Merge pull request #8754 from remitamine/5min

update 5min related web sites info extraction and add support for Aol features.

youtube_dl/extractor/__init__.py

@@ -23,7 +23,10 @@ from .alphaporno import AlphaPornoIE
 from .animeondemand import AnimeOnDemandIE
 from .anitube import AnitubeIE
 from .anysex import AnySexIE
-from .aol import AolIE
+from .aol import (
+    AolIE,
+    AolFeaturesIE,
+)
 from .allocine import AllocineIE
 from .aparat import AparatIE
 from .appleconnect import AppleConnectIE

youtube_dl/extractor/aol.py

@@ -1,24 +1,11 @@
 from __future__ import unicode_literals
 
-import re
-
 from .common import InfoExtractor
 
 
 class AolIE(InfoExtractor):
     IE_NAME = 'on.aol.com'
-    _VALID_URL = r'''(?x)
+    _VALID_URL = r'(?:aol-video:|http://on\.aol\.com/video/.*-)(?P<id>[0-9]+)(?:$|\?)'
-        (?:
-            aol-video:|
-            http://on\.aol\.com/
-            (?:
-                video/.*-|
-                playlist/(?P<playlist_display_id>[^/?#]+?)-(?P<playlist_id>[0-9]+)[?#].*_videoid=
-            )
-        )
-        (?P<id>[0-9]+)
-        (?:$|\?)
-    '''
 
     _TESTS = [{
         'url': 'http://on.aol.com/video/u-s--official-warns-of-largest-ever-irs-phone-scam-518167793?icid=OnHomepageC2Wide_MustSee_Img',
@@ -29,42 +16,31 @@ class AolIE(InfoExtractor):
             'title': 'U.S. Official Warns Of \'Largest Ever\' IRS Phone Scam',
         },
         'add_ie': ['FiveMin'],
-    }, {
-        'url': 'http://on.aol.com/playlist/brace-yourself---todays-weirdest-news-152147?icid=OnHomepageC4_Omg_Img#_videoid=518184316',
-        'info_dict': {
-            'id': '152147',
-            'title': 'Brace Yourself - Today\'s Weirdest News',
-        },
-        'playlist_mincount': 10,
     }]
 
     def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
+        video_id = self._match_id(url)
-        video_id = mobj.group('id')
-        playlist_id = mobj.group('playlist_id')
-        if not playlist_id or self._downloader.params.get('noplaylist'):
         return self.url_result('5min:%s' % video_id)
 
-        self.to_screen('Downloading playlist %s - add --no-playlist to just download video %s' % (playlist_id, video_id))
 
-        webpage = self._download_webpage(url, playlist_id)
+class AolFeaturesIE(InfoExtractor):
-        title = self._html_search_regex(
+    IE_NAME = 'features.aol.com'
-            r'<h1 class="video-title[^"]*">(.+?)</h1>', webpage, 'title')
+    _VALID_URL = r'http://features\.aol\.com/video/(?P<id>[^/?#]+)'
-        playlist_html = self._search_regex(
-            r"(?s)<ul\s+class='video-related[^']*'>(.*?)</ul>", webpage,
-            'playlist HTML')
-        entries = [{
-            '_type': 'url',
-            'url': 'aol-video:%s' % m.group('id'),
-            'ie_key': 'Aol',
-        } for m in re.finditer(
-            r"<a\s+href='.*videoid=(?P<id>[0-9]+)'\s+class='video-thumb'>",
-            playlist_html)]
 
-        return {
+    _TESTS = [{
-            '_type': 'playlist',
+        'url': 'http://features.aol.com/video/behind-secret-second-careers-late-night-talk-show-hosts',
-            'id': playlist_id,
+        'md5': '7db483bb0c09c85e241f84a34238cc75',
-            'display_id': mobj.group('playlist_display_id'),
+        'info_dict': {
-            'title': title,
+            'id': '519507715',
-            'entries': entries,
+            'ext': 'mp4',
-        }
+            'title': 'What To Watch - February 17, 2016',
+        },
+        'add_ie': ['FiveMin'],
+    }]
+
+    def _real_extract(self, url):
+        display_id = self._match_id(url)
+        webpage = self._download_webpage(url, display_id)
+        return self.url_result(self._search_regex(
+            r'<script type="text/javascript" src="(https?://[^/]*?5min\.com/Scripts/PlayerSeed\.js[^"]+)"',
+            webpage, '5min embed url'), 'FiveMin')

youtube_dl/extractor/engadget.py

@@ -1,21 +1,13 @@
 from __future__ import unicode_literals
 
-import re
-
 from .common import InfoExtractor
-from ..utils import (
-    url_basename,
-)
 
 
 class EngadgetIE(InfoExtractor):
-    _VALID_URL = r'''(?x)https?://www.engadget.com/
+    _VALID_URL = r'https?://www.engadget.com/video/(?P<id>\d+)'
-        (?:video(?:/5min)?/(?P<id>\d+)|
-            [\d/]+/.*?)
-        '''
 
     _TEST = {
-        'url': 'http://www.engadget.com/video/5min/518153925/',
+        'url': 'http://www.engadget.com/video/518153925/',
         'md5': 'c6820d4828a5064447a4d9fc73f312c9',
         'info_dict': {
             'id': '518153925',
@@ -27,15 +19,4 @@ class EngadgetIE(InfoExtractor):
 
     def _real_extract(self, url):
         video_id = self._match_id(url)
-
-        if video_id is not None:
         return self.url_result('5min:%s' % video_id)
-        else:
-            title = url_basename(url)
-            webpage = self._download_webpage(url, title)
-            ids = re.findall(r'<iframe[^>]+?playList=(\d+)', webpage)
-            return {
-                '_type': 'playlist',
-                'title': title,
-                'entries': [self.url_result('5min:%s' % vid) for vid in ids]
-            }

youtube_dl/extractor/fivemin.py

@@ -1,5 +1,7 @@
 from __future__ import unicode_literals
 
+import re
+
 from .common import InfoExtractor
 from ..compat import (
     compat_urllib_parse,
@@ -16,12 +18,7 @@ from ..utils import (
 
 class FiveMinIE(InfoExtractor):
     IE_NAME = '5min'
-    _VALID_URL = r'''(?x)
+    _VALID_URL = r'(?:5min:(?P<id>\d+)(?::(?P<sid>\d+))?|https?://[^/]*?5min\.com/Scripts/PlayerSeed\.js\?(?P<query>.*))'
-        (?:https?://[^/]*?5min\.com/Scripts/PlayerSeed\.js\?(?:.*?&)?playList=|
-            https?://(?:(?:massively|www)\.)?joystiq\.com/video/|
-            5min:)
-        (?P<id>\d+)
-        '''
 
     _TESTS = [
         {
@@ -45,6 +42,7 @@ class FiveMinIE(InfoExtractor):
                 'title': 'How to Make a Next-Level Fruit Salad',
                 'duration': 184,
             },
+            'skip': 'no longer available',
         },
     ]
     _ERRORS = {
@@ -91,20 +89,33 @@ class FiveMinIE(InfoExtractor):
     }
 
     def _real_extract(self, url):
-        video_id = self._match_id(url)
+        mobj = re.match(self._VALID_URL, url)
+        video_id = mobj.group('id')
+        sid = mobj.group('sid')
+
+        if mobj.group('query'):
+            qs = compat_parse_qs(mobj.group('query'))
+            if not qs.get('playList'):
+                raise ExtractorError('Invalid URL', expected=True)
+            video_id = qs['playList'][0]
+            if qs.get('sid'):
+                sid = qs['sid'][0]
+
         embed_url = 'https://embed.5min.com/playerseed/?playList=%s' % video_id
+        if not sid:
             embed_page = self._download_webpage(embed_url, video_id,
                                                 'Downloading embed page')
             sid = self._search_regex(r'sid=(\d+)', embed_page, 'sid')
-        query = compat_urllib_parse.urlencode({
+
+        response = self._download_json(
+            'https://syn.5min.com/handlers/SenseHandler.ashx?' +
+            compat_urllib_parse.urlencode({
                 'func': 'GetResults',
                 'playlist': video_id,
                 'sid': sid,
                 'isPlayerSeed': 'true',
                 'url': embed_url,
-        })
+            }),
-        response = self._download_json(
-            'https://syn.5min.com/handlers/SenseHandler.ashx?' + query,
             video_id)
         if not response['success']:
             raise ExtractorError(
@@ -118,9 +129,7 @@ class FiveMinIE(InfoExtractor):
         parsed_video_url = compat_urllib_parse_urlparse(compat_parse_qs(
             compat_urllib_parse_urlparse(info['EmbededURL']).query)['videoUrl'][0])
         for rendition in info['Renditions']:
-            if rendition['RenditionType'] == 'm3u8':
+            if rendition['RenditionType'] == 'aac' or rendition['RenditionType'] == 'm3u8':
-                formats.extend(self._extract_m3u8_formats(rendition['Url'], video_id, m3u8_id='hls'))
-            elif rendition['RenditionType'] == 'aac':
                 continue
             else:
                 rendition_url = compat_urlparse.urlunparse(parsed_video_url._replace(path=replace_extension(parsed_video_url.path.replace('//', '/%s/' % rendition['ID']), rendition['RenditionType'])))