[faz] fix extraction and add support for Perform Group embeds(fixes #14714)

youtube_dl/extractor/extractors.py

@@ -789,6 +789,7 @@
 from .pbs import PBSIE
 from .pearvideo import PearVideoIE
 from .people import PeopleIE
+from .performgroup import PerformGroupIE
 from .periscope import (
     PeriscopeIE,
     PeriscopeUserIE,

youtube_dl/extractor/faz.py

@@ -1,7 +1,10 @@
 # coding: utf-8
 from __future__ import unicode_literals
 
+import re
+
 from .common import InfoExtractor
+from ..compat import compat_etree_fromstring
 from ..utils import (
     xpath_element,
     xpath_text,
@@ -43,10 +46,15 @@ def _real_extract(self, url):
 
         webpage = self._download_webpage(url, video_id)
         description = self._og_search_description(webpage)
-        config_xml_url = self._search_regex(
-            r'videoXMLURL\s*=\s*"([^"]+)', webpage, 'config xml url')
-        config = self._download_xml(
-            config_xml_url, video_id, 'Downloading config xml')
+        media = self._html_search_regex(
+            r"data-videojs-media='([^']+)",
+            webpage, 'media')
+        if media == 'extern':
+            perform_url = self._search_regex(
+                r"<iframe[^>]+?src='((?:http:)?//player\.performgroup\.com/eplayer/eplayer\.html#/?[0-9a-f]{26}\.[0-9a-z]{26})",
+                webpage, 'perform url')
+            return self.url_result(perform_url)
+        config = compat_etree_fromstring(media)
 
         encodings = xpath_element(config, 'ENCODINGS', 'encodings', True)
         formats = []
@@ -55,12 +63,24 @@ def _real_extract(self, url):
             if encoding is not None:
                 encoding_url = xpath_text(encoding, 'FILENAME')
                 if encoding_url:
-                    formats.append({
+                    tbr = xpath_text(encoding, 'AVERAGEBITRATE', 1000)
+                    if tbr:
+                        tbr = int_or_none(tbr.replace(',', '.'))
+                    f = {
                         'url': encoding_url,
                         'format_id': code.lower(),
                         'quality': pref,
-                        'tbr': int_or_none(xpath_text(encoding, 'AVERAGEBITRATE')),
-                    })
+                        'tbr': tbr,
+                        'vcodec': xpath_text(encoding, 'CODEC'),
+                    }
+                    mobj = re.search(r'(\d+)x(\d+)_(\d+)\.mp4', encoding_url)
+                    if mobj:
+                        f.update({
+                            'width': int(mobj.group(1)),
+                            'height': int(mobj.group(2)),
+                            'tbr': tbr or int(mobj.group(3)),
+                        })
+                    formats.append(f)
         self._sort_formats(formats)
 
         return {

youtube_dl/extractor/performgroup.py

@@ -0,0 +1,83 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+import re
+
+from .common import InfoExtractor
+from ..utils import int_or_none
+
+
+class PerformGroupIE(InfoExtractor):
+    _VALID_URL = r'https?://player\.performgroup\.com/eplayer(?:/eplayer\.html|\.js)#/?(?P<id>[0-9a-f]{26})\.(?P<auth_token>[0-9a-z]{26})'
+    _TESTS = [{
+        # http://www.faz.net/aktuell/sport/fussball/wm-2018-playoffs-schweiz-besiegt-nordirland-1-0-15286104.html
+        'url': 'http://player.performgroup.com/eplayer/eplayer.html#d478c41c5d192f56b9aa859de8.1w4crrej5w14e1ed4s1ce4ykab',
+        'md5': '259cb03d142e2e52471e8837ecacb29f',
+        'info_dict': {
+            'id': 'xgrwobuzumes1lwjxtcdpwgxd',
+            'ext': 'mp4',
+            'title': 'Liga MX: Keine Einsicht nach Horrorfoul',
+            'description': 'md5:7cd3b459c82725b021e046ab10bf1c5b',
+            'timestamp': 1511533477,
+            'upload_date': '20171124',
+        }
+    }]
+
+    def _call_api(self, service, auth_token, content_id, referer_url):
+        return self._download_json(
+            'http://ep3.performfeeds.com/ep%s/%s/%s/' % (service, auth_token, content_id),
+            content_id, headers={
+                'Referer': referer_url,
+                'Origin': 'http://player.performgroup.com',
+            }, query={
+                '_fmt': 'json',
+            })
+
+    def _real_extract(self, url):
+        player_id, auth_token = re.search(self._VALID_URL, url).groups()
+        bootstrap = self._call_api('bootstrap', auth_token, player_id, url)
+        video = bootstrap['config']['dataSource']['sourceItems'][0]['videos'][0]
+        video_id = video['uuid']
+        vod = self._call_api('vod', auth_token, video_id, url)
+        media = vod['videos']['video'][0]['media']
+
+        formats = []
+        hls_url = media.get('hls', {}).get('url')
+        if hls_url:
+            formats.extend(self._extract_m3u8_formats(hls_url, video_id, 'mp4', 'm3u8_native', m3u8_id='hls', fatal=False))
+
+        hds_url = media.get('hds', {}).get('url')
+        if hds_url:
+            formats.extend(self._extract_f4m_formats(hds_url + '?hdcore', video_id, f4m_id='hds', fatal=False))
+
+        for c in media.get('content', []):
+            c_url = c.get('url')
+            if not c_url:
+                continue
+            tbr = int_or_none(c.get('bitrate'), 1000)
+            format_id = 'http'
+            if tbr:
+                format_id += '-%d' % tbr
+            formats.append({
+                'format_id': format_id,
+                'url': c_url,
+                'tbr': tbr,
+                'width': int_or_none(c.get('width')),
+                'height': int_or_none(c.get('height')),
+                'filesize': int_or_none(c.get('fileSize')),
+                'vcodec': c.get('type'),
+                'fps': int_or_none(c.get('videoFrameRate')),
+                'vbr': int_or_none(c.get('videoRate'), 1000),
+                'abr': int_or_none(c.get('audioRate'), 1000),
+            })
+        self._sort_formats(formats)
+
+        return {
+            'id': video_id,
+            'title': video['title'],
+            'description': video.get('description'),
+            'thumbnail': video.get('poster'),
+            'duration': int_or_none(video.get('duration')),
+            'timestamp': int_or_none(video.get('publishedTime'), 1000),
+            'formats': formats,
+        }