Update upstream source from tag 'upstream/2019.01.16'

[youtubedl] / youtube_dl / extractor / nuvid.py

diff --git a/youtube_dl/extractor/nuvid.py b/youtube_dl/extractor/nuvid.py
index e3db9fe8c6c643d49aa8c521921bc3f508010f98..ab6bfcd7f485218d19034930607d35d1665e7406 100644 (file)
--- a/youtube_dl/extractor/nuvid.py
+++ b/youtube_dl/extractor/nuvid.py
@@ -3,46 +3,69 @@ from __future__ import unicode_literals
 import re
 
 from .common import InfoExtractor
+from ..utils import (
+    parse_duration,
+)
 
 
 class NuvidIE(InfoExtractor):
-    _VALID_URL = r'^https?://(?:www|m)\.nuvid\.com/video/(?P<id>[0-9]+)'
+    _VALID_URL = r'https?://(?:www|m)\.nuvid\.com/video/(?P<id>[0-9]+)'
     _TEST = {
         'url': 'http://m.nuvid.com/video/1310741/',
         'md5': 'eab207b7ac4fccfb4e23c86201f11277',
         'info_dict': {
             'id': '1310741',
             'ext': 'mp4',
-            "title": "Horny babes show their awesome bodeis and",
-            "age_limit": 18,
+            'title': 'Horny babes show their awesome bodeis and',
+            'duration': 129,
+            'age_limit': 18,
         }
     }
 
     def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
-        video_id = mobj.group('id')
-
-        murl = url.replace('://www.', '://m.')
-        webpage = self._download_webpage(murl, video_id)
+        video_id = self._match_id(url)
 
-        title = self._html_search_regex(
-            r'<div class="title">\s+<h2[^>]*>([^<]+)</h2>',
-            webpage, 'title').strip()
+        page_url = 'http://m.nuvid.com/video/%s' % video_id
+        webpage = self._download_webpage(
+            page_url, video_id, 'Downloading video page')
+        # When dwnld_speed exists and has a value larger than the MP4 file's
+        # bitrate, Nuvid returns the MP4 URL
+        # It's unit is 100bytes/millisecond, see mobile-nuvid-min.js for the algorithm
+        self._set_cookie('nuvid.com', 'dwnld_speed', '10.0')
+        mp4_webpage = self._download_webpage(
+            page_url, video_id, 'Downloading video page for MP4 format')
 
-        url_end = self._html_search_regex(
-            r'href="(/[^"]+)"[^>]*data-link_type="mp4"',
-            webpage, 'video_url')
-        video_url = 'http://m.nuvid.com' + url_end
+        html5_video_re = r'(?s)<(?:video|audio)[^<]*(?:>.*?<source[^>]*)?\s+src=["\'](.*?)["\']',
+        video_url = self._html_search_regex(html5_video_re, webpage, video_id)
+        mp4_video_url = self._html_search_regex(html5_video_re, mp4_webpage, video_id)
+        formats = [{
+            'url': video_url,
+        }]
+        if mp4_video_url != video_url:
+            formats.append({
+                'url': mp4_video_url,
+            })
 
-        thumbnail = self._html_search_regex(
-            r'href="(/thumbs/[^"]+)"[^>]*data-link_type="thumbs"',
-            webpage, 'thumbnail URL', fatal=False)
+        title = self._html_search_regex(
+            [r'<span title="([^"]+)">',
+             r'<div class="thumb-holder video">\s*<h5[^>]*>([^<]+)</h5>',
+             r'<span[^>]+class="title_thumb">([^<]+)</span>'], webpage, 'title').strip()
+        thumbnails = [
+            {
+                'url': thumb_url,
+            } for thumb_url in re.findall(r'<img src="([^"]+)" alt="" />', webpage)
+        ]
+        thumbnail = thumbnails[0]['url'] if thumbnails else None
+        duration = parse_duration(self._html_search_regex(
+            [r'<i class="fa fa-clock-o"></i>\s*(\d{2}:\d{2})',
+             r'<span[^>]+class="view_time">([^<]+)</span>'], webpage, 'duration', fatal=False))
 
         return {
             'id': video_id,
-            'url': video_url,
-            'ext': 'mp4',
             'title': title,
+            'thumbnails': thumbnails,
             'thumbnail': thumbnail,
+            'duration': duration,
             'age_limit': 18,
+            'formats': formats,
         }