From 44f48e75218d40e53ece247f3bd2a724888536ba Mon Sep 17 00:00:00 2001
From: Ales Jirasek <schunkac@gmail.com>
Date: Wed, 10 Oct 2018 23:47:21 +0200
Subject: [PATCH] [MallTV] Add new extractor, better regexe update generic
 JSON_LD_RE to find json_ld in attributes without quotes, closes #18058

---
 youtube_dl/extractor/extractors.py |  3 +-
 youtube_dl/extractor/malltv.py     | 71 ++++++++++++++++++++++++++++++
 youtube_dl/utils.py                |  2 +-
 3 files changed, 74 insertions(+), 2 deletions(-)
 create mode 100644 youtube_dl/extractor/malltv.py

diff --git a/youtube_dl/extractor/extractors.py b/youtube_dl/extractor/extractors.py
index 464c8d690..268b2cbb1 100644
--- a/youtube_dl/extractor/extractors.py
+++ b/youtube_dl/extractor/extractors.py
@@ -595,6 +595,7 @@ from .mailru import (
     MailRuMusicSearchIE,
 )
 from .makertv import MakerTVIE
+from .malltv import MallTVIE
 from .mangomolo import (
     MangomoloVideoIE,
     MangomoloLiveIE,
@@ -1367,7 +1368,7 @@ from .webofstories import (
     WebOfStoriesPlaylistIE,
 )
 from .weibo import (
-    WeiboIE, 
+    WeiboIE,
     WeiboMobileIE
 )
 from .weiqitv import WeiqiTVIE
diff --git a/youtube_dl/extractor/malltv.py b/youtube_dl/extractor/malltv.py
new file mode 100644
index 000000000..512d999c9
--- /dev/null
+++ b/youtube_dl/extractor/malltv.py
@@ -0,0 +1,71 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+from .common import InfoExtractor
+from ..utils import parse_duration
+
+
+class MallTVIE(InfoExtractor):
+    _VALID_URL = r'https?://(?:www\.)?mall\.tv/(?:.+/)?(?P<id>.+)'
+    _TESTS = [
+        {
+            'url': ('https://www.mall.tv/18-miliard-pro-neziskovky'
+                    '-opravdu-jsou-sportovci-nebo-clovek-v-tisni-pijavice'),
+            'info_dict': {
+                'id': ('18-miliard-pro-neziskovky-opravdu-jsou-sportovci-nebo-'
+                       'clovek-v-tisni-pijavice'),
+                'ext': 'mp4',
+                'title': ('18 miliard pro neziskovky. Opravdu jsou sportovci '
+                          'nebo Člověk v tísni pijavice?'),
+                'description': ('Pokud někdo hospodaří s penězmi daňových '
+                                'poplatníků, pak logicky chceme vědět, jak s '
+                                'nimi nakládá. Objem dotací pro neziskovky '
+                                'roste, ale opravdu jsou tyto organizace '
+                                '„pijavice", jak o nich hovoří And')
+                },
+            'params': {
+                'skip_download': True
+            }
+        },
+        {
+            'url': ('https://www.mall.tv/kdo-to-plati/18-miliard-pro-neziskovky'
+                    '-opravdu-jsou-sportovci-nebo-clovek-v-tisni-pijavice'),
+            'info_dict': {
+                'id': ('18-miliard-pro-neziskovky-opravdu-jsou-sportovci-nebo-'
+                       'clovek-v-tisni-pijavice'),
+                'ext': 'mp4',
+                'title': ('18 miliard pro neziskovky. Opravdu jsou sportovci '
+                          'nebo Člověk v tísni pijavice?'),
+                'description': ('Pokud někdo hospodaří s penězmi daňových '
+                                'poplatníků, pak logicky chceme vědět, jak s '
+                                'nimi nakládá. Objem dotací pro neziskovky '
+                                'roste, ale opravdu jsou tyto organizace '
+                                '„pijavice", jak o nich hovoří And')
+                },
+            'params': {
+                'skip_download': True
+            }
+        },
+    ]
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+        webpage = self._download_webpage(url, video_id)
+        info = self._search_json_ld(webpage, video_id, default={})
+
+        format_url = self._html_search_regex(
+            r'<source src=([\"\'])?(?P<src>.+?index)\1?[^>]*?>',
+            webpage, 'm3u8 URL', group='src')
+        formats = self._extract_m3u8_formats(format_url+'.m3u8',
+                                             video_id, 'mp4')
+        self._sort_formats(formats)
+        title = info.get('title', self._og_search_title(webpage, fatal=False))
+        thumbnail = info.get('thumbnailUrl', self._og_search_thumbnail(webpage))
+        return {
+            'id': video_id,
+            'title': title,
+            'thumbnail': thumbnail,
+            'description': self._og_search_description(webpage),
+            'duration': parse_duration(info.get('duration')),
+            'formats': formats
+        }
diff --git a/youtube_dl/utils.py b/youtube_dl/utils.py
index e84d35d4d..4f21d8821 100644
--- a/youtube_dl/utils.py
+++ b/youtube_dl/utils.py
@@ -183,7 +183,7 @@ DATE_FORMATS_MONTH_FIRST.extend([
 ])
 
 PACKED_CODES_RE = r"}\('(.+)',(\d+),(\d+),'([^']+)'\.split\('\|'\)"
-JSON_LD_RE = r'(?is)<script[^>]+type=(["\'])application/ld\+json\1[^>]*>(?P<json_ld>.+?)</script>'
+JSON_LD_RE = r'(?is)<script[^>]+type=(["\'])?application/ld\+json\1?[^>]*>(?P<json_ld>.+?)</script>'
 
 
 def preferredencoding():