wstreaming' -> use 'Windows Media Player' and mms url
# s['media_type'] == 'hstreaming' -> use 'Quicktime' and rtsp url # s['media_type'] == 'hstreaming' -> use 'Quicktime' and rtsp url
# choose first/default media type and highest quality for now # choose first/default media type and highest quality for now
for s in streams: #find 300 - dsl1000mbit def stream_pref(s):
if s['quality'] == '300' and s['media_type'] == 'wstreaming': TYPE_ORDER = ['ostreaming', 'hstreaming', 'wstreaming']
stream_=s try:
break type_pref = TYPE_ORDER.index(s['media_type'])
for s in streams: #find veryhigh - dsl2000mbit except ValueError:
if s['quality'] == 'veryhigh' and s['media_type'] == 'wstreaming': # 'hstreaming' - rtsp is not working type_pref = 999
stream_=s
break QUALITY_ORDER = ['veryhigh', '300']
if stream_ is None: try:
quality_pref = QUALITY_ORDER.index(s['quality'])
except ValueError:
quality_pref = 999
return (type_pref, quality_pref)
sorted_streams = sorted(streams, key=stream_pref)
if not sorted_streams:
raise ExtractorError(u'No stream found.') raise ExtractorError(u'No stream found.')
stream = sorted_streams[0]
media_link = self._download_webpage(stream_['video_url'], video_id,'Get stream URL') media_link = self._download_webpage(
stream['video_url'],
video_id,
u'Get stream URL')
self.report_extraction(video_id) MMS_STREAM = r'href="(?P<video_url>mms://[^"]*)"'
mobj = re.search(self._TITLE, html) RTSP_STREAM = r'(?P<video_url>rtsp://[^"]*.mp4)'
mobj = re.search(self._MEDIA_STREAM, media_link)
if mobj is None: if mobj is None:
raise ExtractorError(u'Cannot extract title') mobj = re.search(RTSP_STREAM, media_link)
title = unescapeHTML(mobj.group('title'))
mobj = re.search(self._MMS_STREAM, media_link)
if mobj is None:
mobj = re.search(self._RTSP_STREAM, media_link)
if mobj is None: if mobj is None:
raise ExtractorError(u'Cannot extract mms:// or rtsp:// URL') raise ExtractorError(u'Cannot extract mms:// or rtsp:// URL')
mms_url = mobj.group('video_url') video_url = mobj.group('video_url')
mobj = re.search('(.*)[.](?P<ext>[^.]+)', mms_url) title = self._html_search_regex(
if mobj is None: r'<h1(?: class="beitragHeadline")?>(.*?)</h1>',
raise ExtractorError(u'Cannot extract extention') html, u'title')
ext = mobj.group('ext')
return [{'id': video_id, return {
'url': mms_url, 'id': video_id,
'title': title, 'url': video_url,
'ext': ext 'title': title,
}] 'ext': determine_ext(video_url)
}