X-Git-Url: http://git.bitcoin.ninja/index.cgi?a=blobdiff_plain;f=youtube_dl%2Fextractor%2Fgeneric.py;h=5b22b6b5e4568fbe60845491845c468f3606d262;hb=b8f67449ecafa76d4d925c7ad72f9c8f8338aba7;hp=8121f04a5e02cf672dc6ab0f152d21df8b98034e;hpb=83548824c29ccdf53a4659260aa3898939833882;p=youtube-dl

diff --git a/youtube_dl/extractor/generic.py b/youtube_dl/extractor/generic.py
index 8121f04a5..5b22b6b5e 100644
--- a/youtube_dl/extractor/generic.py
+++ b/youtube_dl/extractor/generic.py
@@ -59,6 +59,8 @@ from .videomore import VideomoreIE
 from .googledrive import GoogleDriveIE
 from .jwplatform import JWPlatformIE
 from .digiteka import DigitekaIE
+from .instagram import InstagramIE
+from .liveleak import LiveLeakIE
 
 
 class GenericIE(InfoExtractor):
@@ -239,6 +241,35 @@ class GenericIE(InfoExtractor):
                 'format': 'bestvideo',
             },
         },
+        # m3u8 served with Content-Type: audio/x-mpegURL; charset=utf-8
+        {
+            'url': 'http://once.unicornmedia.com/now/master/playlist/bb0b18ba-64f5-4b1b-a29f-0ac252f06b68/77a785f3-5188-4806-b788-0893a61634ed/93677179-2d99-4ef4-9e17-fe70d49abfbf/content.m3u8',
+            'info_dict': {
+                'id': 'content',
+                'ext': 'mp4',
+                'title': 'content',
+                'formats': 'mincount:8',
+            },
+            'params': {
+                # m3u8 downloads
+                'skip_download': True,
+            }
+        },
+        # m3u8 served with Content-Type: text/plain
+        {
+            'url': 'http://www.nacentapps.com/m3u8/index.m3u8',
+            'info_dict': {
+                'id': 'index',
+                'ext': 'mp4',
+                'title': 'index',
+                'upload_date': '20140720',
+                'formats': 'mincount:11',
+            },
+            'params': {
+                # m3u8 downloads
+                'skip_download': True,
+            }
+        },
         # google redirect
         {
             'url': 'http://www.google.com/url?sa=t&rct=j&q=&esrc=s&source=web&cd=1&cad=rja&ved=0CCUQtwIwAA&url=http%3A%2F%2Fwww.youtube.com%2Fwatch%3Fv%3DcmQHVoWB5FY&ei=F-sNU-LLCaXk4QT52ICQBQ&usg=AFQjCNEw4hL29zgOohLXvpJ-Bdh2bils1Q&bvm=bv.61965928,d.bGE',
@@ -376,19 +407,6 @@ class GenericIE(InfoExtractor):
                 'skip_download': True,
             },
         },
-        # multiple ooyala embeds on SBN network websites
-        {
-            'url': 'http://www.sbnation.com/college-football-recruiting/2015/2/3/7970291/national-signing-day-rationalizations-itll-be-ok-itll-be-ok',
-            'info_dict': {
-                'id': 'national-signing-day-rationalizations-itll-be-ok-itll-be-ok',
-                'title': '25 lies you will tell yourself on National Signing Day - SBNation.com',
-            },
-            'playlist_mincount': 3,
-            'params': {
-                'skip_download': True,
-            },
-            'add_ie': ['Ooyala'],
-        },
         # embed.ly video
         {
             'url': 'http://www.tested.com/science/weird/460206-tested-grinding-coffee-2000-frames-second/',
@@ -1094,7 +1112,47 @@ class GenericIE(InfoExtractor):
                 # m3u8 downloads
                 'skip_download': True,
             }
-        }
+        },
+        # Brightcove embed, with no valid 'renditions' but valid 'IOSRenditions'
+        # This video can't be played in browsers if Flash disabled and UA set to iPhone, which is actually a false alarm
+        {
+            'url': 'https://dl.dropboxusercontent.com/u/29092637/interview.html',
+            'info_dict': {
+                'id': '4785848093001',
+                'ext': 'mp4',
+                'title': 'The Cardinal Pell Interview',
+                'description': 'Sky News Contributor Andrew Bolt interviews George Pell in Rome, following the Cardinal\'s evidence before the Royal Commission into Child Abuse. ',
+                'uploader': 'GlobeCast Australia - GlobeStream',
+            },
+            'params': {
+                # m3u8 downloads
+                'skip_download': True,
+            },
+        },
+        # Another form of arte.tv embed
+        {
+            'url': 'http://www.tv-replay.fr/redirection/09-04-16/arte-reportage-arte-11508975.html',
+            'md5': '850bfe45417ddf221288c88a0cffe2e2',
+            'info_dict': {
+                'id': '030273-562_PLUS7-F',
+                'ext': 'mp4',
+                'title': 'ARTE Reportage - Nulle part, en France',
+                'description': 'md5:e3a0e8868ed7303ed509b9e3af2b870d',
+                'upload_date': '20160409',
+            },
+        },
+        # LiveLeak embed
+        {
+            'url': 'http://www.wykop.pl/link/3088787/',
+            'md5': 'ace83b9ed19b21f68e1b50e844fdf95d',
+            'info_dict': {
+                'id': '874_1459135191',
+                'ext': 'mp4',
+                'title': 'Man shows poor quality of new apartment building',
+                'description': 'The wall is like a sand pile.',
+                'uploader': 'Lake8737',
+            }
+        },
     ]
 
     def report_following_redirect(self, new_url):
@@ -1245,14 +1303,13 @@ class GenericIE(InfoExtractor):
         info_dict = {
             'id': video_id,
             'title': compat_urllib_parse_unquote(os.path.splitext(url_basename(url))[0]),
+            'upload_date': unified_strdate(head_response.headers.get('Last-Modified'))
         }
 
         # Check for direct link to a video
-        content_type = head_response.headers.get('Content-Type', '')
-        m = re.match(r'^(?P<type>audio|video|application(?=/(?:ogg$|(?:vnd\.apple\.|x-)?mpegurl)))/(?P<format_id>.+)$', content_type)
+        content_type = head_response.headers.get('Content-Type', '').lower()
+        m = re.match(r'^(?P<type>audio|video|application(?=/(?:ogg$|(?:vnd\.apple\.|x-)?mpegurl)))/(?P<format_id>[^;\s]+)', content_type)
         if m:
-            upload_date = unified_strdate(
-                head_response.headers.get('Last-Modified'))
             format_id = m.group('format_id')
             if format_id.endswith('mpegurl'):
                 formats = self._extract_m3u8_formats(url, video_id, 'mp4')
@@ -1264,11 +1321,9 @@ class GenericIE(InfoExtractor):
                     'url': url,
                     'vcodec': 'none' if m.group('type') == 'audio' else None
                 }]
-            info_dict.update({
-                'direct': True,
-                'formats': formats,
-                'upload_date': upload_date,
-            })
+                info_dict['direct'] = True
+            self._sort_formats(formats)
+            info_dict['formats'] = formats
             return info_dict
 
         if not self._downloader.params.get('test', False) and not is_intentional:
@@ -1289,18 +1344,22 @@ class GenericIE(InfoExtractor):
             request.add_header('Accept-Encoding', '*')
             full_response = self._request_webpage(request, video_id)
 
+        first_bytes = full_response.read(512)
+
+        # Is it an M3U playlist?
+        if first_bytes.startswith(b'#EXTM3U'):
+            info_dict['formats'] = self._extract_m3u8_formats(url, video_id, 'mp4')
+            self._sort_formats(info_dict['formats'])
+            return info_dict
+
         # Maybe it's a direct link to a video?
         # Be careful not to download the whole thing!
-        first_bytes = full_response.read(512)
         if not is_html(first_bytes):
             self._downloader.report_warning(
                 'URL could be a direct video link, returning it as such.')
-            upload_date = unified_strdate(
-                head_response.headers.get('Last-Modified'))
             info_dict.update({
                 'direct': True,
                 'url': url,
-                'upload_date': upload_date,
             })
             return info_dict
 
@@ -1315,15 +1374,19 @@ class GenericIE(InfoExtractor):
             if doc.tag == 'rss':
                 return self._extract_rss(url, video_id, doc)
             elif re.match(r'^(?:{[^}]+})?smil$', doc.tag):
-                return self._parse_smil(doc, url, video_id)
+                smil = self._parse_smil(doc, url, video_id)
+                self._sort_formats(smil['formats'])
+                return smil
             elif doc.tag == '{http://xspf.org/ns/0/}playlist':
                 return self.playlist_result(self._parse_xspf(doc, video_id), video_id)
             elif re.match(r'(?i)^(?:{[^}]+})?MPD$', doc.tag):
                 info_dict['formats'] = self._parse_mpd_formats(
                     doc, video_id, mpd_base_url=url.rpartition('/')[0])
+                self._sort_formats(info_dict['formats'])
                 return info_dict
             elif re.match(r'^{http://ns\.adobe\.com/f4m/[12]\.0}manifest$', doc.tag):
                 info_dict['formats'] = self._parse_f4m_formats(doc, url, video_id)
+                self._sort_formats(info_dict['formats'])
                 return info_dict
         except compat_xml_parse_error:
             pass
@@ -1664,7 +1727,7 @@ class GenericIE(InfoExtractor):
 
         # Look for embedded arte.tv player
         mobj = re.search(
-            r'<script [^>]*?src="(?P<url>http://www\.arte\.tv/playerv2/embed[^"]+)"',
+            r'<(?:script|iframe) [^>]*?src="(?P<url>http://www\.arte\.tv/(?:playerv2/embed|arte_vp/index)[^"]+)"',
             webpage)
         if mobj is not None:
             return self.url_result(mobj.group('url'), 'ArteTVEmbed')
@@ -1881,6 +1944,24 @@ class GenericIE(InfoExtractor):
                 self._proto_relative_url(unescapeHTML(mobj.group(1))),
                 'AdobeTVVideo')
 
+        # Look for Vine embeds
+        mobj = re.search(
+            r'<iframe[^>]+src=[\'"]((?:https?:)?//(?:www\.)?vine\.co/v/[^/]+/embed/(?:simple|postcard))',
+            webpage)
+        if mobj is not None:
+            return self.url_result(
+                self._proto_relative_url(unescapeHTML(mobj.group(1))), 'Vine')
+
+        # Look for Instagram embeds
+        instagram_embed_url = InstagramIE._extract_embed_url(webpage)
+        if instagram_embed_url is not None:
+            return self.url_result(instagram_embed_url, InstagramIE.ie_key())
+
+        # Look for LiveLeak embeds
+        liveleak_url = LiveLeakIE._extract_url(webpage)
+        if liveleak_url:
+            return self.url_result(liveleak_url, 'LiveLeak')
+
         def check_video(vurl):
             if YoutubeIE.suitable(vurl):
                 return True
@@ -1995,6 +2076,9 @@ class GenericIE(InfoExtractor):
             else:
                 entry_info_dict['url'] = video_url
 
+            if entry_info_dict.get('formats'):
+                self._sort_formats(entry_info_dict['formats'])
+
             entries.append(entry_info_dict)
 
         if len(entries) == 1: