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

1 files changed, 42 insertions, 32 deletions

diff --git a/youtube_dl/extractor/zdf.py b/youtube_dl/extractor/zdf.py
index 418509cb9..faed7ff7f 100644
--- a/youtube_dl/extractor/zdf.py
+++ b/youtube_dl/extractor/zdf.py
@@ -2,16 +2,14 @@ import re
 
 from .common import InfoExtractor
 from ..utils import (
+    determine_ext,
     ExtractorError,
-    unescapeHTML,
 )
 
+
 class ZDFIE(InfoExtractor):
-    _VALID_URL = r'^http://www\.zdf\.de\/ZDFmediathek\/(.*beitrag\/video\/)(?P<video_id>[^/\?]+)(?:\?.*)?'
-    _TITLE = r'<h1(?: class="beitragHeadline")?>(?P<title>.*)</h1>'
+    _VALID_URL = r'^http://www\.zdf\.de\/ZDFmediathek(?P<hash>#)?\/(.*beitrag\/video\/)(?P<video_id>[^/\?]+)(?:\?.*)?'
     _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):
         mobj = re.match(self._VALID_URL, url)
@@ -19,6 +17,9 @@ class ZDFIE(InfoExtractor):
             raise ExtractorError(u'Invalid URL: %s' % url)
         video_id = mobj.group('video_id')
 
+        if mobj.group('hash'):
+            url = url.replace(u'#', u'', 1)
+
         html = self._download_webpage(url, video_id)
         streams = [m.groupdict() for m in re.finditer(self._MEDIA_STREAM, html)]
         if streams is None:
@@ -27,39 +28,48 @@ class ZDFIE(InfoExtractor):
         # s['media_type'] == 'wstreaming' -> use 'Windows Media Player' and mms url
         # s['media_type'] == 'hstreaming' -> use 'Quicktime' and rtsp url
         # choose first/default media type and highest quality for now
-        for s in streams:        #find 300 - dsl1000mbit
-            if s['quality'] == '300' and s['media_type'] == 'wstreaming':
-                stream_=s
-                break
-        for s in streams:        #find veryhigh - dsl2000mbit
-            if s['quality'] == 'veryhigh' and s['media_type'] == 'wstreaming': # 'hstreaming' - rtsp is not working
-                stream_=s
-                break
-        if stream_ is None:
+        def stream_pref(s):
+            TYPE_ORDER = ['ostreaming', 'hstreaming', 'wstreaming']
+            try:
+                type_pref = TYPE_ORDER.index(s['media_type'])
+            except ValueError:
+                type_pref = 999
+
+            QUALITY_ORDER = ['veryhigh', '300']
+            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.')
+        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)
-        mobj = re.search(self._TITLE, html)
-        if mobj is None:
-            raise ExtractorError(u'Cannot extract title')
-        title = unescapeHTML(mobj.group('title'))
+        MMS_STREAM = r'href="(?P<video_url>mms://[^"]*)"'
+        RTSP_STREAM = r'(?P<video_url>rtsp://[^"]*.mp4)'
 
-        mobj = re.search(self._MMS_STREAM, media_link)
+        mobj = re.search(self._MEDIA_STREAM, media_link)
         if mobj is None:
-            mobj = re.search(self._RTSP_STREAM, media_link)
+            mobj = re.search(RTSP_STREAM, media_link)
             if mobj is None:
                 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)
-        if mobj is None:
-            raise ExtractorError(u'Cannot extract extention')
-        ext = mobj.group('ext')
+        title = self._html_search_regex(
+            r'<h1(?: class="beitragHeadline")?>(.*?)</h1>',
+            html, u'title')
 
-        return [{'id': video_id,
-                 'url': mms_url,
-                 'title': title,
-                 'ext': ext
-                 }]
+        return {
+            'id': video_id,
+            'url': video_url,
+            'title': title,
+            'ext': determine_ext(video_url)
+        }