[tumblr] Added support for HD video where available (#7036)

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

diff --git a/youtube_dl/extractor/tumblr.py b/youtube_dl/extractor/tumblr.py
index 2a1ae5a717cf7b2af16bf5a1ce3ef7494e28a7a6..71cced56207892c80463546e877d598376a6a93f 100644 (file)
--- a/youtube_dl/extractor/tumblr.py
+++ b/youtube_dl/extractor/tumblr.py
@@ -28,6 +28,54 @@ class TumblrIE(InfoExtractor):
             'description': 'md5:dba62ac8639482759c8eb10ce474586a',
             'thumbnail': 're:http://.*\.jpg',
         }
+    }, {
+        'url': 'http://hdvideotest.tumblr.com/post/130323439814/test-description-for-my-hd-video',
+        'md5': '99a84522f60972bf064a0b80f87bcbb5',
+        'info_dict': {
+            'id': '130323439814',
+            'ext': 'mp4',
+            'title': 'HD Video Testing \u2014 Test description for my HD video',
+            'description': 'md5:97cc3ab5fcd27ee4af6356701541319c',
+            'thumbnail': 're:http://.*\.jpg',
+        },
+        'params': {
+            'format': 'sd',
+        },
+    }, {
+        'url': 'http://hdvideotest.tumblr.com/post/130323439814/test-description-for-my-hd-video',
+        'md5': '7ae503065ad150122dc3089f8cf1546c',
+        'info_dict': {
+            'id': '130323439814',
+            'ext': 'mp4',
+            'title': 'HD Video Testing \u2014 Test description for my HD video',
+            'description': 'md5:97cc3ab5fcd27ee4af6356701541319c',
+            'thumbnail': 're:http://.*\.jpg',
+        },
+        'params': {
+            'format': 'hd',
+        },
+    }, {
+        'url': 'http://naked-yogi.tumblr.com/post/118312946248/naked-smoking-stretching',
+        'md5': 'de07e5211d60d4f3a2c3df757ea9f6ab',
+        'info_dict': {
+            'id': 'Wmur',
+            'ext': 'mp4',
+            'title': 'naked smoking & stretching',
+            'upload_date': '20150506',
+            'timestamp': 1430931613,
+        },
+        'add_ie': ['Vidme'],
+    }, {
+        'url': 'http://camdamage.tumblr.com/post/98846056295/',
+        'md5': 'a9e0c8371ea1ca306d6554e3fecf50b6',
+        'info_dict': {
+            'id': '105463834',
+            'ext': 'mp4',
+            'title': 'Cam Damage-HD 720p',
+            'uploader': 'John Moyer',
+            'uploader_id': 'user32021558',
+        },
+        'add_ie': ['Vimeo'],
     }]
 
     def _real_extract(self, url):
@@ -35,15 +83,45 @@ class TumblrIE(InfoExtractor):
         video_id = m_url.group('id')
         blog = m_url.group('blog_name')
 
+        video_urls = []
+
         url = 'http://%s.tumblr.com/post/%s/' % (blog, video_id)
-        webpage = self._download_webpage(url, video_id)
+        webpage, urlh = self._download_webpage_handle(url, video_id)
 
         iframe_url = self._search_regex(
             r'src=\'(https?://www\.tumblr\.com/video/[^\']+)\'',
-            webpage, 'iframe url')
-        iframe = self._download_webpage(iframe_url, video_id)
-        video_url = self._search_regex(r'<source src="([^"]+)"',
-                                       iframe, 'video url')
+            webpage, 'iframe url', default=None)
+        if iframe_url is None:
+            return self.url_result(urlh.geturl(), 'Generic')
+
+        iframe = self._download_webpage(iframe_url, video_id,
+                                        'Downloading iframe page')
+
+        sd_video_url = self._search_regex(r'<source src="([^"]+)"',
+                                          iframe, 'sd video url')
+        resolution_id = sd_video_url.split("/")[-1] + 'p'
+        if len(resolution_id) != 4:
+            resolution_id = None
+        video_urls.append({
+            'ext': 'mp4',
+            'format_id': 'sd',
+            'url': sd_video_url,
+            'resolution': resolution_id,
+        })
+
+        hd_video_url = self._search_regex(r'hdUrl":"([^"]+)"', iframe,
+                                          'hd video url', default=None)
+        if hd_video_url:
+            hd_video_url = hd_video_url.replace("\\", "")
+            resolution_id = hd_video_url.split("/")[-1] + 'p'
+            if len(resolution_id) != 4:
+                resolution_id = None
+            video_urls.append({
+                'ext': 'mp4',
+                'format_id': 'hd',
+                'url': hd_video_url,
+                'resolution': resolution_id,
+            })
 
         # The only place where you can get a title, it's not complete,
         # but searching in other places doesn't work for all videos
@@ -53,9 +131,9 @@ class TumblrIE(InfoExtractor):
 
         return {
             'id': video_id,
-            'url': video_url,
+            'formats': video_urls,
             'ext': 'mp4',
             'title': video_title,
-            'description': self._og_search_description(webpage),
-            'thumbnail': self._og_search_thumbnail(webpage),
+            'description': self._og_search_description(webpage, default=None),
+            'thumbnail': self._og_search_thumbnail(webpage, default=None),
         }