[nexx] Add support for shortcuts and relax domain id extraction

[youtube-dl] / youtube_dl / extractor / nexx.py

diff --git a/youtube_dl/extractor/nexx.py b/youtube_dl/extractor/nexx.py
index 60b42cb7d0de0296b2da2d3b3470b1aa66da048b..071879ba4f5317326ebed99ee59e8a6f4e6b4ce4 100644 (file)
--- a/youtube_dl/extractor/nexx.py
+++ b/youtube_dl/extractor/nexx.py
@@ -18,7 +18,13 @@ from ..utils import (
 
 
 class NexxIE(InfoExtractor):
-    _VALID_URL = r'https?://api\.nexx(?:\.cloud|cdn\.com)/v3/(?P<domain_id>\d+)/videos/byid/(?P<id>\d+)'
+    _VALID_URL = r'''(?x)
+                        (?:
+                            https?://api\.nexx(?:\.cloud|cdn\.com)/v3/(?P<domain_id>\d+)/videos/byid/|
+                            nexx:(?P<domain_id_s>\d+):
+                        )
+                        (?P<id>\d+)
+                    '''
     _TESTS = [{
         # movie
         'url': 'https://api.nexx.cloud/v3/748/videos/byid/128907',
@@ -62,8 +68,18 @@ class NexxIE(InfoExtractor):
     }, {
         'url': 'https://api.nexxcdn.com/v3/748/videos/byid/128907',
         'only_matching': True,
+    }, {
+        'url': 'nexx:748:128907',
+        'only_matching': True,
     }]
 
+    @staticmethod
+    def _extract_domain_id(webpage):
+        mobj = re.search(
+            r'<script\b[^>]+\bsrc=["\'](?:https?:)?//require\.nexx(?:\.cloud|cdn\.com)/(?P<id>\d+)',
+            webpage)
+        return mobj.group('id') if mobj else None
+
     @staticmethod
     def _extract_urls(webpage):
         # Reference:
@@ -72,18 +88,23 @@ class NexxIE(InfoExtractor):
         entries = []
 
         # JavaScript Integration
-        for domain_id, video_id in re.findall(
-                r'''(?isx)
-                    <script\b[^>]+\bsrc=["\']https?://require\.nexx(?:\.cloud|cdn\.com)/(\d+).+?
-                    onPLAYReady.+?
-                    _play\.init\s*\(.+?\s*,\s*(\d+)\s*,\s*.+?\)
-                ''', webpage):
-            entries.append('https://api.nexx.cloud/v3/%s/videos/byid/%s' % (domain_id, video_id))
+        domain_id = NexxIE._extract_domain_id(webpage)
+        if domain_id:
+            for video_id in re.findall(
+                    r'(?is)onPLAYReady.+?_play\.init\s*\(.+?\s*,\s*["\']?(\d+)',
+                    webpage):
+                entries.append(
+                    'https://api.nexx.cloud/v3/%s/videos/byid/%s'
+                    % (domain_id, video_id))
 
         # TODO: support more embed formats
 
         return entries
 
+    @staticmethod
+    def _extract_url(webpage):
+        return NexxIE._extract_urls(webpage)[0]
+
     def _handle_error(self, response):
         status = int_or_none(try_get(
             response, lambda x: x['metadata']['status']) or 200)
@@ -104,7 +125,8 @@ class NexxIE(InfoExtractor):
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
-        domain_id, video_id = mobj.group('domain_id', 'id')
+        domain_id = mobj.group('domain_id') or mobj.group('domain_id_s')
+        video_id = mobj.group('id')
 
         # Reverse engineered from JS code (see getDeviceID function)
         device_id = '%d:%d:%d%d' % (
@@ -219,3 +241,45 @@ class NexxIE(InfoExtractor):
                 video, lambda x: x['episodedata']['season'])),
             'formats': formats,
         }
+
+
+class NexxEmbedIE(InfoExtractor):
+    _VALID_URL = r'https?://embed\.nexx(?:\.cloud|cdn\.com)/\d+/(?P<id>[^/?#&]+)'
+    _TEST = {
+        'url': 'http://embed.nexx.cloud/748/KC1614647Z27Y7T?autoplay=1',
+        'md5': '16746bfc28c42049492385c989b26c4a',
+        'info_dict': {
+            'id': '161464',
+            'ext': 'mp4',
+            'title': 'Nervenkitzel Achterbahn',
+            'alt_title': 'Karussellbauer in Deutschland',
+            'description': 'md5:ffe7b1cc59a01f585e0569949aef73cc',
+            'release_year': 2005,
+            'creator': 'SPIEGEL TV',
+            'thumbnail': r're:^https?://.*\.jpg$',
+            'duration': 2761,
+            'timestamp': 1394021479,
+            'upload_date': '20140305',
+        },
+        'params': {
+            'format': 'bestvideo',
+            'skip_download': True,
+        },
+    }
+
+    @staticmethod
+    def _extract_urls(webpage):
+        # Reference:
+        # 1. https://nx-s.akamaized.net/files/201510/44.pdf
+
+        # iFrame Embed Integration
+        return [mobj.group('url') for mobj in re.finditer(
+            r'<iframe[^>]+\bsrc=(["\'])(?P<url>(?:https?:)?//embed\.nexx(?:\.cloud|cdn\.com)/\d+/(?:(?!\1).)+)\1',
+            webpage)]
+
+    def _real_extract(self, url):
+        embed_id = self._match_id(url)
+
+        webpage = self._download_webpage(url, embed_id)
+
+        return self.url_result(NexxIE._extract_url(webpage), ie=NexxIE.ie_key())