Imported Debian patch 2015.11.27.1-1.1

[youtubedl] / youtube_dl / extractor / condenast.py

diff --git a/youtube_dl/extractor/condenast.py b/youtube_dl/extractor/condenast.py
index 91c1c1348f587798131459676ffe5444727c5c3b..6f92ae2ed0cd1383f66d21b3a9274be151ca32b2 100644 (file)
--- a/youtube_dl/extractor/condenast.py
+++ b/youtube_dl/extractor/condenast.py
@@ -2,15 +2,17 @@
 from __future__ import unicode_literals
 
 import re
 from __future__ import unicode_literals
 
 import re

-import json

 
 from .common import InfoExtractor
 
 from .common import InfoExtractor

-from ..utils import (
+from ..compat import (

     compat_urllib_parse,
     compat_urllib_parse,

-    orderedSet,

     compat_urllib_parse_urlparse,
     compat_urlparse,
 )
     compat_urllib_parse_urlparse,
     compat_urlparse,
 )

+from ..utils import (
+    orderedSet,
+    remove_end,
+)

 
 
 class CondeNastIE(InfoExtractor):
 
 
 class CondeNastIE(InfoExtractor):


@@ -22,26 +24,51 @@ class CondeNastIE(InfoExtractor):
     # The keys are the supported sites and the values are the name to be shown
     # to the user and in the extractor description.
     _SITES = {
     # The keys are the supported sites and the values are the name to be shown
     # to the user and in the extractor description.
     _SITES = {

-        'wired': 'WIRED',
+        'allure': 'Allure',
+        'architecturaldigest': 'Architectural Digest',
+        'arstechnica': 'Ars Technica',
+        'bonappetit': 'Bon Appétit',
+        'brides': 'Brides',
+        'cnevids': 'Condé Nast',
+        'cntraveler': 'Condé Nast Traveler',
+        'details': 'Details',
+        'epicurious': 'Epicurious',
+        'glamour': 'Glamour',
+        'golfdigest': 'Golf Digest',

         'gq': 'GQ',
         'gq': 'GQ',

+        'newyorker': 'The New Yorker',
+        'self': 'SELF',
+        'teenvogue': 'Teen Vogue',
+        'vanityfair': 'Vanity Fair',

         'vogue': 'Vogue',
         'vogue': 'Vogue',

-        'glamour': 'Glamour',
+        'wired': 'WIRED',

         'wmagazine': 'W Magazine',
         'wmagazine': 'W Magazine',

-        'vanityfair': 'Vanity Fair',

     }
 
     }
 

-    _VALID_URL = r'http://(video|www)\.(?P<site>%s)\.com/(?P<type>watch|series|video)/(?P<id>.+)' % '|'.join(_SITES.keys())
+    _VALID_URL = r'http://(?:video|www|player)\.(?P<site>%s)\.com/(?P<type>watch|series|video|embed(?:js)?)/(?P<id>[^/?#]+)' % '|'.join(_SITES.keys())

     IE_DESC = 'Condé Nast media group: %s' % ', '.join(sorted(_SITES.values()))
 
     IE_DESC = 'Condé Nast media group: %s' % ', '.join(sorted(_SITES.values()))
 

-    _TEST = {
+    EMBED_URL = r'(?:https?:)?//player\.(?P<site>%s)\.com/(?P<type>embed(?:js)?)/.+?' % '|'.join(_SITES.keys())
+
+    _TESTS = [{

         'url': 'http://video.wired.com/watch/3d-printed-speakers-lit-with-led',
         'url': 'http://video.wired.com/watch/3d-printed-speakers-lit-with-led',

-        'file': '5171b343c2b4c00dd0c1ccb3.mp4',

         'md5': '1921f713ed48aabd715691f774c451f7',
         'info_dict': {
         'md5': '1921f713ed48aabd715691f774c451f7',
         'info_dict': {

+            'id': '5171b343c2b4c00dd0c1ccb3',
+            'ext': 'mp4',

             'title': '3D Printed Speakers Lit With LED',
             'description': 'Check out these beautiful 3D printed LED speakers.  You can\'t actually buy them, but LumiGeek is working on a board that will let you make you\'re own.',
         }
             'title': '3D Printed Speakers Lit With LED',
             'description': 'Check out these beautiful 3D printed LED speakers.  You can\'t actually buy them, but LumiGeek is working on a board that will let you make you\'re own.',
         }

-    }
+    }, {
+        # JS embed
+        'url': 'http://player.cnevids.com/embedjs/55f9cf8b61646d1acf00000c/5511d76261646d5566020000.js',
+        'md5': 'f1a6f9cafb7083bab74a710f65d08999',
+        'info_dict': {
+            'id': '55f9cf8b61646d1acf00000c',
+            'ext': 'mp4',
+            'title': '3D printed TSA Travel Sentry keys really do open TSA locks',
+        }
+    }]

 
     def _extract_series(self, url, webpage):
         title = self._html_search_regex(r'<div class="cne-series-info">.*?<h1>(.+?)</h1>',
 
     def _extract_series(self, url, webpage):
         title = self._html_search_regex(r'<div class="cne-series-info">.*?<h1>(.+?)</h1>',


@@ -55,12 +82,16 @@ class CondeNastIE(InfoExtractor):
         entries = [self.url_result(build_url(path), 'CondeNast') for path in paths]
         return self.playlist_result(entries, playlist_title=title)
 
         entries = [self.url_result(build_url(path), 'CondeNast') for path in paths]
         return self.playlist_result(entries, playlist_title=title)
 

-    def _extract_video(self, webpage):
-        description = self._html_search_regex([r'<div class="cne-video-description">(.+?)</div>',
-                                               r'<div class="video-post-content">(.+?)</div>',
-                                               ],
-                                              webpage, 'description',
-                                              fatal=False, flags=re.DOTALL)
+    def _extract_video(self, webpage, url_type):
+        if url_type != 'embed':
+            description = self._html_search_regex(
+                [
+                    r'<div class="cne-video-description">(.+?)</div>',
+                    r'<div class="video-post-content">(.+?)</div>',
+                ],
+                webpage, 'description', fatal=False, flags=re.DOTALL)
+        else:
+            description = None

         params = self._search_regex(r'var params = {(.+?)}[;,]', webpage,
                                     'player params', flags=re.DOTALL)
         video_id = self._search_regex(r'videoId: [\'"](.+?)[\'"]', params, 'video id')
         params = self._search_regex(r'var params = {(.+?)}[;,]', webpage,
                                     'player params', flags=re.DOTALL)
         video_id = self._search_regex(r'videoId: [\'"](.+?)[\'"]', params, 'video id')


@@ -76,8 +107,8 @@ class CondeNastIE(InfoExtractor):
         info_url = base_info_url + data
         info_page = self._download_webpage(info_url, video_id,
                                            'Downloading video info')
         info_url = base_info_url + data
         info_page = self._download_webpage(info_url, video_id,
                                            'Downloading video info')

-        video_info = self._search_regex(r'var video = ({.+?});', info_page, 'video info')
-        video_info = json.loads(video_info)
+        video_info = self._search_regex(r'var\s+video\s*=\s*({.+?});', info_page, 'video info')
+        video_info = self._parse_json(video_info, video_id)

 
         formats = [{
             'format_id': '%s-%s' % (fdata['type'].split('/')[-1], fdata['quality']),
 
         formats = [{
             'format_id': '%s-%s' % (fdata['type'].split('/')[-1], fdata['quality']),


@@ -99,12 +130,19 @@ class CondeNastIE(InfoExtractor):
         mobj = re.match(self._VALID_URL, url)
         site = mobj.group('site')
         url_type = mobj.group('type')
         mobj = re.match(self._VALID_URL, url)
         site = mobj.group('site')
         url_type = mobj.group('type')

-        id = mobj.group('id')
+        item_id = mobj.group('id')
+
+        # Convert JS embed to regular embed
+        if url_type == 'embedjs':
+            parsed_url = compat_urlparse.urlparse(url)
+            url = compat_urlparse.urlunparse(parsed_url._replace(
+                path=remove_end(parsed_url.path, '.js').replace('/embedjs/', '/embed/')))
+            url_type = 'embed'

 
 

-        self.to_screen(u'Extracting from %s with the Condé Nast extractor' % self._SITES[site])
-        webpage = self._download_webpage(url, id)
+        self.to_screen('Extracting from %s with the Condé Nast extractor' % self._SITES[site])
+        webpage = self._download_webpage(url, item_id)

 
         if url_type == 'series':
             return self._extract_series(url, webpage)
         else:
 
         if url_type == 'series':
             return self._extract_series(url, webpage)
         else:

-            return self._extract_video(webpage)
+            return self._extract_video(webpage, url_type)