debian/NEWS: Add link to my webpage to give more details.

[youtubedl] / youtube_dl / extractor / zdf.py

diff --git a/youtube_dl/extractor/zdf.py b/youtube_dl/extractor/zdf.py
index 418509cb96df7822b6120909fa440e2186c677c4..829f002cf02f9c908a5057ab3c6b20f520e2f2ea 100644 (file)
--- a/youtube_dl/extractor/zdf.py
+++ b/youtube_dl/extractor/zdf.py
@@ -1,65 +1,111 @@
+# coding: utf-8
+

 import re
 
 from .common import InfoExtractor
 from ..utils import (
 import re
 
 from .common import InfoExtractor
 from ..utils import (

-    ExtractorError,
-    unescapeHTML,
+    int_or_none,
+    unified_strdate,

 )
 
 )
 

+

 class ZDFIE(InfoExtractor):
 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)'
+    _VALID_URL = r'^https?://www\.zdf\.de/ZDFmediathek(?P<hash>#)?/(.*beitrag/(?:video/)?)(?P<video_id>[0-9]+)(?:/[^/?]+)?(?:\?.*)?'
+
+    _TEST = {
+        u"url": u"http://www.zdf.de/ZDFmediathek/beitrag/video/2037704/ZDFspezial---Ende-des-Machtpokers--?bc=sts;stt",
+        u"file": u"2037704.webm",
+        u"info_dict": {
+            u"upload_date": u"20131127",
+            u"description": u"Union und SPD haben sich auf einen Koalitionsvertrag geeinigt. Aber was bedeutet das für die Bürger? Sehen Sie hierzu das ZDFspezial \"Ende des Machtpokers - Große Koalition für Deutschland\".",
+            u"uploader": u"spezial",
+            u"title": u"ZDFspezial - Ende des Machtpokers"
+        },
+        u"skip": u"Videos on ZDF.de are depublicised in short order",
+    }

 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
 
     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')
 
         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
-                 }]
+        xml_url = u'http://www.zdf.de/ZDFmediathek/xmlservice/web/beitragsDetails?ak=web&id=%s' % video_id
+        doc = self._download_xml(
+            xml_url, video_id,
+            note=u'Downloading video info',
+            errnote=u'Failed to download video info')
+
+        title = doc.find('.//information/title').text
+        description = doc.find('.//information/detail').text
+        uploader_node = doc.find('.//details/originChannelTitle')
+        uploader = None if uploader_node is None else uploader_node.text
+        duration_str = doc.find('.//details/length').text
+        duration_m = re.match(r'''(?x)^
+            (?P<hours>[0-9]{2})
+            :(?P<minutes>[0-9]{2})
+            :(?P<seconds>[0-9]{2})
+            (?:\.(?P<ms>[0-9]+)?)
+            ''', duration_str)
+        duration = (
+            (
+                (int(duration_m.group('hours')) * 60 * 60) +
+                (int(duration_m.group('minutes')) * 60) +
+                int(duration_m.group('seconds'))
+            )
+            if duration_m
+            else None
+        )
+        upload_date = unified_strdate(doc.find('.//details/airtime').text)
+
+        def xml_to_format(fnode):
+            video_url = fnode.find('url').text
+            is_available = u'http://www.metafilegenerator' not in video_url
+
+            format_id = fnode.attrib['basetype']
+            format_m = re.match(r'''(?x)
+                (?P<vcodec>[^_]+)_(?P<acodec>[^_]+)_(?P<container>[^_]+)_
+                (?P<proto>[^_]+)_(?P<index>[^_]+)_(?P<indexproto>[^_]+)
+            ''', format_id)
+
+            ext = format_m.group('container')
+            proto = format_m.group('proto').lower()
+
+            quality = fnode.find('./quality').text
+            abr = int(fnode.find('./audioBitrate').text) // 1000
+            vbr = int(fnode.find('./videoBitrate').text) // 1000
+
+            format_note = u''
+            if not format_note:
+                format_note = None
+
+            return {
+                'format_id': format_id + u'-' + quality,
+                'url': video_url,
+                'ext': ext,
+                'acodec': format_m.group('acodec'),
+                'vcodec': format_m.group('vcodec'),
+                'abr': abr,
+                'vbr': vbr,
+                'width': int_or_none(fnode.find('./width').text),
+                'height': int_or_none(fnode.find('./height').text),
+                'filesize': int_or_none(fnode.find('./filesize').text),
+                'format_note': format_note,
+                'protocol': proto,
+                '_available': is_available,
+            }
+
+        format_nodes = doc.findall('.//formitaeten/formitaet')
+        formats = list(filter(
+            lambda f: f['_available'],
+            map(xml_to_format, format_nodes)))
+
+        self._sort_formats(formats)
+
+        return {
+            'id': video_id,
+            'title': title,
+            'formats': formats,
+            'description': description,
+            'uploader': uploader,
+            'duration': duration,
+            'upload_date': upload_date,
+        }