Register new version in the changelog.

[youtubedl] / youtube_dl / extractor / twitter.py

diff --git a/youtube_dl/extractor/twitter.py b/youtube_dl/extractor/twitter.py
index 1aaa0630519fff52ed51f4964121552592ffbaa6..9d3e46b946843ae0da6b9de525c4aa4b8b3f4cbb 100644 (file)
--- a/youtube_dl/extractor/twitter.py
+++ b/youtube_dl/extractor/twitter.py
@@ -1,3 +1,4 @@
+# coding: utf-8

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


@@ -6,23 +7,51 @@ from .common import InfoExtractor
 from ..compat import compat_urllib_request
 from ..utils import (
     float_or_none,
 from ..compat import compat_urllib_request
 from ..utils import (
     float_or_none,

-    unescapeHTML,
+    xpath_text,
+    remove_end,

 )
 
 
 class TwitterCardIE(InfoExtractor):
 )
 
 
 class TwitterCardIE(InfoExtractor):

+    IE_NAME = 'twitter:card'

     _VALID_URL = r'https?://(?:www\.)?twitter\.com/i/cards/tfw/v1/(?P<id>\d+)'
     _VALID_URL = r'https?://(?:www\.)?twitter\.com/i/cards/tfw/v1/(?P<id>\d+)'

-    _TEST = {
-        'url': 'https://twitter.com/i/cards/tfw/v1/560070183650213889',
-        'md5': 'a74f50b310c83170319ba16de6955192',
-        'info_dict': {
-            'id': '560070183650213889',
-            'ext': 'mp4',
-            'title': 'TwitterCard',
-            'thumbnail': 're:^https?://.*\.jpg$',
-            'duration': 30.033,
+    _TESTS = [
+        {
+            'url': 'https://twitter.com/i/cards/tfw/v1/560070183650213889',
+            'md5': '7d2f6b4d2eb841a7ccc893d479bfceb4',
+            'info_dict': {
+                'id': '560070183650213889',
+                'ext': 'mp4',
+                'title': 'TwitterCard',
+                'thumbnail': 're:^https?://.*\.jpg$',
+                'duration': 30.033,
+            }

         },
         },

-    }
+        {
+            'url': 'https://twitter.com/i/cards/tfw/v1/623160978427936768',
+            'md5': '7ee2a553b63d1bccba97fbed97d9e1c8',
+            'info_dict': {
+                'id': '623160978427936768',
+                'ext': 'mp4',
+                'title': 'TwitterCard',
+                'thumbnail': 're:^https?://.*\.jpg',
+                'duration': 80.155,
+            },
+        },
+        {
+            'url': 'https://twitter.com/i/cards/tfw/v1/654001591733886977',
+            'md5': 'b6f35e8b08a0bec6c8af77a2f4b3a814',
+            'info_dict': {
+                'id': 'dq4Oj5quskI',
+                'ext': 'mp4',
+                'title': 'Ubuntu 11.10 Overview',
+                'description': 'Take a quick peek at what\'s new and improved in Ubuntu 11.10.\n\nOnce installed take a look at 10 Things to Do After Installing: http://www.omgubuntu.co.uk/2011/10/10-things-to-do-after-installing-ubuntu-11-10/',
+                'upload_date': '20111013',
+                'uploader': 'OMG! Ubuntu!',
+                'uploader_id': 'omgubuntu',
+            },
+        }
+    ]

 
     def _real_extract(self, url):
         video_id = self._match_id(url)
 
     def _real_extract(self, url):
         video_id = self._match_id(url)


@@ -40,10 +69,24 @@ class TwitterCardIE(InfoExtractor):
             request.add_header('User-Agent', user_agent)
             webpage = self._download_webpage(request, video_id)
 
             request.add_header('User-Agent', user_agent)
             webpage = self._download_webpage(request, video_id)
 

-            config = self._parse_json(
-                unescapeHTML(self._search_regex(
-                    r'data-player-config="([^"]+)"', webpage, 'data player config')),
+            youtube_url = self._html_search_regex(
+                r'<iframe[^>]+src="((?:https?:)?//www.youtube.com/embed/[^"]+)"',
+                webpage, 'youtube iframe', default=None)
+            if youtube_url:
+                return self.url_result(youtube_url, 'Youtube')
+
+            config = self._parse_json(self._html_search_regex(
+                r'data-player-config="([^"]+)"', webpage, 'data player config'),

                 video_id)
                 video_id)

+            if 'playlist' not in config:
+                if 'vmapUrl' in config:
+                    vmap_data = self._download_xml(config['vmapUrl'], video_id)
+                    video_url = xpath_text(vmap_data, './/MediaFile').strip()
+                    formats.append({
+                        'url': video_url,
+                    })
+                    break   # same video regardless of UA
+                continue

 
             video_url = config['playlist'][0]['source']
 
 
             video_url = config['playlist'][0]['source']
 


@@ -70,3 +113,54 @@ class TwitterCardIE(InfoExtractor):
             'duration': duration,
             'formats': formats,
         }
             'duration': duration,
             'formats': formats,
         }

+
+
+class TwitterIE(InfoExtractor):
+    IE_NAME = 'twitter'
+    _VALID_URL = r'https?://(?:www\.|m\.|mobile\.)?twitter\.com/(?P<user_id>[^/]+)/status/(?P<id>\d+)'
+    _TEMPLATE_URL = 'https://twitter.com/%s/status/%s'
+
+    _TEST = {
+        'url': 'https://twitter.com/freethenipple/status/643211948184596480',
+        'md5': '31cd83a116fc41f99ae3d909d4caf6a0',
+        'info_dict': {
+            'id': '643211948184596480',
+            'ext': 'mp4',
+            'title': 'FREE THE NIPPLE - FTN supporters on Hollywood Blvd today!',
+            'thumbnail': 're:^https?://.*\.jpg',
+            'duration': 12.922,
+            'description': 'FREE THE NIPPLE on Twitter: "FTN supporters on Hollywood Blvd today! http://t.co/c7jHH749xJ"',
+            'uploader': 'FREE THE NIPPLE',
+            'uploader_id': 'freethenipple',
+        },
+    }
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        user_id = mobj.group('user_id')
+        twid = mobj.group('id')
+
+        webpage = self._download_webpage(self._TEMPLATE_URL % (user_id, twid), twid)
+
+        username = remove_end(self._og_search_title(webpage), ' on Twitter')
+
+        title = self._og_search_description(webpage).strip('').replace('\n', ' ')
+
+        # strip  'https -_t.co_BJYgOjSeGA' junk from filenames
+        mobj = re.match(r'“(.*)\s+(https?://[^ ]+)”', title)
+        title, short_url = mobj.groups()
+
+        card_id = self._search_regex(
+            r'["\']/i/cards/tfw/v1/(\d+)', webpage, 'twitter card url')
+        card_url = 'https://twitter.com/i/cards/tfw/v1/' + card_id
+
+        return {
+            '_type': 'url_transparent',
+            'ie_key': 'TwitterCard',
+            'uploader_id': user_id,
+            'uploader': username,
+            'url': card_url,
+            'webpage_url': url,
+            'description': '%s on Twitter: "%s %s"' % (username, title, short_url),
+            'title': username + ' - ' + title,
+        }