Merge branch 'pr-democracynow' of https://github.com/atomicdryad/youtube-dl into...

[youtube-dl] / youtube_dl / extractor / twitter.py

diff --git a/youtube_dl/extractor/twitter.py b/youtube_dl/extractor/twitter.py
index 37a9fd5fdf7fcfb79b17063c298aedcd017bdfc7..9d3e46b946843ae0da6b9de525c4aa4b8b3f4cbb 100644 (file)
--- a/youtube_dl/extractor/twitter.py
+++ b/youtube_dl/extractor/twitter.py
@@ -7,12 +7,13 @@ from .common import InfoExtractor
 from ..compat import compat_urllib_request
 from ..utils import (
     float_or_none,
-    unescapeHTML,
     xpath_text,
+    remove_end,
 )
 
 
 class TwitterCardIE(InfoExtractor):
+    IE_NAME = 'twitter:card'
     _VALID_URL = r'https?://(?:www\.)?twitter\.com/i/cards/tfw/v1/(?P<id>\d+)'
     _TESTS = [
         {
@@ -36,6 +37,19 @@ class TwitterCardIE(InfoExtractor):
                 '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',
+            },
         }
     ]
 
@@ -55,21 +69,22 @@ class TwitterCardIE(InfoExtractor):
             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)
             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()
-                    f = {
+                    formats.append({
                         'url': video_url,
-                    }
-                    ext = re.search(r'\.([a-z0-9]{2,4})(\?.+)?$', video_url)
-                    if ext:
-                        f['ext'] = ext.group(1)
-                    formats.append(f)
+                    })
                     break   # same video regardless of UA
                 continue
 
@@ -100,45 +115,52 @@ class TwitterCardIE(InfoExtractor):
         }
 
 
-class TwitterIE(TwitterCardIE):
-    _VALID_URL = r'https?://(?:www|m|mobile)?\.?twitter\.com/(?P<id>[^/]+/status/\d+)'
+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'
 
-    _TESTS = [{
+    _TEST = {
         'url': 'https://twitter.com/freethenipple/status/643211948184596480',
         'md5': '31cd83a116fc41f99ae3d909d4caf6a0',
         'info_dict': {
             'id': '643211948184596480',
             'ext': 'mp4',
-            'title': 'freethenipple - FTN supporters on Hollywood Blvd today!',
+            '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):
-        id = self._match_id(url)
-        username, twid = re.match(r'([^/]+)/status/(\d+)', id).groups()
-        name = username
-        url = re.sub(r'https?://(m|mobile)\.', 'https://', url)
-        webpage = self._download_webpage(url, 'tweet: ' + url)
-        description = unescapeHTML(self._search_regex('<title>\s*(.+?)\s*</title>', webpage, 'title'))
-        title = description.replace('\n', ' ')
-        splitdesc = re.match(r'^(.+?)\s*on Twitter:\s* "(.+?)"$', title)
-        if splitdesc:
-            name, title = splitdesc.groups()
-        title = re.sub(r'\s*https?://[^ ]+', '', title)  # strip  'https -_t.co_BJYgOjSeGA' junk from filenames
-        card_id = self._search_regex(r'["\']/i/cards/tfw/v1/(\d+)', webpage, '/i/card/...')
+        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': username,
-            'uploader': name,
+            'uploader_id': user_id,
+            'uploader': username,
             'url': card_url,
             'webpage_url': url,
-            'description': description,
+            'description': '%s on Twitter: "%s %s"' % (username, title, short_url),
             'title': username + ' - ' + title,
         }