youtube_dl/extractor/extractors.py

@@ -1399,6 +1399,7 @@ from .vidme import (
     VidmeUserIE,
     VidmeUserLikesIE,
 )
+from .vidzi import VidziIE
 from .vier import VierIE, VierVideosIE
 from .viewlift import (
     ViewLiftIE,

youtube_dl/extractor/vidio.py

@@ -4,13 +4,7 @@ from __future__ import unicode_literals
 import re
 
 from .common import InfoExtractor
-from ..utils import (
-    int_or_none,
-    parse_iso8601,
-    str_or_none,
-    strip_or_none,
-    try_get,
-)
+from ..utils import int_or_none
 
 
 class VidioIE(InfoExtractor):
@@ -27,63 +21,57 @@ class VidioIE(InfoExtractor):
             'thumbnail': r're:^https?://.*\.jpg$',
             'duration': 149,
             'like_count': int,
-            'uploader': 'TWELVE Pic',
-            'timestamp': 1444902800,
-            'upload_date': '20151015',
-            'uploader_id': 'twelvepictures',
-            'channel': 'Cover Music Video',
-            'channel_id': '280236',
-            'view_count': int,
-            'dislike_count': int,
-            'comment_count': int,
-            'tags': 'count:4',
         },
     }, {
         'url': 'https://www.vidio.com/watch/77949-south-korea-test-fires-missile-that-can-strike-all-of-the-north',
         'only_matching': True,
     }]
 
-    def _real_initialize(self):
-        self._api_key = self._download_json(
-            'https://www.vidio.com/auth', None, data=b'')['api_key']
-
     def _real_extract(self, url):
-        video_id, display_id = re.match(self._VALID_URL, url).groups()
-        data = self._download_json(
-            'https://api.vidio.com/videos/' + video_id, display_id, headers={
-                'Content-Type': 'application/vnd.api+json',
-                'X-API-KEY': self._api_key,
-            })
-        video = data['videos'][0]
-        title = video['title'].strip()
+        mobj = re.match(self._VALID_URL, url)
+        video_id, display_id = mobj.group('id', 'display_id')
 
+        webpage = self._download_webpage(url, display_id)
+
+        title = self._og_search_title(webpage)
+
+        m3u8_url, duration, thumbnail = [None] * 3
+
+        clips = self._parse_json(
+            self._html_search_regex(
+                r'data-json-clips\s*=\s*(["\'])(?P<data>\[.+?\])\1',
+                webpage, 'video data', default='[]', group='data'),
+            display_id, fatal=False)
+        if clips:
+            clip = clips[0]
+            m3u8_url = clip.get('sources', [{}])[0].get('file')
+            duration = clip.get('clip_duration')
+            thumbnail = clip.get('image')
+
+        m3u8_url = m3u8_url or self._search_regex(
+            r'data(?:-vjs)?-clip-hls-url=(["\'])(?P<url>(?:(?!\1).)+)\1',
+            webpage, 'hls url', group='url')
         formats = self._extract_m3u8_formats(
-            data['clips'][0]['hls_url'], display_id, 'mp4', 'm3u8_native')
+            m3u8_url, display_id, 'mp4', entry_protocol='m3u8_native')
         self._sort_formats(formats)
 
-        get_first = lambda x: try_get(data, lambda y: y[x + 's'][0], dict) or {}
-        channel = get_first('channel')
-        user = get_first('user')
-        username = user.get('username')
-        get_count = lambda x: int_or_none(video.get('total_' + x))
+        duration = int_or_none(duration or self._search_regex(
+            r'data-video-duration=(["\'])(?P<duration>\d+)\1', webpage,
+            'duration', fatal=False, group='duration'))
+        thumbnail = thumbnail or self._og_search_thumbnail(webpage)
+
+        like_count = int_or_none(self._search_regex(
+            (r'<span[^>]+data-comment-vote-count=["\'](\d+)',
+             r'<span[^>]+class=["\'].*?\blike(?:__|-)count\b.*?["\'][^>]*>\s*(\d+)'),
+            webpage, 'like count', fatal=False))
 
         return {
             'id': video_id,
             'display_id': display_id,
             'title': title,
-            'description': strip_or_none(video.get('description')),
-            'thumbnail': video.get('image_url_medium'),
-            'duration': int_or_none(video.get('duration')),
-            'like_count': get_count('likes'),
+            'description': self._og_search_description(webpage),
+            'thumbnail': thumbnail,
+            'duration': duration,
+            'like_count': like_count,
             'formats': formats,
-            'uploader': user.get('name'),
-            'timestamp': parse_iso8601(video.get('created_at')),
-            'uploader_id': username,
-            'uploader_url': 'https://www.vidio.com/@' + username if username else None,
-            'channel': channel.get('name'),
-            'channel_id': str_or_none(channel.get('id')),
-            'view_count': get_count('view_count'),
-            'dislike_count': get_count('dislikes'),
-            'comment_count': get_count('comments'),
-            'tags': video.get('tag_list'),
         }

youtube_dl/extractor/vidzi.py

@@ -0,0 +1,68 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+import re
+
+from .common import InfoExtractor
+from ..utils import (
+    decode_packed_codes,
+    js_to_json,
+    NO_DEFAULT,
+    PACKED_CODES_RE,
+)
+
+
+class VidziIE(InfoExtractor):
+    _VALID_URL = r'https?://(?:www\.)?vidzi\.(?:tv|cc|si|nu)/(?:embed-)?(?P<id>[0-9a-zA-Z]+)'
+    _TESTS = [{
+        'url': 'http://vidzi.tv/cghql9yq6emu.html',
+        'md5': '4f16c71ca0c8c8635ab6932b5f3f1660',
+        'info_dict': {
+            'id': 'cghql9yq6emu',
+            'ext': 'mp4',
+            'title': 'youtube-dl test video  1\\\\2\'3/4<5\\\\6ä7↭',
+        },
+        'params': {
+            # m3u8 download
+            'skip_download': True,
+        },
+    }, {
+        'url': 'http://vidzi.tv/embed-4z2yb0rzphe9-600x338.html',
+        'only_matching': True,
+    }, {
+        'url': 'http://vidzi.cc/cghql9yq6emu.html',
+        'only_matching': True,
+    }, {
+        'url': 'https://vidzi.si/rph9gztxj1et.html',
+        'only_matching': True,
+    }, {
+        'url': 'http://vidzi.nu/cghql9yq6emu.html',
+        'only_matching': True,
+    }]
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+
+        webpage = self._download_webpage(
+            'http://vidzi.tv/%s' % video_id, video_id)
+        title = self._html_search_regex(
+            r'(?s)<h2 class="video-title">(.*?)</h2>', webpage, 'title')
+
+        codes = [webpage]
+        codes.extend([
+            decode_packed_codes(mobj.group(0)).replace('\\\'', '\'')
+            for mobj in re.finditer(PACKED_CODES_RE, webpage)])
+        for num, code in enumerate(codes, 1):
+            jwplayer_data = self._parse_json(
+                self._search_regex(
+                    r'setup\(([^)]+)\)', code, 'jwplayer data',
+                    default=NO_DEFAULT if num == len(codes) else '{}'),
+                video_id, transform_source=lambda s: js_to_json(
+                    re.sub(r'\s*\+\s*window\[.+?\]', '', s)))
+            if jwplayer_data:
+                break
+
+        info_dict = self._parse_jwplayer_data(jwplayer_data, video_id, require_title=False)
+        info_dict['title'] = title
+
+        return info_dict