Merge branch 'dcn' of github.com:remitamine/youtube-dl into remitamine-dcn

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

diff --git a/youtube_dl/extractor/lrt.py b/youtube_dl/extractor/lrt.py
index fca0bfef0726b472e484d7be2fd890fb5ce50f8c..863efd896b12afa1b674dfb1bcde6d6ffe0214fb 100644 (file)
--- a/youtube_dl/extractor/lrt.py
+++ b/youtube_dl/extractor/lrt.py
@@ -1,13 +1,9 @@
 # coding: utf-8
 from __future__ import unicode_literals
 
-import re
-import json
-
 from .common import InfoExtractor
 from ..utils import (
-    determine_ext,
-    js_to_json,
+    int_or_none,
     parse_duration,
     remove_end,
 )
@@ -22,42 +18,38 @@ class LRTIE(InfoExtractor):
             'id': '54391',
             'ext': 'mp4',
             'title': 'Septynios Kauno dienos',
-            'description': 'Kauno miesto ir apskrities naujienos',
+            'description': 'md5:24d84534c7dc76581e59f5689462411a',
             'duration': 1783,
+            'view_count': int,
+            'like_count': int,
         },
         'params': {
-            'skip_download': True,  # HLS download
+            'skip_download': True,  # m3u8 download
         },
-
     }
 
     def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
-        video_id = mobj.group('id')
-
+        video_id = self._match_id(url)
         webpage = self._download_webpage(url, video_id)
 
         title = remove_end(self._og_search_title(webpage), ' - LRT')
+        m3u8_url = self._search_regex(
+            r'file\s*:\s*(["\'])(?P<url>.+?)\1\s*\+\s*location\.hash\.substring\(1\)',
+            webpage, 'm3u8 url', group='url')
+        formats = self._extract_m3u8_formats(m3u8_url, video_id, 'mp4')
+
         thumbnail = self._og_search_thumbnail(webpage)
         description = self._og_search_description(webpage)
         duration = parse_duration(self._search_regex(
-            r"'duration':\s*'([^']+)',", webpage,
-            'duration', fatal=False, default=None))
+            r'var\s+record_len\s*=\s*(["\'])(?P<duration>[0-9]+:[0-9]+:[0-9]+)\1',
+            webpage, 'duration', default=None, group='duration'))
 
-        formats = []
-        for js in re.findall(r'(?s)config:\s*(\{.*?\})', webpage):
-            data = json.loads(js_to_json(js))
-            if data['provider'] == 'rtmp':
-                formats.append({
-                    'format_id': 'rtmp',
-                    'ext': determine_ext(data['file']),
-                    'url': data['streamer'],
-                    'play_path': 'mp4:%s' % data['file'],
-                    'preference': -1,
-                })
-            else:
-                formats.extend(
-                    self._extract_m3u8_formats(data['file'], video_id, 'mp4'))
+        view_count = int_or_none(self._html_search_regex(
+            r'<div[^>]+class=(["\']).*?record-desc-seen.*?\1[^>]*>(?P<count>.+?)</div>',
+            webpage, 'view count', fatal=False, group='count'))
+        like_count = int_or_none(self._search_regex(
+            r'<span[^>]+id=(["\'])flikesCount.*?\1>(?P<count>\d+)<',
+            webpage, 'like count', fatal=False, group='count'))
 
         return {
             'id': video_id,
@@ -66,4 +58,6 @@ class LRTIE(InfoExtractor):
             'thumbnail': thumbnail,
             'description': description,
             'duration': duration,
+            'view_count': view_count,
+            'like_count': like_count,
         }