[lynda] minor changes

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

diff --git a/youtube_dl/extractor/imdb.py b/youtube_dl/extractor/imdb.py
index 07e4f7d29fba010b26965e0656798bc243b809bc..16926b4d391bdc11801510797c26481610b928e3 100644 (file)
--- a/youtube_dl/extractor/imdb.py
+++ b/youtube_dl/extractor/imdb.py
@@ -11,7 +11,7 @@ from ..utils import (
 class ImdbIE(InfoExtractor):
     IE_NAME = u'imdb'
     IE_DESC = u'Internet Movie Database trailers'
-    _VALID_URL = r'http://www\.imdb\.com/video/imdb/vi(?P<id>\d+)'
+    _VALID_URL = r'http://(?:www|m)\.imdb\.com/video/imdb/vi(?P<id>\d+)'
 
     _TEST = {
         u'url': u'http://www.imdb.com/video/imdb/vi2524815897',
@@ -21,32 +21,31 @@ class ImdbIE(InfoExtractor):
             u'ext': u'mp4',
             u'title': u'Ice Age: Continental Drift Trailer (No. 2) - IMDb',
             u'description': u'md5:9061c2219254e5d14e03c25c98e96a81',
-            u'duration': 151,
         }
     }
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
         video_id = mobj.group('id')
-        webpage = self._download_webpage(url,video_id)
+        webpage = self._download_webpage('http://www.imdb.com/video/imdb/vi%s' % video_id, video_id)
         descr = get_element_by_attribute('itemprop', 'description', webpage)
         available_formats = re.findall(
             r'case \'(?P<f_id>.*?)\' :$\s+url = \'(?P<path>.*?)\'', webpage,
             flags=re.MULTILINE)
         formats = []
         for f_id, f_path in available_formats:
+            f_path = f_path.strip()
             format_page = self._download_webpage(
                 compat_urlparse.urljoin(url, f_path),
                 u'Downloading info for %s format' % f_id)
-            json_data = get_element_by_attribute('class', 'imdb-player-data',
-                format_page)
+            json_data = self._search_regex(
+                r'<script[^>]+class="imdb-player-data"[^>]*?>(.*?)</script>',
+                format_page, u'json data', flags=re.DOTALL)
             info = json.loads(json_data)
             format_info = info['videoPlayerObject']['video']
             formats.append({
                 'format_id': f_id,
                 'url': format_info['url'],
-                'height': format_info['height'],
-                'width': format_info['width'],
             })
 
         return {
@@ -55,5 +54,33 @@ class ImdbIE(InfoExtractor):
             'formats': formats,
             'description': descr,
             'thumbnail': format_info['slate'],
-            'duration': int(info['titleObject']['title']['duration_seconds']),
         }
+
+class ImdbListIE(InfoExtractor):
+    IE_NAME = u'imdb:list'
+    IE_DESC = u'Internet Movie Database lists'
+    _VALID_URL = r'http://www\.imdb\.com/list/(?P<id>[\da-zA-Z_-]{11})'
+    
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        list_id = mobj.group('id')
+        
+        # RSS XML is sometimes malformed
+        rss = self._download_webpage('http://rss.imdb.com/list/%s' % list_id, list_id, u'Downloading list RSS')
+        list_title = self._html_search_regex(r'<title>(.*?)</title>', rss, u'list title')
+        
+        # Export is independent of actual author_id, but returns 404 if no author_id is provided.
+        # However, passing dummy author_id seems to be enough.
+        csv = self._download_webpage('http://www.imdb.com/list/export?list_id=%s&author_id=ur00000000' % list_id,
+                                     list_id, u'Downloading list CSV')
+        
+        entries = []
+        for item in csv.split('\n')[1:]:
+            cols = item.split(',')
+            if len(cols) < 2:
+                continue
+            item_id = cols[1][1:-1]
+            if item_id.startswith('vi'):
+                entries.append(self.url_result('http://www.imdb.com/video/imdb/%s' % item_id, 'Imdb'))
+        
+        return self.playlist_result(entries, list_id, list_title)
\ No newline at end of file