youtube-dl

commit: f61496863d2207718a2a6cb1591dcbc7abc282de
parent 799756a3b3c794284ca52b9af482e1f03fc46833
Author: Sergey M․ <dstftw@gmail.com>
Date:   Mon, 15 Jul 2019 23:56:05 +0700

[asiancrush] Add support for yuyutv.com, midnightpulp.com and cocoro.tv (closes #21281, closes #21290)

Diffstat:
M
youtube_dl/extractor/asiancrush.py
80
+++++++++++++++++++++++++++++++++++++++++++++++++++++++------------------------

1 file changed, 56 insertions(+), 24 deletions(-)
diff --git a/youtube_dl/extractor/asiancrush.py b/youtube_dl/extractor/asiancrush.py
@@ -5,14 +5,12 @@ import re
 
 from .common import InfoExtractor
 from .kaltura import KalturaIE
-from ..utils import (
-    extract_attributes,
-    remove_end,
-)
+from ..utils import extract_attributes
 
 
 class AsianCrushIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?asiancrush\.com/video/(?:[^/]+/)?0+(?P<id>\d+)v\b'
+    _VALID_URL_BASE = r'https?://(?:www\.)?(?P<host>(?:(?:asiancrush|yuyutv|midnightpulp)\.com|cocoro\.tv))'
+    _VALID_URL = r'%s/video/(?:[^/]+/)?0+(?P<id>\d+)v\b' % _VALID_URL_BASE
     _TESTS = [{
         'url': 'https://www.asiancrush.com/video/012869v/women-who-flirt/',
         'md5': 'c3b740e48d0ba002a42c0b72857beae6',
@@ -20,7 +18,7 @@ class AsianCrushIE(InfoExtractor):
             'id': '1_y4tmjm5r',
             'ext': 'mp4',
             'title': 'Women Who Flirt',
-            'description': 'md5:3db14e9186197857e7063522cb89a805',
+            'description': 'md5:7e986615808bcfb11756eb503a751487',
             'timestamp': 1496936429,
             'upload_date': '20170608',
             'uploader_id': 'craig@crifkin.com',
@@ -28,10 +26,27 @@ class AsianCrushIE(InfoExtractor):
     }, {
         'url': 'https://www.asiancrush.com/video/she-was-pretty/011886v-pretty-episode-3/',
         'only_matching': True,
+    }, {
+        'url': 'https://www.yuyutv.com/video/013886v/the-act-of-killing/',
+        'only_matching': True,
+    }, {
+        'url': 'https://www.yuyutv.com/video/peep-show/013922v-warring-factions/',
+        'only_matching': True,
+    }, {
+        'url': 'https://www.midnightpulp.com/video/010400v/drifters/',
+        'only_matching': True,
+    }, {
+        'url': 'https://www.midnightpulp.com/video/mononoke/016378v-zashikiwarashi-part-1/',
+        'only_matching': True,
+    }, {
+        'url': 'https://www.cocoro.tv/video/the-wonderful-wizard-of-oz/008878v-the-wonderful-wizard-of-oz-ep01/',
+        'only_matching': True,
     }]
 
     def _real_extract(self, url):
-        video_id = self._match_id(url)
+        mobj = re.match(self._VALID_URL, url)
+        host = mobj.group('host')
+        video_id = mobj.group('id')
 
         webpage = self._download_webpage(url, video_id)
 
@@ -51,7 +66,7 @@ class AsianCrushIE(InfoExtractor):
                 r'\bentry_id["\']\s*:\s*["\'](\d+)', webpage, 'entry id')
 
         player = self._download_webpage(
-            'https://api.asiancrush.com/embeddedVideoPlayer', video_id,
+            'https://api.%s/embeddedVideoPlayer' % host, video_id,
             query={'id': entry_id})
 
         kaltura_id = self._search_regex(
@@ -63,15 +78,23 @@ class AsianCrushIE(InfoExtractor):
                 r'/p(?:artner_id)?/(\d+)', player, 'partner id',
                 default='513551')
 
-        return self.url_result(
-            'kaltura:%s:%s' % (partner_id, kaltura_id),
-            ie=KalturaIE.ie_key(), video_id=kaltura_id,
-            video_title=title)
+        description = self._html_search_regex(
+            r'(?s)<div[^>]+\bclass=["\']description["\'][^>]*>(.+?)</div>',
+            webpage, 'description', fatal=False)
+
+        return {
+            '_type': 'url_transparent',
+            'url': 'kaltura:%s:%s' % (partner_id, kaltura_id),
+            'ie_key': KalturaIE.ie_key(),
+            'id': video_id,
+            'title': title,
+            'description': description,
+        }
 
 
 class AsianCrushPlaylistIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?asiancrush\.com/series/0+(?P<id>\d+)s\b'
-    _TEST = {
+    _VALID_URL = r'%s/series/0+(?P<id>\d+)s\b' % AsianCrushIE._VALID_URL_BASE
+    _TESTS = [{
         'url': 'https://www.asiancrush.com/series/012481s/scholar-walks-night/',
         'info_dict': {
             'id': '12481',
@@ -79,7 +102,16 @@ class AsianCrushPlaylistIE(InfoExtractor):
             'description': 'md5:7addd7c5132a09fd4741152d96cce886',
         },
         'playlist_count': 20,
-    }
+    }, {
+        'url': 'https://www.yuyutv.com/series/013920s/peep-show/',
+        'only_matching': True,
+    }, {
+        'url': 'https://www.midnightpulp.com/series/016375s/mononoke/',
+        'only_matching': True,
+    }, {
+        'url': 'https://www.cocoro.tv/series/008549s/the-wonderful-wizard-of-oz/',
+        'only_matching': True,
+    }]
 
     def _real_extract(self, url):
         playlist_id = self._match_id(url)
@@ -96,15 +128,15 @@ class AsianCrushPlaylistIE(InfoExtractor):
                 entries.append(self.url_result(
                     mobj.group('url'), ie=AsianCrushIE.ie_key()))
 
-        title = remove_end(
-            self._html_search_regex(
-                r'(?s)<h1\b[^>]\bid=["\']movieTitle[^>]+>(.+?)</h1>', webpage,
-                'title', default=None) or self._og_search_title(
-                webpage, default=None) or self._html_search_meta(
-                'twitter:title', webpage, 'title',
-                default=None) or self._search_regex(
-                r'<title>([^<]+)</title>', webpage, 'title', fatal=False),
-            ' | AsianCrush')
+        title = self._html_search_regex(
+            r'(?s)<h1\b[^>]\bid=["\']movieTitle[^>]+>(.+?)</h1>', webpage,
+            'title', default=None) or self._og_search_title(
+            webpage, default=None) or self._html_search_meta(
+            'twitter:title', webpage, 'title',
+            default=None) or self._search_regex(
+            r'<title>([^<]+)</title>', webpage, 'title', fatal=False)
+        if title:
+            title = re.sub(r'\s*\|\s*.+?$', '', title)
 
         description = self._og_search_description(
             webpage, default=None) or self._html_search_meta(