X-Git-Url: http://git.cielonegro.org/gitweb.cgi?a=blobdiff_plain;f=youtube_dl%2FInfoExtractors.py;h=4aec8c6879e79ccc13cf5e9bcfbf9abc17c4d1d1;hb=01ba4b80a732308e8da66ba89bac9273181db1ad;hp=2f926f24363947d38c129af8828cfab27f3dad7f;hpb=0251f9c9c085234505b8a65b066ff54052d5fcdb;p=youtube-dl.git

diff --git a/youtube_dl/InfoExtractors.py b/youtube_dl/InfoExtractors.py
index 2f926f243..4aec8c687 100755
--- a/youtube_dl/InfoExtractors.py
+++ b/youtube_dl/InfoExtractors.py
@@ -420,7 +420,7 @@ class YoutubeIE(InfoExtractor):
     def _request_automatic_caption(self, video_id, webpage):
         """We need the webpage for getting the captions url, pass it as an
            argument to speed up the process."""
-        sub_lang = self._downloader.params.get('subtitleslang')
+        sub_lang = self._downloader.params.get('subtitleslang') or 'en'
         sub_format = self._downloader.params.get('subtitlesformat')
         self.to_screen(u'%s: Looking for automatic captions' % video_id)
         mobj = re.search(r';ytplayer.config = ({.*?});', webpage)
@@ -699,14 +699,14 @@ class YoutubeIE(InfoExtractor):
                         pass
                     else:
                         # We report the original error
-                        self._downloader.report_error(sub_error)
+                        self._downloader.report_warning(sub_error)
 
         if self._downloader.params.get('allsubtitles', False):
             video_subtitles = self._extract_all_subtitles(video_id)
             for video_subtitle in video_subtitles:
                 (sub_error, sub_lang, sub) = video_subtitle
                 if sub_error:
-                    self._downloader.report_error(sub_error)
+                    self._downloader.report_warning(sub_error)
 
         if self._downloader.params.get('listsubtitles', False):
             sub_lang_list = self._list_available_subtitles(video_id)
@@ -732,8 +732,11 @@ class YoutubeIE(InfoExtractor):
             for url_data_str in video_info['url_encoded_fmt_stream_map'][0].split(','):
                 url_data = compat_parse_qs(url_data_str)
                 if 'itag' in url_data and 'url' in url_data:
-                    url = url_data['url'][0] + '&signature=' + url_data['sig'][0]
-                    if not 'ratebypass' in url: url += '&ratebypass=yes'
+                    url = url_data['url'][0]
+                    if 'sig' in url_data:
+                        url += '&signature=' + url_data['sig'][0]
+                    if 'ratebypass' not in url:
+                        url += '&ratebypass=yes'
                     url_map[url_data['itag'][0]] = url
 
             format_limit = self._downloader.params.get('format_limit', None)
@@ -940,16 +943,10 @@ class DailymotionIE(InfoExtractor):
         video_title = unescapeHTML(mobj.group('title'))
 
         video_uploader = None
-        mobj = re.search(r'(?im)<span class="owner[^\"]+?">[^<]+?<a [^>]+?>([^<]+?)</a>', webpage)
-        if mobj is None:
-            # lookin for official user
-            mobj_official = re.search(r'<span rel="author"[^>]+?>([^<]+?)</span>', webpage)
-            if mobj_official is None:
-                self._downloader.report_warning(u'unable to extract uploader nickname')
-            else:
-                video_uploader = mobj_official.group(1)
-        else:
-            video_uploader = mobj.group(1)
+        video_uploader = self._search_regex([r'(?im)<span class="owner[^\"]+?">[^<]+?<a [^>]+?>([^<]+?)</a>',
+                                             # Looking for official user
+                                             r'<(?:span|a) .*?rel="author".*?>([^<]+?)</'],
+                                            webpage, 'video uploader')
 
         video_upload_date = None
         mobj = re.search(r'<div class="[^"]*uploaded_cont[^"]*" title="[^"]*">([0-9]{2})-([0-9]{2})-([0-9]{4})</div>', webpage)
@@ -1412,6 +1409,13 @@ class GenericIE(InfoExtractor):
         if mobj is None:
             # Try to find twitter cards info
             mobj = re.search(r'<meta (?:property|name)="twitter:player:stream" (?:content|value)="(.+?)"', webpage)
+        if mobj is None:
+            # We look for Open Graph info:
+            # We have to match any number spaces between elements, some sites try to align them (eg.: statigr.am)
+            m_video_type = re.search(r'<meta.*?property="og:video:type".*?content="video/(.*?)"', webpage)
+            # We only look in og:video if the MIME type is a video, don't try if it's a Flash player:
+            if m_video_type is not None:
+                mobj = re.search(r'<meta.*?property="og:video".*?content="(.*?)"', webpage)
         if mobj is None:
             raise ExtractorError(u'Invalid URL: %s' % url)
 
@@ -1576,7 +1580,7 @@ class YoutubePlaylistIE(InfoExtractor):
                      |
                         ((?:PL|EC|UU)[0-9A-Za-z-_]{10,})
                      )"""
-    _TEMPLATE_URL = 'https://gdata.youtube.com/feeds/api/playlists/%s?max-results=%i&start-index=%i&v=2&alt=json'
+    _TEMPLATE_URL = 'https://gdata.youtube.com/feeds/api/playlists/%s?max-results=%i&start-index=%i&v=2&alt=json&safeSearch=none'
     _MAX_RESULTS = 50
     IE_NAME = u'youtube:playlist'
 
@@ -1612,9 +1616,10 @@ class YoutubePlaylistIE(InfoExtractor):
                 # Number of videos is a multiple of self._MAX_RESULTS
                 break
 
-            videos += [ (entry['yt$position']['$t'], entry['content']['src'])
-                        for entry in response['feed']['entry']
-                        if 'content' in entry ]
+            for entry in response['feed']['entry']:
+                index = entry['yt$position']['$t']
+                if 'media$group' in entry and 'media$player' in entry['media$group']:
+                    videos.append((index, entry['media$group']['media$player']['url']))
 
             if len(response['feed']['entry']) < self._MAX_RESULTS:
                 break
@@ -3365,6 +3370,8 @@ class SteamIE(InfoExtractor):
                 (?P<gameID>\d+)/?
                 (?P<videoID>\d*)(?P<extra>\??) #For urltype == video we sometimes get the videoID
                 """
+    _VIDEO_PAGE_TEMPLATE = 'http://store.steampowered.com/video/%s/'
+    _AGECHECK_TEMPLATE = 'http://store.steampowered.com/agecheck/video/%s/?snr=1_agecheck_agecheck__age-gate&ageDay=1&ageMonth=January&ageYear=1970'
 
     @classmethod
     def suitable(cls, url):
@@ -3374,11 +3381,19 @@ class SteamIE(InfoExtractor):
     def _real_extract(self, url):
         m = re.match(self._VALID_URL, url, re.VERBOSE)
         gameID = m.group('gameID')
-        videourl = 'http://store.steampowered.com/agecheck/video/%s/?snr=1_agecheck_agecheck__age-gate&ageDay=1&ageMonth=January&ageYear=1970' % gameID
-        self.report_age_confirmation()
+
+        videourl = self._VIDEO_PAGE_TEMPLATE % gameID
         webpage = self._download_webpage(videourl, gameID)
-        game_title = re.search(r'<h2 class="pageheader">(?P<game_title>.*?)</h2>', webpage).group('game_title')
-        
+
+        if re.search('<h2>Please enter your birth date to continue:</h2>', webpage) is not None:
+            videourl = self._AGECHECK_TEMPLATE % gameID
+            self.report_age_confirmation()
+            webpage = self._download_webpage(videourl, gameID)
+
+        self.report_extraction(gameID)
+        game_title = self._html_search_regex(r'<h2 class="pageheader">(.*?)</h2>',
+                                             webpage, 'game title')
+
         urlRE = r"'movie_(?P<videoID>\d+)': \{\s*FILENAME: \"(?P<videoURL>[\w:/\.\?=]+)\"(,\s*MOVIE_NAME: \"(?P<videoName>[\w:/\.\?=\+-]+)\")?\s*\},"
         mweb = re.finditer(urlRE, webpage)
         namesRE = r'<span class="title">(?P<videoName>.+?)</span>'
@@ -3484,8 +3499,8 @@ class RBMARadioIE(InfoExtractor):
 
         webpage = self._download_webpage(url, video_id)
 
-        json_data = self._search_regex(r'<script>window.gon = {.*?};gon\.show=(.+?);</script>',
-            webpage, u'json data')
+        json_data = self._search_regex(r'window\.gon.*?gon\.show=(.+?);$',
+            webpage, u'json data', flags=re.MULTILINE)
 
         try:
             data = json.loads(json_data)
@@ -3788,10 +3803,6 @@ class TEDIE(InfoExtractor):
             self.to_screen(u'Getting info of playlist %s: "%s"' % (playlist_id,name))
             return [self._playlist_videos_info(url,name,playlist_id)]
 
-    def _talk_video_link(self,mediaSlug):
-        '''Returns the video link for that mediaSlug'''
-        return 'http://download.ted.com/talks/%s.mp4' % mediaSlug
-
     def _playlist_videos_info(self,url,name,playlist_id=0):
         '''Returns the videos of the playlist'''
         video_RE=r'''
@@ -3804,9 +3815,8 @@ class TEDIE(InfoExtractor):
         m_videos=re.finditer(video_RE,webpage,re.VERBOSE)
         m_names=re.finditer(video_name_RE,webpage)
 
-        playlist_RE = r'div class="headline">(\s*?)<h1>(\s*?)<span>(?P<playlist_title>.*?)</span>'
-        m_playlist = re.search(playlist_RE, webpage)
-        playlist_title = m_playlist.group('playlist_title')
+        playlist_title = self._html_search_regex(r'div class="headline">\s*?<h1>\s*?<span>(.*?)</span>',
+                                                 webpage, 'playlist title')
 
         playlist_entries = []
         for m_video, m_name in zip(m_videos,m_names):
@@ -3817,27 +3827,28 @@ class TEDIE(InfoExtractor):
 
     def _talk_info(self, url, video_id=0):
         """Return the video for the talk in the url"""
-        m=re.match(self._VALID_URL, url,re.VERBOSE)
-        videoName=m.group('name')
-        webpage=self._download_webpage(url, video_id, 'Downloading \"%s\" page' % videoName)
+        m = re.match(self._VALID_URL, url,re.VERBOSE)
+        video_name = m.group('name')
+        webpage = self._download_webpage(url, video_id, 'Downloading \"%s\" page' % video_name)
+        self.report_extraction(video_name)
         # If the url includes the language we get the title translated
-        title_RE=r'<span id="altHeadline" >(?P<title>.*)</span>'
-        title=re.search(title_RE, webpage).group('title')
-        info_RE=r'''<script\ type="text/javascript">var\ talkDetails\ =(.*?)
-                        "id":(?P<videoID>[\d]+).*?
-                        "mediaSlug":"(?P<mediaSlug>[\w\d]+?)"'''
-        thumb_RE=r'</span>[\s.]*</div>[\s.]*<img src="(?P<thumbnail>.*?)"'
-        thumb_match=re.search(thumb_RE,webpage)
-        info_match=re.search(info_RE,webpage,re.VERBOSE)
-        video_id=info_match.group('videoID')
-        mediaSlug=info_match.group('mediaSlug')
-        video_url=self._talk_video_link(mediaSlug)
+        title = self._html_search_regex(r'<span id="altHeadline" >(?P<title>.*)</span>',
+                                        webpage, 'title')
+        json_data = self._search_regex(r'<script.*?>var talkDetails = ({.*?})</script>',
+                                    webpage, 'json data')
+        info = json.loads(json_data)
+        desc = self._html_search_regex(r'<div class="talk-intro">.*?<p.*?>(.*?)</p>',
+                                       webpage, 'description', flags = re.DOTALL)
+        
+        thumbnail = self._search_regex(r'</span>[\s.]*</div>[\s.]*<img src="(.*?)"',
+                                       webpage, 'thumbnail')
         info = {
-                'id': video_id,
-                'url': video_url,
+                'id': info['id'],
+                'url': info['htmlStreams'][-1]['file'],
                 'ext': 'mp4',
                 'title': title,
-                'thumbnail': thumb_match.group('thumbnail')
+                'thumbnail': thumbnail,
+                'description': desc,
                 }
         return info
 
@@ -4541,6 +4552,29 @@ class GametrailersIE(InfoExtractor):
                 'description': video_description,
                 }
 
+class StatigrIE(InfoExtractor):
+    _VALID_URL = r'(?:http://)?(?:www\.)?statigr\.am/p/([^/]+)'
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        if mobj is None:
+            raise ExtractorError(u'Invalid URL: %s' % url)
+        video_id = mobj.group(1)
+        webpage = self._download_webpage(url, video_id)
+        video_url = re.search(r'<meta property="og:video:secure_url" content="(.+?)">',webpage).group(1)
+        thumbnail_url = re.search(r'<meta property="og:image" content="(.+?)" />',webpage).group(1)
+        title = (re.search(r'<title>(.+?)</title>',webpage).group(1)).strip("| Statigram")
+        uploader = re.search(r'@(.+) \(Videos\)',title).group(1)
+        ext = "mp4"
+        return [{
+            'id':        video_id,
+            'url':       video_url,
+            'ext':       ext,
+            'title':     title,
+            'thumbnail': thumbnail_url,
+            'uploader' : uploader
+        }]
+
 def gen_extractors():
     """ Return a list of an instance of every supported extractor.
     The order does matter; the first extractor matched is the one handling the URL.
@@ -4607,6 +4641,7 @@ def gen_extractors():
         HypemIE(),
         Vbox7IE(),
         GametrailersIE(),
+        StatigrIE(),
         GenericIE()
     ]