[trutube] Support multiple formats (#2433)

test/test_download.py

@@ -18,6 +18,7 @@ from test.helper import (
 import hashlib
 import io
 import json
+import re
 import socket
 
 import youtube_dl.YoutubeDL
@@ -137,12 +138,21 @@ def generator(test_case):
                 with io.open(info_json_fn, encoding='utf-8') as infof:
                     info_dict = json.load(infof)
                 for (info_field, expected) in tc.get('info_dict', {}).items():
-                    if isinstance(expected, compat_str) and expected.startswith('md5:'):
+                    if isinstance(expected, compat_str) and expected.startswith('re:'):
-                        got = 'md5:' + md5(info_dict.get(info_field))
-                    else:
                         got = info_dict.get(info_field)
-                    self.assertEqual(expected, got,
+                        match_str = expected[len('re:'):]
-                        u'invalid value for field %s, expected %r, got %r' % (info_field, expected, got))
+                        match_rex = re.compile(match_str)
+
+                        self.assertTrue(
+                            isinstance(got, compat_str) and match_rex.match(got),
+                            u'field %s (value: %r) should match %r' % (info_field, got, match_str))
+                    else:
+                        if isinstance(expected, compat_str) and expected.startswith('md5:'):
+                            got = 'md5:' + md5(info_dict.get(info_field))
+                        else:
+                            got = info_dict.get(info_field)
+                        self.assertEqual(expected, got,
+                            u'invalid value for field %s, expected %r, got %r' % (info_field, expected, got))
 
                 # If checkable fields are missing from the test case, print the info_dict
                 test_info_dict = dict((key, value if not isinstance(value, compat_str) or len(value) < 250 else 'md5:' + md5(value))

youtube_dl/extractor/trutube.py

@@ -1,3 +1,5 @@
+from __future__ import unicode_literals
+
 import re
 
 from .common import InfoExtractor
@@ -7,37 +9,39 @@ from ..utils import (
 
 
 class TruTubeIE(InfoExtractor):
-    _VALID_URL = r'(?:https?://)?(?:www\.)?(?P<url>trutube\.tv/video/(?P<videoid>.*/.*))'
+    _VALID_URL = r'https?://(?:www\.)?trutube\.tv/video/(?P<id>[0-9]+)/.*'
     _TEST = {
-        'url': ('http://www.trutube.tv/video/20814/Ernst-Zundel-met-les-Jui'
+        'url': 'http://trutube.tv/video/14880/Ramses-II-Proven-To-Be-A-Red-Headed-Caucasoid-',
-                'fs-en-guarde-VOSTFR'),
+        'md5': 'c5b6e301b0a2040b074746cbeaa26ca1',
-        'md5': '9973aa3c2870626799d2ac4e36cfc3dc',
         'info_dict': {
-            u"title": u"TruTube.TV - Spitting in the face of die-versity",
+            'id': '14880',
-            u"ext": u"mp4"
+            'ext': 'flv',
+            'title': 'Ramses II - Proven To Be A Red Headed Caucasoid',
+            'thumbnail': 're:^http:.*\.jpg$',
         }
     }
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
+        video_id = mobj.group('id')
 
-        video_id = mobj.group('videoid')
-
-        # Get webpage content
         webpage = self._download_webpage(url, video_id)
+        video_title = self._og_search_title(webpage).strip()
+        thumbnail = self._search_regex(
+            r"var splash_img = '([^']+)';", webpage, 'thumbnail', fatal=False)
 
-        # Get the video title
+        all_formats = re.finditer(
-        video_title = self._html_search_regex(r'<title>(?P<title>.*)</title>',
+            r"var (?P<key>[a-z]+)_video_file\s*=\s*'(?P<url>[^']+)';", webpage)
-                                              webpage, 'title').strip()
+        formats = [{
-
+            'format_id': m.group('key'),
-        video_url = self._search_regex(r'(http://.*\.(?:mp4|flv))',
+            'quality': -i,
-                                       webpage, u'video URL')
+            'url': m.group('url'),
-
+        } for i, m in enumerate(all_formats)]
-        ext = video_url[-3:]
+        self._sort_formats(formats)
 
         return {
             'id': video_id,
-            'url': video_url,
             'title': video_title,
-            'ext': ext
+            'formats': formats,
-            }
+            'thumbnail': thumbnail,
+        }