[viewlift] fix extraction for snagfils.com(closes #15766)

youtube_dl/extractor/viewlift.py

@@ -1,24 +1,27 @@
 from __future__ import unicode_literals
 
+import base64
 import re
 
 from .common import InfoExtractor
+from ..compat import compat_urllib_parse_unquote
 from ..utils import (
     ExtractorError,
     clean_html,
     determine_ext,
     int_or_none,
     js_to_json,
+    parse_age_limit,
     parse_duration,
 )
 
 
 class ViewLiftBaseIE(InfoExtractor):
-    _DOMAINS_REGEX = r'(?:snagfilms|snagxtreme|funnyforfree|kiddovid|winnersview|monumentalsportsnetwork|vayafilm)\.com|kesari\.tv'
+    _DOMAINS_REGEX = r'(?:snagfilms|snagxtreme|funnyforfree|kiddovid|winnersview|(?:monumental|lax)sportsnetwork|vayafilm)\.com'
 
 
 class ViewLiftEmbedIE(ViewLiftBaseIE):
-    _VALID_URL = r'https?://(?:(?:www|embed)\.)?(?:%s)/embed/player\?.*\bfilmId=(?P<id>[\da-f-]{36})' % ViewLiftBaseIE._DOMAINS_REGEX
+    _VALID_URL = r'https?://(?:(?:www|embed)\.)?(?:%s)/embed/player\?.*\bfilmId=(?P<id>[\da-f]{8}-(?:[\da-f]{4}-){3}[\da-f]{12})' % ViewLiftBaseIE._DOMAINS_REGEX
     _TESTS = [{
         'url': 'http://embed.snagfilms.com/embed/player?filmId=74849a00-85a9-11e1-9660-123139220831&w=500',
         'md5': '2924e9215c6eff7a55ed35b72276bd93',
@@ -60,8 +63,10 @@ def _real_extract(self, url):
 
         formats = []
         has_bitrate = False
-        for source in self._parse_json(js_to_json(self._search_regex(
+        sources = self._parse_json(self._search_regex(
-                r'(?s)sources:\s*(\[.+?\]),', webpage, 'json')), video_id):
+            r'(?s)sources:\s*(\[.+?\]),', webpage,
+            'sources', default='[]'), video_id, js_to_json)
+        for source in sources:
             file_ = source.get('file')
             if not file_:
                 continue
@@ -70,7 +75,8 @@ def _real_extract(self, url):
             format_id = source.get('label') or ext
             if all(v in ('m3u8', 'hls') for v in (type_, ext)):
                 formats.extend(self._extract_m3u8_formats(
-                    file_, video_id, 'mp4', m3u8_id='hls'))
+                    file_, video_id, 'mp4', 'm3u8_native',
+                    m3u8_id='hls', fatal=False))
             else:
                 bitrate = int_or_none(self._search_regex(
                     [r'(\d+)kbps', r'_\d{1,2}x\d{1,2}_(\d{3,})\.%s' % ext],
@@ -85,6 +91,13 @@ def _real_extract(self, url):
                     'tbr': bitrate,
                     'height': height,
                 })
+        if not formats:
+            hls_url = self._parse_json(self._search_regex(
+                r'filmInfo\.src\s*=\s*({.+?});',
+                webpage, 'src'), video_id, js_to_json)['src']
+            formats = self._extract_m3u8_formats(
+                hls_url, video_id, 'mp4', 'm3u8_native',
+                m3u8_id='hls', fatal=False)
         field_preference = None if has_bitrate else ('height', 'tbr', 'format_id')
         self._sort_formats(formats, field_preference)
 
@@ -109,10 +122,13 @@ class ViewLiftIE(ViewLiftBaseIE):
             'display_id': 'lost_for_life',
             'ext': 'mp4',
             'title': 'Lost for Life',
-            'description': 'md5:fbdacc8bb6b455e464aaf98bc02e1c82',
+            'description': 'md5:ea10b5a50405ae1f7b5269a6ec594102',
             'thumbnail': r're:^https?://.*\.jpg',
             'duration': 4489,
-            'categories': ['Documentary', 'Crime', 'Award Winning', 'Festivals']
+            'categories': 'mincount:3',
+            'age_limit': 14,
+            'upload_date': '20150421',
+            'timestamp': 1429656819,
         }
     }, {
         'url': 'http://www.snagfilms.com/show/the_world_cut_project/india',
@@ -125,7 +141,9 @@ class ViewLiftIE(ViewLiftBaseIE):
             'description': 'md5:5c168c5a8f4719c146aad2e0dfac6f5f',
             'thumbnail': r're:^https?://.*\.jpg',
             'duration': 979,
-            'categories': ['Documentary', 'Sports', 'Politics']
+            'categories': 'mincount:2',
+            'timestamp': 1399478279,
+            'upload_date': '20140507',
         }
     }, {
         # Film is not playable in your area.
@@ -138,9 +156,6 @@ class ViewLiftIE(ViewLiftBaseIE):
     }, {
         'url': 'http://www.winnersview.com/videos/the-good-son',
         'only_matching': True,
-    }, {
-        'url': 'http://www.kesari.tv/news/video/1461919076414',
-        'only_matching': True,
     }, {
         # Was once Kaltura embed
         'url': 'https://www.monumentalsportsnetwork.com/videos/john-carlson-postgame-2-25-15',
@@ -156,45 +171,96 @@ def _real_extract(self, url):
             raise ExtractorError(
                 'Film %s is not available.' % display_id, expected=True)
 
-        film_id = self._search_regex(r'filmId=([\da-f-]{36})"', webpage, 'film id')
+        initial_store_state = self._search_regex(
+            r"window\.initialStoreState\s*=.*?JSON\.parse\(unescape\(atob\('([^']+)'\)\)\)",
+            webpage, 'Initial Store State', default=None)
+        if initial_store_state:
+            modules = self._parse_json(compat_urllib_parse_unquote(base64.b64decode(
+                initial_store_state).decode()), display_id)['page']['data']['modules']
+            content_data = next(m['contentData'][0] for m in modules if m.get('moduleType') == 'VideoDetailModule')
+            gist = content_data['gist']
+            film_id = gist['id']
+            title = gist['title']
+            video_assets = content_data['streamingInfo']['videoAssets']
 
-        snag = self._parse_json(
+            formats = []
-            self._search_regex(
+            mpeg_video_assets = video_assets.get('mpeg') or []
-                r'Snag\.page\.data\s*=\s*(\[.+?\]);', webpage, 'snag'),
+            for video_asset in mpeg_video_assets:
-            display_id)
+                video_asset_url = video_asset.get('url')
+                if not video_asset:
+                    continue
+                bitrate = int_or_none(video_asset.get('bitrate'))
+                height = int_or_none(self._search_regex(
+                    r'^_?(\d+)[pP]$', video_asset.get('renditionValue'),
+                    'height', default=None))
+                formats.append({
+                    'url': video_asset_url,
+                    'format_id': 'http%s' % ('-%d' % bitrate if bitrate else ''),
+                    'tbr': bitrate,
+                    'height': height,
+                    'vcodec': video_asset.get('codec'),
+                })
 
-        for item in snag:
+            hls_url = video_assets.get('hls')
-            if item.get('data', {}).get('film', {}).get('id') == film_id:
+            if hls_url:
-                data = item['data']['film']
+                formats.extend(self._extract_m3u8_formats(
-                title = data['title']
+                    hls_url, film_id, 'mp4', 'm3u8_native', m3u8_id='hls', fatal=False))
-                description = clean_html(data.get('synopsis'))
+            self._sort_formats(formats, ('height', 'tbr', 'format_id'))
-                thumbnail = data.get('image')
+
-                duration = int_or_none(data.get('duration') or data.get('runtime'))
+            info = {
-                categories = [
+                'id': film_id,
-                    category['title'] for category in data.get('categories', [])
+                'display_id': display_id,
-                    if category.get('title')]
+                'title': title,
-                break
+                'description': gist.get('description'),
+                'thumbnail': gist.get('videoImageUrl'),
+                'duration': int_or_none(gist.get('runtime')),
+                'age_limit': parse_age_limit(content_data.get('parentalRating', '').replace('_', '-')),
+                'timestamp': int_or_none(gist.get('publishDate'), 1000),
+                'formats': formats,
+            }
+            for k in ('categories', 'tags'):
+                info[k] = [v['title'] for v in content_data.get(k, []) if v.get('title')]
+            return info
         else:
-            title = self._search_regex(
+            film_id = self._search_regex(r'filmId=([\da-f-]{36})"', webpage, 'film id')
-                r'itemprop="title">([^<]+)<', webpage, 'title')
-            description = self._html_search_regex(
-                r'(?s)<div itemprop="description" class="film-synopsis-inner ">(.+?)</div>',
-                webpage, 'description', default=None) or self._og_search_description(webpage)
-            thumbnail = self._og_search_thumbnail(webpage)
-            duration = parse_duration(self._search_regex(
-                r'<span itemprop="duration" class="film-duration strong">([^<]+)<',
-                webpage, 'duration', fatal=False))
-            categories = re.findall(r'<a href="/movies/[^"]+">([^<]+)</a>', webpage)
 
-        return {
+            snag = self._parse_json(
-            '_type': 'url_transparent',
+                self._search_regex(
-            'url': 'http://%s/embed/player?filmId=%s' % (domain, film_id),
+                    r'Snag\.page\.data\s*=\s*(\[.+?\]);', webpage, 'snag', default='[]'),
-            'id': film_id,
+                display_id)
-            'display_id': display_id,
+
-            'title': title,
+            for item in snag:
-            'description': description,
+                if item.get('data', {}).get('film', {}).get('id') == film_id:
-            'thumbnail': thumbnail,
+                    data = item['data']['film']
-            'duration': duration,
+                    title = data['title']
-            'categories': categories,
+                    description = clean_html(data.get('synopsis'))
-            'ie_key': 'ViewLiftEmbed',
+                    thumbnail = data.get('image')
-        }
+                    duration = int_or_none(data.get('duration') or data.get('runtime'))
+                    categories = [
+                        category['title'] for category in data.get('categories', [])
+                        if category.get('title')]
+                    break
+            else:
+                title = self._search_regex(
+                    r'itemprop="title">([^<]+)<', webpage, 'title')
+                description = self._html_search_regex(
+                    r'(?s)<div itemprop="description" class="film-synopsis-inner ">(.+?)</div>',
+                    webpage, 'description', default=None) or self._og_search_description(webpage)
+                thumbnail = self._og_search_thumbnail(webpage)
+                duration = parse_duration(self._search_regex(
+                    r'<span itemprop="duration" class="film-duration strong">([^<]+)<',
+                    webpage, 'duration', fatal=False))
+                categories = re.findall(r'<a href="/movies/[^"]+">([^<]+)</a>', webpage)
+
+            return {
+                '_type': 'url_transparent',
+                'url': 'http://%s/embed/player?filmId=%s' % (domain, film_id),
+                'id': film_id,
+                'display_id': display_id,
+                'title': title,
+                'description': description,
+                'thumbnail': thumbnail,
+                'duration': duration,
+                'categories': categories,
+                'ie_key': 'ViewLiftEmbed',
+            }