[twitter] Use _html_search_regex
[youtube-dl] / youtube_dl / extractor / twitter.py
index 1dd43ff3cbcdaa370aab4016105671e0e8f70cc9..6ff15369c4b58922559173e4d6b3cd211c5b3efe 100644 (file)
@@ -1,3 +1,4 @@
+# coding: utf-8
 from __future__ import unicode_literals
 
 import re
@@ -7,15 +8,17 @@ 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+)'
     _TESTS = [
         {
             'url': 'https://twitter.com/i/cards/tfw/v1/560070183650213889',
-            'md5': 'a74f50b310c83170319ba16de6955192',
+            'md5': '7d2f6b4d2eb841a7ccc893d479bfceb4',
             'info_dict': {
                 'id': '560070183650213889',
                 'ext': 'mp4',
@@ -59,16 +62,11 @@ class TwitterCardIE(InfoExtractor):
                 video_id)
             if 'playlist' not in config:
                 if 'vmapUrl' in config:
-                    webpage = self._download_webpage(config['vmapUrl'], video_id + ' (xml)')
-                    video_url = self._search_regex(
-                        r'<MediaFile>\s*<!\[CDATA\[(https?://.+?)\]\]>', webpage, 'data player config (xml)')
-                    f = {
+                    vmap_data = self._download_xml(config['vmapUrl'], video_id)
+                    video_url = xpath_text(vmap_data, './/MediaFile').strip()
+                    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
 
@@ -99,23 +97,24 @@ class TwitterCardIE(InfoExtractor):
         }
 
 
-class TwitterIE(TwitterCardIE):
+class TwitterIE(InfoExtractor):
+    IE_NAME = 'twitter'
     _VALID_URL = r'https?://(?:www|m|mobile)?\.?twitter\.com/(?P<id>[^/]+/status/\d+)'
 
-    _TESTS = [{
-        'url': 'https://m.twitter.com/thereaIbanksy/status/614301758345490432',
-        'md5': '8bbccb487bd7a31349b775915fcd412f',
+    _TEST = {
+        'url': 'https://twitter.com/freethenipple/status/643211948184596480',
+        'md5': '31cd83a116fc41f99ae3d909d4caf6a0',
         'info_dict': {
-            'id': '614301758345490432',
+            'id': '643211948184596480',
             'ext': 'mp4',
-            'title': 'thereaIbanksy - This time lapse is so pretty \U0001f60d\U0001f60d',
+            'title': 'freethenipple - FTN supporters on Hollywood Blvd today!',
             'thumbnail': 're:^https?://.*\.jpg',
-            'duration': 29.5,
-            'description': 'banksy on Twitter: "This time lapse is so pretty \U0001f60d\U0001f60d http://t.co/QB8DDbqiR1"',
-            'uploader': 'banksy',
-            'uploader_id': 'thereaIbanksy',
+            '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)
@@ -123,7 +122,7 @@ class TwitterIE(TwitterCardIE):
         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'))
+        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: