[utils] Introduce merge_dicts

[youtube-dl] / youtube_dl / extractor / generic.py

diff --git a/youtube_dl/extractor/generic.py b/youtube_dl/extractor/generic.py
index 4b210da72eb52e7e216fee67579bbf38b1d4e60f..d48914495c4e02a62061a0bf97e3e6cf494fbb27 100644 (file)
--- a/youtube_dl/extractor/generic.py
+++ b/youtube_dl/extractor/generic.py
@@ -23,6 +23,7 @@ from ..utils import (
     is_html,
     js_to_json,
     KNOWN_EXTENSIONS,
+    merge_dicts,
     mimetype2ext,
     orderedSet,
     sanitized_Request,
@@ -58,6 +59,7 @@ from .xhamster import XHamsterEmbedIE
 from .tnaflix import TNAFlixNetworkEmbedIE
 from .drtuber import DrTuberIE
 from .redtube import RedTubeIE
+from .tube8 import Tube8IE
 from .vimeo import VimeoIE
 from .dailymotion import DailymotionIE
 from .dailymail import DailyMailIE
@@ -1219,7 +1221,7 @@ class GenericIE(InfoExtractor):
                 'title': '35871',
                 'timestamp': 1355743100,
                 'upload_date': '20121217',
-                'uploader_id': 'batchUser',
+                'uploader_id': 'cplapp@learn360.com',
             },
             'add_ie': ['Kaltura'],
         },
@@ -1270,6 +1272,22 @@ class GenericIE(InfoExtractor):
             },
             'add_ie': ['Kaltura'],
         },
+        {
+            # meta twitter:player
+            'url': 'http://thechive.com/2017/12/08/all-i-want-for-christmas-is-more-twerk/',
+            'info_dict': {
+                'id': '0_01b42zps',
+                'ext': 'mp4',
+                'title': 'Main Twerk (Video)',
+                'upload_date': '20171208',
+                'uploader_id': 'sebastian.salinas@thechive.com',
+                'timestamp': 1512713057,
+            },
+            'params': {
+                'skip_download': True,
+            },
+            'add_ie': ['Kaltura'],
+        },
         # referrer protected EaglePlatform embed
         {
             'url': 'https://tvrain.ru/lite/teleshow/kak_vse_nachinalos/namin-418921/',
@@ -1974,10 +1992,10 @@ class GenericIE(InfoExtractor):
             'info_dict': {
                 'id': '83645793',
                 'title': 'Lock up and get excited',
-                'thumbnail': r're:^https?://.*\.jpg(\?.*)?$',
                 'ext': 'mp4'
-            }
-        }
+            },
+            'skip': 'TODO: fix nested playlists processing in tests',
+        },
         # {
         #     # TODO: find another test
         #     # http://schema.org/VideoObject
@@ -2556,6 +2574,11 @@ class GenericIE(InfoExtractor):
         if redtube_urls:
             return self.playlist_from_matches(redtube_urls, video_id, video_title, ie=RedTubeIE.ie_key())
 
+        # Look for embedded Tube8 player
+        tube8_urls = Tube8IE._extract_urls(webpage)
+        if tube8_urls:
+            return self.playlist_from_matches(tube8_urls, video_id, video_title, ie=Tube8IE.ie_key())
+
         # Look for embedded Tvigle player
         mobj = re.search(
             r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//cloud\.tvigle\.ru/video/.+?)\1', webpage)
@@ -2973,28 +2996,12 @@ class GenericIE(InfoExtractor):
             return self.playlist_from_matches(
                 xfileshare_urls, video_id, video_title, ie=XFileShareIE.ie_key())
 
-        def merge_dicts(dict1, dict2):
-            merged = {}
-            for k, v in dict1.items():
-                if v is not None:
-                    merged[k] = v
-            for k, v in dict2.items():
-                if v is None:
-                    continue
-                if (k not in merged or
-                        (isinstance(v, compat_str) and v and
-                            isinstance(merged[k], compat_str) and
-                            not merged[k])):
-                    merged[k] = v
-            return merged
-
-        # Look for Share-Videos.se embeds
-        sharevideosse_urls = [m.group('url') for m in re.finditer(
-            r'<iframe[^>]+?src\s*=\s*(["\'])(?P<url>https?://embed\.share-videos\.se/auto/embed/\d+.+?)\1',
+        sharevideos_urls = [mobj.group('url') for mobj in re.finditer(
+            r'<iframe[^>]+?\bsrc\s*=\s*(["\'])(?P<url>(?:https?:)?//embed\.share-videos\.se/auto/embed/\d+\?.*?\buid=\d+.*?)\1',
             webpage)]
-        if sharevideosse_urls:
+        if sharevideos_urls:
             return self.playlist_from_matches(
-                sharevideosse_urls, video_id, video_title)
+                sharevideos_urls, video_id, video_title)
 
         # Look for HTML5 media
         entries = self._parse_html5_media_entries(url, webpage, video_id, m3u8_id='hls')