youtube-dl

commit 3054ff0cbebc7969d918f0deee76cd37ea6f5b40
parent cd453d38bbd038076c1b301396c2f996e77e261f
Author: Jaime Marquínez Ferrándiz <jaime.marquinez.ferrandiz@gmail.com>
Date:   Sat,  8 Jun 2013 11:11:49 +0200

Merge pull request #853 from mc2avr/master

add ZDFIE

Diffstat:
M
youtube_dl/FileDownloader.py
 | 
35
+++++++++++++++++++++++++++++++++++
M
youtube_dl/InfoExtractors.py
 | 
59
+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

2 files changed, 94 insertions(+), 0 deletions(-)
diff --git a/youtube_dl/FileDownloader.py b/youtube_dl/FileDownloader.py
@@ -810,6 +810,37 @@ class FileDownloader(object):
             self.report_error(u'rtmpdump exited with code %d' % retval)
             return False
 
+    def _download_with_mplayer(self, filename, url):
+        self.report_destination(filename)
+        tmpfilename = self.temp_name(filename)
+
+        args = ['mplayer', '-really-quiet', '-vo', 'null', '-vc', 'dummy', '-dumpstream', '-dumpfile', tmpfilename, url]
+        # Check for mplayer first
+        try:
+            subprocess.call(['mplayer', '-h'], stdout=(open(os.path.devnull, 'w')), stderr=subprocess.STDOUT)
+        except (OSError, IOError):
+            self.report_error(u'MMS or RTSP download detected but "%s" could not be run' % args[0] )
+            return False
+
+        # Download using mplayer. 
+        retval = subprocess.call(args)
+        if retval == 0:
+            fsize = os.path.getsize(encodeFilename(tmpfilename))
+            self.to_screen(u'\r[%s] %s bytes' % (args[0], fsize))
+            self.try_rename(tmpfilename, filename)
+            self._hook_progress({
+                'downloaded_bytes': fsize,
+                'total_bytes': fsize,
+                'filename': filename,
+                'status': 'finished',
+            })
+            return True
+        else:
+            self.to_stderr(u"\n")
+            self.report_error(u'mplayer exited with code %d' % retval)
+            return False
+
+
     def _do_download(self, filename, info_dict):
         url = info_dict['url']
 
@@ -830,6 +861,10 @@ class FileDownloader(object):
                                                 info_dict.get('play_path', None),
                                                 info_dict.get('tc_url', None))
 
+        # Attempt to download using mplayer
+        if url.startswith('mms') or url.startswith('rtsp'):
+            return self._download_with_mplayer(filename, url)
+
         tmpfilename = self.temp_name(filename)
         stream = None
 
diff --git a/youtube_dl/InfoExtractors.py b/youtube_dl/InfoExtractors.py
@@ -4091,6 +4091,64 @@ class ARDIE(InfoExtractor):
             info["url"] = stream["video_url"]
         return [info]
 
+class ZDFIE(InfoExtractor):
+    _VALID_URL = r'^http://www\.zdf\.de\/ZDFmediathek\/(.*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".+>'
+    _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)
+        if mobj is None:
+            raise ExtractorError(u'Invalid URL: %s' % url)
+        video_id = mobj.group('video_id')
+
+        html = self._download_webpage(url, video_id)
+        streams = [m.groupdict() for m in re.finditer(self._MEDIA_STREAM, html)]
+        if streams is None:
+            raise ExtractorError(u'No media url found.')
+
+        # 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:
+            raise ExtractorError(u'No stream found.')
+
+        media_link = self._download_webpage(stream_['video_url'], video_id,'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'))
+
+        mobj = re.search(self._MMS_STREAM, media_link)
+        if mobj is None:
+            mobj = re.search(self._RTSP_STREAM, media_link)
+            if mobj is None:
+                raise ExtractorError(u'Cannot extract mms:// or rtsp:// URL')
+        mms_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')
+
+        return [{'id': video_id,
+                 'url': mms_url,
+                 'title': title,
+                 'ext': ext
+                 }]
+
 class TumblrIE(InfoExtractor):
     _VALID_URL = r'http://(?P<blog_name>.*?)\.tumblr\.com/((post)|(video))/(?P<id>\d*)/(.*?)'
 
@@ -4622,6 +4680,7 @@ def gen_extractors():
         SpiegelIE(),
         LiveLeakIE(),
         ARDIE(),
+        ZDFIE(),
         TumblrIE(),
         BandcampIE(),
         RedTubeIE(),