[twitter] Use _html_search_regex

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

diff --git a/youtube_dl/extractor/twitter.py b/youtube_dl/extractor/twitter.py
index 1aaa0630519fff52ed51f4964121552592ffbaa6..6ff15369c4b58922559173e4d6b3cd211c5b3efe 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
@@ -7,22 +8,37 @@ from ..compat import compat_urllib_request
 from ..utils import (
     float_or_none,
     unescapeHTML,
+    xpath_text,
 )
 
 
 class TwitterCardIE(InfoExtractor):
+    IE_NAME = 'twitter:card'
     _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,
+            },
+        }
+    ]
 
     def _real_extract(self, url):
         video_id = self._match_id(url)
@@ -44,6 +60,15 @@ class TwitterCardIE(InfoExtractor):
                 unescapeHTML(self._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()
+                    formats.append({
+                        'url': video_url,
+                    })
+                    break   # same video regardless of UA
+                continue
 
             video_url = config['playlist'][0]['source']
 
@@ -70,3 +95,48 @@ class TwitterCardIE(InfoExtractor):
             'duration': duration,
             'formats': formats,
         }
+
+
+class TwitterIE(InfoExtractor):
+    IE_NAME = 'twitter'
+    _VALID_URL = r'https?://(?:www|m|mobile)?\.?twitter\.com/(?P<id>[^/]+/status/\d+)'
+
+    _TEST = {
+        'url': 'https://twitter.com/freethenipple/status/643211948184596480',
+        'md5': '31cd83a116fc41f99ae3d909d4caf6a0',
+        'info_dict': {
+            'id': '643211948184596480',
+            'ext': 'mp4',
+            'title': 'freethenipple - 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 = self._html_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/...')
+        card_url = 'https://twitter.com/i/cards/tfw/v1/' + card_id
+        return {
+            '_type': 'url_transparent',
+            'ie_key': 'TwitterCard',
+            'uploader_id': username,
+            'uploader': name,
+            'url': card_url,
+            'webpage_url': url,
+            'description': description,
+            'title': username + ' - ' + title,
+        }