Fix extraction from youtube.

[youtubedl] / youtube_dl / extractor / ninecninemedia.py

diff --git a/youtube_dl/extractor/ninecninemedia.py b/youtube_dl/extractor/ninecninemedia.py
index d889245ad6885d9bbd88d0a1b335aa4b5a56f612..65754c5e703acfc38e075f5bc974b5d4815c12d2 100644 (file)
--- a/youtube_dl/extractor/ninecninemedia.py
+++ b/youtube_dl/extractor/ninecninemedia.py
@@ -6,50 +6,97 @@ import re
 from .common import InfoExtractor
 from ..utils import (
     parse_iso8601,
-    parse_duration,
-    ExtractorError
+    float_or_none,
+    ExtractorError,
+    int_or_none,
 )
 
 
 class NineCNineMediaIE(InfoExtractor):
+    IE_NAME = '9c9media'
+    _GEO_COUNTRIES = ['CA']
     _VALID_URL = r'9c9media:(?P<destination_code>[^:]+):(?P<id>\d+)'
+    _API_BASE_TEMPLATE = 'http://capi.9c9media.com/destinations/%s/platforms/desktop/contents/%s/'
 
     def _real_extract(self, url):
-        destination_code, video_id = re.match(self._VALID_URL, url).groups()
-        api_base_url = 'http://capi.9c9media.com/destinations/%s/platforms/desktop/contents/%s/' % (destination_code, video_id)
-        content = self._download_json(api_base_url, video_id, query={
-            '$include': '[contentpackages]',
+        destination_code, content_id = re.match(self._VALID_URL, url).groups()
+        api_base_url = self._API_BASE_TEMPLATE % (destination_code, content_id)
+        content = self._download_json(api_base_url, content_id, query={
+            '$include': '[Media,Season,ContentPackages]',
         })
         title = content['Name']
         if len(content['ContentPackages']) > 1:
             raise ExtractorError('multiple content packages')
         content_package = content['ContentPackages'][0]
-        stacks_base_url = api_base_url + 'contentpackages/%s/stacks/' % content_package['Id']
-        stacks = self._download_json(stacks_base_url, video_id)['Items']
-        if len(stacks) > 1:
-            raise ExtractorError('multiple stacks')
-        stack = stacks[0]
-        stack_base_url = '%s%s/manifest.' % (stacks_base_url, stack['Id'])
+        package_id = content_package['Id']
+        content_package_url = api_base_url + 'contentpackages/%s/' % package_id
+        content_package = self._download_json(
+            content_package_url, content_id, query={
+                '$include': '[HasClosedCaptions]',
+            })
+
+        if content_package.get('Constraints', {}).get('Security', {}).get('Type'):
+            raise ExtractorError('This video is DRM protected.', expected=True)
+
+        manifest_base_url = content_package_url + 'manifest.'
         formats = []
         formats.extend(self._extract_m3u8_formats(
-            stack_base_url + 'm3u8', video_id, 'mp4',
+            manifest_base_url + 'm3u8', content_id, 'mp4',
             'm3u8_native', m3u8_id='hls', fatal=False))
         formats.extend(self._extract_f4m_formats(
-            stack_base_url + 'f4m', video_id,
+            manifest_base_url + 'f4m', content_id,
             f4m_id='hds', fatal=False))
-        mp4_url = self._download_webpage(stack_base_url + 'pd', video_id, fatal=False)
-        if mp4_url:
-            formats.append({
-                'url': mp4_url,
-                'format_id': 'mp4',
-            })
+        formats.extend(self._extract_mpd_formats(
+            manifest_base_url + 'mpd', content_id,
+            mpd_id='dash', fatal=False))
         self._sort_formats(formats)
 
-        return {
-            'id': video_id,
+        thumbnails = []
+        for image in content.get('Images', []):
+            image_url = image.get('Url')
+            if not image_url:
+                continue
+            thumbnails.append({
+                'url': image_url,
+                'width': int_or_none(image.get('Width')),
+                'height': int_or_none(image.get('Height')),
+            })
+
+        tags, categories = [], []
+        for source_name, container in (('Tags', tags), ('Genres', categories)):
+            for e in content.get(source_name, []):
+                e_name = e.get('Name')
+                if not e_name:
+                    continue
+                container.append(e_name)
+
+        season = content.get('Season', {})
+
+        info = {
+            'id': content_id,
             'title': title,
             'description': content.get('Desc') or content.get('ShortDesc'),
             'timestamp': parse_iso8601(content.get('BroadcastDateTime')),
-            'duration': parse_duration(content.get('BroadcastTime')),
+            'episode_number': int_or_none(content.get('Episode')),
+            'season': season.get('Name'),
+            'season_number': season.get('Number'),
+            'season_id': season.get('Id'),
+            'series': content.get('Media', {}).get('Name'),
+            'tags': tags,
+            'categories': categories,
+            'duration': float_or_none(content_package.get('Duration')),
             'formats': formats,
         }
+
+        if content_package.get('HasClosedCaptions'):
+            info['subtitles'] = {
+                'en': [{
+                    'url': manifest_base_url + 'vtt',
+                    'ext': 'vtt',
+                }, {
+                    'url': manifest_base_url + 'srt',
+                    'ext': 'srt',
+                }]
+            }
+
+        return info