youtube-dl

commit f484c5fa257420201768158aaec31af25f904f9c
parent 88d9f6c0c4c3d1d2179ee4fe0af560f500e62579
Author: Sergey M․ <dstftw@gmail.com>
Date:   Sun, 26 Jun 2016 16:59:28 +0700

[vidbit] Improve (Closes #9759)

Diffstat:
M
youtube_dl/extractor/vidbit.py
 | 
84
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++-----------------

1 file changed, 66 insertions(+), 18 deletions(-)
diff --git a/youtube_dl/extractor/vidbit.py b/youtube_dl/extractor/vidbit.py
@@ -1,36 +1,84 @@
-# coding: utf-8
 from __future__ import unicode_literals
 
-import re
-
 from .common import InfoExtractor
-from ..utils import url_basename
 from ..compat import compat_urlparse
+from ..utils import (
+    int_or_none,
+    js_to_json,
+    remove_end,
+    unified_strdate,
+)
 
 
 class VidbitIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?vidbit\.co/watch\?v=(?P<id>[\w-]+)'
-    _TEST = {
-        'url': 'http://www.vidbit.co/watch?v=MrM7LeaMJq',
-        'md5': 'f1a579a93282a78de7e1c53220ef0f12',
+    _VALID_URL = r'https?://(?:www\.)?vidbit\.co/(?:watch|embed)\?.*?\bv=(?P<id>[\da-zA-Z]+)'
+    _TESTS = [{
+        'url': 'http://www.vidbit.co/watch?v=jkL2yDOEq2',
+        'md5': '1a34b7f14defe3b8fafca9796892924d',
         'info_dict': {
-            'id': 'MrM7LeaMJq',
+            'id': 'jkL2yDOEq2',
             'ext': 'mp4',
-            'title': 'RoboCop (1987) - Dick You\'re Fired',
-            'thumbnail': 'http://www.vidbit.co/thumbnails/MrM7LeaMJq.jpg',
+            'title': 'Intro to VidBit',
+            'description': 'md5:5e0d6142eec00b766cbf114bfd3d16b7',
+            'thumbnail': 're:https?://.*\.jpg$',
+            'upload_date': '20160618',
+            'view_count': int,
+            'comment_count': int,
         }
-    }
+    }, {
+        'url': 'http://www.vidbit.co/embed?v=jkL2yDOEq2&auto=0&water=0',
+        'only_matching': True,
+    }]
 
     def _real_extract(self, url):
         video_id = self._match_id(url)
-        webpage = self._download_webpage(url, video_id)
+
+        webpage = self._download_webpage(
+            compat_urlparse.urljoin(url, '/watch?v=%s' % video_id), video_id)
+
+        video_url, title = [None] * 2
+
+        config = self._parse_json(self._search_regex(
+            r'(?s)\.setup\(({.+?})\);', webpage, 'setup', default='{}'),
+            video_id, transform_source=js_to_json)
+        if config:
+            if config.get('file'):
+                video_url = compat_urlparse.urljoin(url, config['file'])
+            title = config.get('title')
+
+        if not video_url:
+            video_url = compat_urlparse.urljoin(url, self._search_regex(
+                r'file\s*:\s*(["\'])(?P<url>(?:(?!\1).)+)\1',
+                webpage, 'video URL', group='url'))
+
+        if not title:
+            title = remove_end(
+                self._html_search_regex(
+                    (r'<h1>(.+?)</h1>', r'<title>(.+?)</title>'),
+                    webpage, 'title', default=None) or self._og_search_title(webpage),
+                ' - VidBit')
+
+        description = self._html_search_meta(
+            ('description', 'og:description', 'twitter:description'),
+            webpage, 'description')
+
+        upload_date = unified_strdate(self._html_search_meta(
+            'datePublished', webpage, 'upload date'))
+
+        view_count = int_or_none(self._search_regex(
+            r'<strong>(\d+)</strong> views',
+            webpage, 'view count', fatal=False))
+        comment_count = int_or_none(self._search_regex(
+            r'id=["\']cmt_num["\'][^>]*>\((\d+)\)',
+            webpage, 'comment count', fatal=False))
 
         return {
             'id': video_id,
-            'title': self._html_search_regex(r'<h1>(.+)</h1>', webpage, 'title'),
-            'url': compat_urlparse.urljoin(url, self._html_search_regex(r'file:\s*(["\'])((?:(?!\1).)+)\1',
-                webpage, 'video URL', group=2)),
+            'url': video_url,
+            'title': title,
+            'description': description,
             'thumbnail': self._og_search_thumbnail(webpage),
-            'description': self._html_search_regex(r'description:(["\'])((?:(?!\1).)+)\1',
-                webpage, 'description', None, group=2),
+            'upload_date': upload_date,
+            'view_count': view_count,
+            'comment_count': comment_count,
         }