[varzesh3] Fix metadata extraction (closes #9197)

youtube_dl/extractor/varzesh3.py

@@ -2,11 +2,19 @@
 from __future__ import unicode_literals
 
 from .common import InfoExtractor
+from ..compat import (
+    compat_urllib_parse_urlparse,
+    compat_parse_qs,
+)
+from ..utils import (
+    clean_html,
+    remove_start,
+)
 
 
 class Varzesh3IE(InfoExtractor):
     _VALID_URL = r'https?://(?:www\.)?video\.varzesh3\.com/(?:[^/]+/)+(?P<id>[^/]+)/?'
-    _TEST = {
+    _TESTS = [{
         'url': 'http://video.varzesh3.com/germany/bundesliga/5-%D9%88%D8%A7%DA%A9%D9%86%D8%B4-%D8%A8%D8%B1%D8%AA%D8%B1-%D8%AF%D8%B1%D9%88%D8%A7%D8%B2%D9%87%E2%80%8C%D8%A8%D8%A7%D9%86%D8%A7%D9%86%D8%9B%D9%87%D9%81%D8%AA%D9%87-26-%D8%A8%D9%88%D9%86%D8%AF%D8%B3/',
         'md5': '2a933874cb7dce4366075281eb49e855',
         'info_dict': {
@@ -15,8 +23,18 @@ class Varzesh3IE(InfoExtractor):
             'title': '۵ واکنش برتر دروازه‌بانان؛هفته ۲۶ بوندسلیگا',
             'description': 'فصل ۲۰۱۵-۲۰۱۴',
             'thumbnail': 're:^https?://.*\.jpg$',
-        }
+        },
-    }
+        'skip': 'HTTP 404 Error',
+    }, {
+        'url': 'http://video.varzesh3.com/video/112785/%D8%AF%D9%84%D9%87-%D8%B9%D9%84%DB%8C%D8%9B-%D8%B3%D8%AA%D8%A7%D8%B1%D9%87-%D9%86%D9%88%D8%B8%D9%87%D9%88%D8%B1-%D9%84%DB%8C%DA%AF-%D8%A8%D8%B1%D8%AA%D8%B1-%D8%AC%D8%B2%DB%8C%D8%B1%D9%87',
+        'info_dict': {
+            'id': '112785',
+            'ext': 'mp4',
+            'title': 'دله علی؛ ستاره نوظهور لیگ برتر جزیره',
+            'description': 'فوتبال 120',
+        },
+        'expected_warnings': ['description'],
+    }]
 
     def _real_extract(self, url):
         display_id = self._match_id(url)
@@ -26,15 +44,30 @@ def _real_extract(self, url):
         video_url = self._search_regex(
             r'<source[^>]+src="([^"]+)"', webpage, 'video url')
 
-        title = self._og_search_title(webpage)
+        title = remove_start(self._html_search_regex(
+            r'<title>([^<]+)</title>', webpage, 'title'), 'ویدیو ورزش 3 | ')
+
         description = self._html_search_regex(
             r'(?s)<div class="matn">(.+?)</div>',
-            webpage, 'description', fatal=False)
+            webpage, 'description', default=None)
-        thumbnail = self._og_search_thumbnail(webpage)
+        if description is None:
+            description = clean_html(self._html_search_meta('description', webpage))
+
+        thumbnail = self._og_search_thumbnail(webpage, default=None)
+        if thumbnail is None:
+            fb_sharer_url = self._search_regex(
+                r'<a[^>]+href="(https?://www\.facebook\.com/sharer/sharer\.php?[^"]+)"',
+                webpage, 'facebook sharer URL', fatal=False)
+            sharer_params = compat_parse_qs(compat_urllib_parse_urlparse(fb_sharer_url).query)
+            thumbnail = sharer_params.get('p[images][0]', [None])[0]
 
         video_id = self._search_regex(
             r"<link[^>]+rel='(?:canonical|shortlink)'[^>]+href='/\?p=([^']+)'",
-            webpage, display_id, default=display_id)
+            webpage, display_id, default=None)
+        if video_id is None:
+            video_id = self._search_regex(
+                'var\s+VideoId\s*=\s*(\d+);', webpage, 'video id',
+                default=display_id)
 
         return {
             'url': video_url,