[phoenix] Add new extractor (Fixes #4036)

youtube_dl/extractor/__init__.py

@@ -280,6 +280,7 @@ from .orf import (
 from .parliamentliveuk import ParliamentLiveUKIE
 from .patreon import PatreonIE
 from .pbs import PBSIE
+from .phoenix import PhoenixIE
 from .photobucket import PhotobucketIE
 from .planetaplay import PlanetaPlayIE
 from .played import PlayedIE

youtube_dl/extractor/phoenix.py

@@ -0,0 +1,31 @@
+from __future__ import unicode_literals
+
+from .common import InfoExtractor
+from .zdf import extract_from_xml_url
+
+
+class PhoenixIE(InfoExtractor):
+    _VALID_URL = r'https?://(?:www\.)?phoenix\.de/content/(?P<id>[0-9]+)'
+    _TEST = {
+        'url': 'http://www.phoenix.de/content/884301',
+        'md5': 'ed249f045256150c92e72dbb70eadec6',
+        'info_dict': {
+            'id': '884301',
+            'ext': 'mp4',
+            'title': 'Michael Krons mit Hans-Werner Sinn',
+            'description': 'Im Dialog - Sa. 25.10.14, 00.00 - 00.35 Uhr',
+            'upload_date': '20141025',
+            'uploader': 'Im Dialog',
+        }
+    }
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+        webpage = self._download_webpage(url, video_id)
+
+        internal_id = self._search_regex(
+            r'<div class="phx_vod" id="phx_vod_([0-9]+)"',
+            webpage, 'internal video ID')
+
+        api_url = 'http://www.phoenix.de/php/zdfplayer-v1.3/data/beitragsDetails.php?ak=web&id=%s' % internal_id
+        return extract_from_xml_url(self, video_id, api_url)

youtube_dl/extractor/zdf.py

@@ -10,6 +10,82 @@ from ..utils import (
 )
 
 
+def extract_from_xml_url(ie, video_id, xml_url):
+    doc = ie._download_xml(
+        xml_url, video_id,
+        note='Downloading video info',
+        errnote='Failed to download video info')
+
+    title = doc.find('.//information/title').text
+    description = doc.find('.//information/detail').text
+    duration = int(doc.find('.//details/lengthSec').text)
+    uploader_node = doc.find('.//details/originChannelTitle')
+    uploader = None if uploader_node is None else uploader_node.text
+    uploader_id_node = doc.find('.//details/originChannelId')
+    uploader_id = None if uploader_id_node is None else uploader_id_node.text
+    upload_date = unified_strdate(doc.find('.//details/airtime').text)
+
+    def xml_to_format(fnode):
+        video_url = fnode.find('url').text
+        is_available = '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_node = fnode.find('./videoBitrate')
+        vbr = None if vbr_node is None else int(vbr_node.text) // 1000
+
+        width_node = fnode.find('./width')
+        width = None if width_node is None else int_or_none(width_node.text)
+        height_node = fnode.find('./height')
+        height = None if height_node is None else int_or_none(height_node.text)
+
+        format_note = ''
+        if not format_note:
+            format_note = None
+
+        return {
+            'format_id': format_id + '-' + quality,
+            'url': video_url,
+            'ext': ext,
+            'acodec': format_m.group('acodec'),
+            'vcodec': format_m.group('vcodec'),
+            'abr': abr,
+            'vbr': vbr,
+            'width': width,
+            'height': height,
+            '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)))
+    ie._sort_formats(formats)
+
+    return {
+        'id': video_id,
+        'title': title,
+        'description': description,
+        'duration': duration,
+        'uploader': uploader,
+        'uploader_id': uploader_id,
+        'upload_date': upload_date,
+        'formats': formats,
+    }
+
+
 class ZDFIE(InfoExtractor):
     _VALID_URL = r'^https?://www\.zdf\.de/ZDFmediathek(?P<hash>#)?/(.*beitrag/(?:video/)?)(?P<id>[0-9]+)(?:/[^/?]+)?(?:\?.*)?'
 
@@ -32,77 +108,4 @@ class ZDFIE(InfoExtractor):
         video_id = self._match_id(url)
 
         xml_url = 'http://www.zdf.de/ZDFmediathek/xmlservice/web/beitragsDetails?ak=web&id=%s' % video_id
-        doc = self._download_xml(
+        return extract_from_xml_url(self, video_id, xml_url)
-            xml_url, video_id,
-            note='Downloading video info',
-            errnote='Failed to download video info')
-
-        title = doc.find('.//information/title').text
-        description = doc.find('.//information/detail').text
-        duration = int(doc.find('.//details/lengthSec').text)
-        uploader_node = doc.find('.//details/originChannelTitle')
-        uploader = None if uploader_node is None else uploader_node.text
-        uploader_id_node = doc.find('.//details/originChannelId')
-        uploader_id = None if uploader_id_node is None else uploader_id_node.text
-        upload_date = unified_strdate(doc.find('.//details/airtime').text)
-
-        def xml_to_format(fnode):
-            video_url = fnode.find('url').text
-            is_available = '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_node = fnode.find('./videoBitrate')
-            vbr = None if vbr_node is None else int(vbr_node.text) // 1000
-
-            width_node = fnode.find('./width')
-            width = None if width_node is None else int_or_none(width_node.text)
-            height_node = fnode.find('./height')
-            height = None if height_node is None else int_or_none(height_node.text)
-
-            format_note = ''
-            if not format_note:
-                format_note = None
-
-            return {
-                'format_id': format_id + '-' + quality,
-                'url': video_url,
-                'ext': ext,
-                'acodec': format_m.group('acodec'),
-                'vcodec': format_m.group('vcodec'),
-                'abr': abr,
-                'vbr': vbr,
-                'width': width,
-                'height': height,
-                '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,
-            'description': description,
-            'duration': duration,
-            'uploader': uploader,
-            'uploader_id': uploader_id,
-            'upload_date': upload_date,
-            'formats': formats,
-        }