From 4cfa7883a3b125023bb20f9195e361bef10eb94f Mon Sep 17 00:00:00 2001
From: Laura Liberda <laura@selfisekai.rocks>
Date: Thu, 11 Feb 2021 14:08:07 +0100
Subject: [PATCH] [clip.rs] fix extraction

---
 haruhi_dl/extractor/cliprs.py    | 42 ++++++++++++++++++++++++++------
 haruhi_dl/extractor/pulsembed.py | 15 +++++++++++-
 2 files changed, 49 insertions(+), 8 deletions(-)

diff --git a/haruhi_dl/extractor/cliprs.py b/haruhi_dl/extractor/cliprs.py
index 8dda33c87..4e211298f 100644
--- a/haruhi_dl/extractor/cliprs.py
+++ b/haruhi_dl/extractor/cliprs.py
@@ -2,14 +2,16 @@
 from __future__ import unicode_literals
 
 from .common import InfoExtractor
-from .pulsembed import PulseVideoIE
+from ..utils import (
+    ExtractorError,
+)
+from .pulsembed import PulseVideoIE, PulsEmbedIE
 
 
 class ClipRsIE(InfoExtractor):
     _VALID_URL = r'https?://(?:www\.)?clip\.rs/(?P<id>[^/]+)/\d+'
     _TESTS = [{
-        'url': 'http://www.clip.rs/premijera-frajle-predstavljaju-novi-spot-za-pesmu-moli-me-moli/3732',
-        'md5': 'c412d57815ba07b56f9edc7b5d6a14e5',
+        'url': 'https://www.clip.rs/premijera-frajle-predstavljaju-novi-spot-za-pesmu-moli-me-moli/3732',
         'info_dict': {
             'id': '1488842.1399140381',
             'ext': 'mp4',
@@ -19,16 +21,42 @@ class ClipRsIE(InfoExtractor):
             'timestamp': 1459850243,
             'upload_date': '20160405',
         }
+    }, {
+        'url': 'https://www.clip.rs/u-novom-sadu-se-sinoc-desio-jedan-zimski-neum-svi-su-zaboravili-na-koronu-uhvatili-se-u-kolo-i-nastao-je-hit-video/15686',
+        'info_dict': {
+            'id': '2210721.1689293351',
+            'ext': 'mp4',
+            'title': 'U Novom Sadu se sinoć desio jedan zimski Neum: Svi su zaboravili na koronu, uhvatili se u kolo i nastao je HIT VIDEO',
+            'description': 'md5:b1d7d6c0b029b922f06a2a08c9761852',
+            'timestamp': 1609405068,
+            'upload_date': '20201231',
+        },
     }]
 
     def _real_extract(self, url):
         display_id = self._match_id(url)
 
         webpage = self._download_webpage(url, display_id)
+        info_dict = {}
 
-        return {
+        mvp_id = PulseVideoIE._search_mvp_id(webpage, default=None)
+        if mvp_id:
+            info_dict.update({
+                'url': 'pulsevideo:%s' % PulseVideoIE._search_mvp_id(webpage),
+                'ie_key': PulseVideoIE.ie_key(),
+            })
+        else:
+            entries = PulsEmbedIE._extract_entries(webpage)
+            if not entries:
+                raise ExtractorError('Video ID not found on webpage')
+            if len(entries) > 1:
+                raise ExtractorError('More than 1 PulsEmbed')
+            info_dict.update(entries[0])
+
+        info_dict.update({
             '_type': 'url_transparent',
-            'url': 'pulsevideo:%s' % PulseVideoIE._search_mvp_id(webpage),
-            'ie_key': PulseVideoIE.ie_key(),
+            'title': self._og_search_title(webpage),
+            'description': self._og_search_description(webpage),
             'display_id': display_id,
-        }
+        })
+        return info_dict
diff --git a/haruhi_dl/extractor/pulsembed.py b/haruhi_dl/extractor/pulsembed.py
index 0fac0a7ad..0151053eb 100644
--- a/haruhi_dl/extractor/pulsembed.py
+++ b/haruhi_dl/extractor/pulsembed.py
@@ -43,6 +43,10 @@ class PulseVideoIE(InfoExtractor):
             r'data-(?:params-)?mvp=["\'](\d+\.\d+)', webpage)
         if mvp:
             return mvp.group(1)
+        mvp = re.search(
+            r'\sid=(["\']?)mvp:(\d+\.\d+)\1', webpage)
+        if mvp:
+            return mvp.group(2)
         if default != NO_DEFAULT:
             return default
         raise ExtractorError('Could not extract mvp')
@@ -222,7 +226,16 @@ class PulsEmbedIE(InfoExtractor):
                     'url': smuggle_url('pulsembed:%s' % embed.group('id'), {'referer': url}),
                     'ie_key': 'PulsEmbed',
                 })
-        return entries
+
+        ids = []
+
+        def dedupe(entry):
+            if entry['url'] not in ids:
+                ids.append(entry['url'])
+                return True
+            return False
+
+        return list(filter(dedupe, entries))
 
     def _real_extract(self, url):
         video_id = self._match_id(url)