Updated version 2017.10.15.1 from 'upstream/2017.10.15.1'

[youtubedl] / youtube_dl / extractor / nuvid.py

diff --git a/youtube_dl/extractor/nuvid.py b/youtube_dl/extractor/nuvid.py
index 280328b78306e5ab332cbb7111127f832c6c9aba..ab6bfcd7f485218d19034930607d35d1665e7406 100644 (file)
--- a/youtube_dl/extractor/nuvid.py
+++ b/youtube_dl/extractor/nuvid.py
@@ -5,13 +5,11 @@ import re
 from .common import InfoExtractor
 from ..utils import (
     parse_duration,
 from .common import InfoExtractor
 from ..utils import (
     parse_duration,

-    unified_strdate,
-    compat_urllib_request,

 )
 
 
 class NuvidIE(InfoExtractor):
 )
 
 
 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',
     _TEST = {
         'url': 'http://m.nuvid.com/video/1310741/',
         'md5': 'eab207b7ac4fccfb4e23c86201f11277',


@@ -20,50 +18,54 @@ class NuvidIE(InfoExtractor):
             'ext': 'mp4',
             'title': 'Horny babes show their awesome bodeis and',
             'duration': 129,
             'ext': 'mp4',
             'title': 'Horny babes show their awesome bodeis and',
             'duration': 129,

-            'upload_date': '20140508',

             'age_limit': 18,
         }
     }
 
     def _real_extract(self, url):
             'age_limit': 18,
         }
     }
 
     def _real_extract(self, url):

-        mobj = re.match(self._VALID_URL, url)
-        video_id = mobj.group('id')
+        video_id = self._match_id(url)

 
 

-        formats = []
+        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')

 
 

-        for dwnld_speed, format_id in [(0, '3gp'), (5, 'mp4')]:
-            request = compat_urllib_request.Request(
-                'http://m.nuvid.com/play/%s' % video_id)
-            request.add_header('Cookie', 'skip_download_page=1; dwnld_speed=%d; adv_show=1' % dwnld_speed)
-            webpage = self._download_webpage(
-                request, video_id, 'Downloading %s page' % format_id)
-            video_url = self._html_search_regex(
-                r'<a href="([^"]+)"\s*>Continue to watch video', webpage, '%s video URL' % format_id, fatal=False)
-            if not video_url:
-                continue
+        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({
             formats.append({

-                'url': video_url,
-                'format_id': format_id,
+                'url': mp4_video_url,

             })
 
             })
 

-        webpage = self._download_webpage(
-            'http://m.nuvid.com/video/%s' % video_id, video_id, 'Downloading video page')

         title = self._html_search_regex(
         title = self._html_search_regex(

-            r'<div class="title">\s+<h2[^>]*>([^<]+)</h2>', webpage, 'title').strip()
-        thumbnail = self._html_search_regex(
-            r'href="(/thumbs/[^"]+)"[^>]*data-link_type="thumbs"',
-            webpage, 'thumbnail URL', fatal=False)
+            [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(
         duration = parse_duration(self._html_search_regex(

-            r'Length:\s*<span>(\d{2}:\d{2})</span>',webpage, 'duration', fatal=False))
-        upload_date = unified_strdate(self._html_search_regex(
-            r'Added:\s*<span>(\d{4}-\d{2}-\d{2})</span>', webpage, 'upload date', fatal=False))
+            [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,
             'title': title,
 
         return {
             'id': video_id,
             'title': title,

-            'thumbnail': 'http://m.nuvid.com%s' % thumbnail,
+            'thumbnails': thumbnails,
+            'thumbnail': thumbnail,

             'duration': duration,
             'duration': duration,

-            'upload_date': upload_date,

             'age_limit': 18,
             'formats': formats,
             'age_limit': 18,
             'formats': formats,

-        }
\ No newline at end of file
+        }