Prepare to release.

[youtubedl] / youtube_dl / extractor / funnyordie.py

diff --git a/youtube_dl/extractor/funnyordie.py b/youtube_dl/extractor/funnyordie.py
index ec6d96adaeff666bf0fea7fe78e766c6a6ac2808..49409369cc5e72e626b5fbe64f8a346d9695f802 100644 (file)
--- a/youtube_dl/extractor/funnyordie.py
+++ b/youtube_dl/extractor/funnyordie.py
@@ -8,7 +8,7 @@ from ..utils import ExtractorError
 
 
 class FunnyOrDieIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?funnyordie\.com/(?P<type>embed|videos)/(?P<id>[0-9a-f]+)(?:$|[?#/])'
+    _VALID_URL = r'https?://(?:www\.)?funnyordie\.com/(?P<type>embed|articles|videos)/(?P<id>[0-9a-f]+)(?:$|[?#/])'
     _TESTS = [{
         'url': 'http://www.funnyordie.com/videos/0732f586d7/heart-shaped-box-literal-video-version',
         'md5': 'bcd81e0c4f26189ee09be362ad6e6ba9',
@@ -17,18 +17,23 @@ class FunnyOrDieIE(InfoExtractor):
             'ext': 'mp4',
             'title': 'Heart-Shaped Box: Literal Video Version',
             'description': 'md5:ea09a01bc9a1c46d9ab696c01747c338',
-            'thumbnail': 're:^http:.*\.jpg$',
+            'thumbnail': r're:^http:.*\.jpg$',
         },
     }, {
         'url': 'http://www.funnyordie.com/embed/e402820827',
-        'md5': '29f4c5e5a61ca39dfd7e8348a75d0aad',
         'info_dict': {
             'id': 'e402820827',
             'ext': 'mp4',
             'title': 'Please Use This Song (Jon Lajoie)',
             'description': 'Please use this to sell something.  www.jonlajoie.com',
-            'thumbnail': 're:^http:.*\.jpg$',
+            'thumbnail': r're:^http:.*\.jpg$',
         },
+        'params': {
+            'skip_download': True,
+        },
+    }, {
+        'url': 'http://www.funnyordie.com/articles/ebf5e34fc8/10-hours-of-walking-in-nyc-as-a-man',
+        'only_matching': True,
     }]
 
     def _real_extract(self, url):
@@ -43,19 +48,57 @@ class FunnyOrDieIE(InfoExtractor):
 
         links.sort(key=lambda link: 1 if link[1] == 'mp4' else 0)
 
-        bitrates = self._html_search_regex(r'<source src="[^"]+/v,((?:\d+,)+)\.mp4\.csmil', webpage, 'video bitrates')
-        bitrates = [int(b) for b in bitrates.rstrip(',').split(',')]
-        bitrates.sort()
+        m3u8_url = self._search_regex(
+            r'<source[^>]+src=(["\'])(?P<url>.+?/master\.m3u8[^"\']*)\1',
+            webpage, 'm3u8 url', group='url')
 
         formats = []
 
-        for bitrate in bitrates:
-            for link in links:
-                formats.append({
-                    'url': '%s%d.%s' % (link[0], bitrate, link[1]),
-                    'format_id': '%s-%d' % (link[1], bitrate),
-                    'vbr': bitrate,
-                })
+        m3u8_formats = self._extract_m3u8_formats(
+            m3u8_url, video_id, 'mp4', 'm3u8_native',
+            m3u8_id='hls', fatal=False)
+        source_formats = list(filter(
+            lambda f: f.get('vcodec') != 'none', m3u8_formats))
+
+        bitrates = [int(bitrate) for bitrate in re.findall(r'[,/]v(\d+)(?=[,/])', m3u8_url)]
+        bitrates.sort()
+
+        if source_formats:
+            self._sort_formats(source_formats)
+
+        for bitrate, f in zip(bitrates, source_formats or [{}] * len(bitrates)):
+            for path, ext in links:
+                ff = f.copy()
+                if ff:
+                    if ext != 'mp4':
+                        ff = dict(
+                            [(k, v) for k, v in ff.items()
+                             if k in ('height', 'width', 'format_id')])
+                    ff.update({
+                        'format_id': ff['format_id'].replace('hls', ext),
+                        'ext': ext,
+                        'protocol': 'http',
+                    })
+                else:
+                    ff.update({
+                        'format_id': '%s-%d' % (ext, bitrate),
+                        'vbr': bitrate,
+                    })
+                ff['url'] = self._proto_relative_url(
+                    '%s%d.%s' % (path, bitrate, ext))
+                formats.append(ff)
+        self._check_formats(formats, video_id)
+
+        formats.extend(m3u8_formats)
+        self._sort_formats(
+            formats, field_preference=('height', 'width', 'tbr', 'format_id'))
+
+        subtitles = {}
+        for src, src_lang in re.findall(r'<track kind="captions" src="([^"]+)" srclang="([^"]+)"', webpage):
+            subtitles[src_lang] = [{
+                'ext': src.split('/')[-1],
+                'url': 'http://www.funnyordie.com%s' % src,
+            }]
 
         post_json = self._search_regex(
             r'fb_post\s*=\s*(\{.*?\});', webpage, 'post details')
@@ -67,4 +110,5 @@ class FunnyOrDieIE(InfoExtractor):
             'description': post.get('description'),
             'thumbnail': post.get('picture'),
             'formats': formats,
+            'subtitles': subtitles,
         }