Update upstream source from tag 'upstream/2017.11.06'

[youtubedl] / youtube_dl / extractor / bloomberg.py

diff --git a/youtube_dl/extractor/bloomberg.py b/youtube_dl/extractor/bloomberg.py
index 4a88ccd13caf604f3ea892c6784d603434fb06ee..2fbfad1ba04adf915c41441173a6464c6b96b100 100644 (file)
--- a/youtube_dl/extractor/bloomberg.py
+++ b/youtube_dl/extractor/bloomberg.py
@@ -1,3 +1,4 @@
+# coding: utf-8

 from __future__ import unicode_literals
 
 import re
 from __future__ import unicode_literals
 
 import re


@@ -6,32 +7,77 @@ from .common import InfoExtractor
 
 
 class BloombergIE(InfoExtractor):
 
 
 class BloombergIE(InfoExtractor):

-    _VALID_URL = r'https?://www\.bloomberg\.com/video/(?P<id>.+?)\.html'
+    _VALID_URL = r'https?://(?:www\.)?bloomberg\.com/(?:[^/]+/)*(?P<id>[^/?#]+)'

 
 

-    _TEST = {
-        'url': 'http://www.bloomberg.com/video/shah-s-presentation-on-foreign-exchange-strategies-qurhIVlJSB6hzkVi229d8g.html',
+    _TESTS = [{
+        'url': 'http://www.bloomberg.com/news/videos/b/aaeae121-5949-481e-a1ce-4562db6f5df2',

         # The md5 checksum changes
         'info_dict': {
             'id': 'qurhIVlJSB6hzkVi229d8g',
             'ext': 'flv',
             'title': 'Shah\'s Presentation on Foreign-Exchange Strategies',
         # The md5 checksum changes
         'info_dict': {
             'id': 'qurhIVlJSB6hzkVi229d8g',
             'ext': 'flv',
             'title': 'Shah\'s Presentation on Foreign-Exchange Strategies',

-            'description': 'md5:0681e0d30dcdfc6abf34594961d8ea88',
+            'description': 'md5:a8ba0302912d03d246979735c17d2761',

         },
         },

-    }
+        'params': {
+            'format': 'best[format_id^=hds]',
+        },
+    }, {
+        # video ID in BPlayer(...)
+        'url': 'http://www.bloomberg.com/features/2016-hello-world-new-zealand/',
+        'info_dict': {
+            'id': '938c7e72-3f25-4ddb-8b85-a9be731baa74',
+            'ext': 'flv',
+            'title': 'Meet the Real-Life Tech Wizards of Middle Earth',
+            'description': 'Hello World, Episode 1: New Zealand’s freaky AI babies, robot exoskeletons, and a virtual you.',
+        },
+        'params': {
+            'format': 'best[format_id^=hds]',
+        },
+    }, {
+        # data-bmmrid=
+        'url': 'https://www.bloomberg.com/politics/articles/2017-02-08/le-pen-aide-briefed-french-central-banker-on-plan-to-print-money',
+        'only_matching': True,
+    }, {
+        'url': 'http://www.bloomberg.com/news/articles/2015-11-12/five-strange-things-that-have-been-happening-in-financial-markets',
+        'only_matching': True,
+    }, {
+        'url': 'http://www.bloomberg.com/politics/videos/2015-11-25/karl-rove-on-jeb-bush-s-struggles-stopping-trump',
+        'only_matching': True,
+    }]

 
     def _real_extract(self, url):
         name = self._match_id(url)
         webpage = self._download_webpage(url, name)
 
     def _real_extract(self, url):
         name = self._match_id(url)
         webpage = self._download_webpage(url, name)

-
-        f4m_url = self._search_regex(
-            r'<source src="(https?://[^"]+\.f4m.*?)"', webpage,
-            'f4m url')
+        video_id = self._search_regex(
+            (r'["\']bmmrId["\']\s*:\s*(["\'])(?P<id>(?:(?!\1).)+)\1',
+             r'videoId\s*:\s*(["\'])(?P<id>(?:(?!\1).)+)\1',
+             r'data-bmmrid=(["\'])(?P<id>(?:(?!\1).)+)\1'),
+            webpage, 'id', group='id', default=None)
+        if not video_id:
+            bplayer_data = self._parse_json(self._search_regex(
+                r'BPlayer\(null,\s*({[^;]+})\);', webpage, 'id'), name)
+            video_id = bplayer_data['id']

         title = re.sub(': Video$', '', self._og_search_title(webpage))
 
         title = re.sub(': Video$', '', self._og_search_title(webpage))
 

+        embed_info = self._download_json(
+            'http://www.bloomberg.com/api/embed?id=%s' % video_id, video_id)
+        formats = []
+        for stream in embed_info['streams']:
+            stream_url = stream.get('url')
+            if not stream_url:
+                continue
+            if stream['muxing_format'] == 'TS':
+                formats.extend(self._extract_m3u8_formats(
+                    stream_url, video_id, 'mp4', m3u8_id='hls', fatal=False))
+            else:
+                formats.extend(self._extract_f4m_formats(
+                    stream_url, video_id, f4m_id='hds', fatal=False))
+        self._sort_formats(formats)
+

         return {
         return {

-            'id': name.split('-')[-1],
+            'id': video_id,

             'title': title,
             'title': title,

-            'formats': self._extract_f4m_formats(f4m_url, name),
+            'formats': formats,

             'description': self._og_search_description(webpage),
             'thumbnail': self._og_search_thumbnail(webpage),
         }
             'description': self._og_search_description(webpage),
             'thumbnail': self._og_search_thumbnail(webpage),
         }