release 2014.09.06

Conflicts:
	youtube_dl/extractor/__init__.py

test/test_cache.py

@@ -43,16 +43,16 @@ class TestCache(unittest.TestCase):
         })
         c = Cache(ydl)
         obj = {'x': 1, 'y': ['ä', '\\a', True]}
-        self.assertEqual(c.load('test_cache', 'k'), None)
-        c.store('test_cache', 'k', obj)
+        self.assertEqual(c.load('test_cache', 'k.'), None)
+        c.store('test_cache', 'k.', obj)
         self.assertEqual(c.load('test_cache', 'k2'), None)
         self.assertFalse(_is_empty(self.test_dir))
-        self.assertEqual(c.load('test_cache', 'k'), obj)
+        self.assertEqual(c.load('test_cache', 'k.'), obj)
         self.assertEqual(c.load('test_cache', 'y'), None)
-        self.assertEqual(c.load('test_cache2', 'k'), None)
+        self.assertEqual(c.load('test_cache2', 'k.'), None)
         c.remove()
         self.assertFalse(os.path.exists(self.test_dir))
-        self.assertEqual(c.load('test_cache', 'k'), None)
+        self.assertEqual(c.load('test_cache', 'k.'), None)
 
 
 if __name__ == '__main__':

youtube_dl/__init__.py

@@ -74,6 +74,7 @@ __authors__  = (
     'Keith Beckman',
     'Ole Ernst',
     'Aaron McDaniel (mcd1992)',
+    'Magnus Kolstad',
 )
 
 __license__ = 'Public Domain'

youtube_dl/downloader/http.py

@@ -193,7 +193,8 @@ class HttpFD(FileDownloader):
             self.to_stderr(u"\n")
             self.report_error(u'Did not get any data blocks')
             return False
-        stream.close()
+        if tmpfilename != u'-':
+            stream.close()
         self.report_finish(data_len_str, (time.time() - start))
         if data_len is not None and byte_counter != data_len:
             raise ContentTooShortError(byte_counter, int(data_len))

youtube_dl/extractor/__init__.py

@@ -67,6 +67,7 @@ from .dailymotion import (
     DailymotionUserIE,
 )
 from .daum import DaumIE
+from .dbtv import DBTVIE
 from .dfb import DFBIE
 from .dotsub import DotsubIE
 from .dreisat import DreiSatIE

youtube_dl/extractor/arte.py

@@ -78,7 +78,8 @@ class ArteTVPlus7IE(InfoExtractor):
 
     def _extract_from_webpage(self, webpage, video_id, lang):
         json_url = self._html_search_regex(
-            r'arte_vp_url="(.*?)"', webpage, 'json vp url')
+            [r'arte_vp_url=["\'](.*?)["\']', r'data-url=["\']([^"]+)["\']'],
+            webpage, 'json vp url')
         return self._extract_from_json_url(json_url, video_id, lang)
 
     def _extract_from_json_url(self, json_url, video_id, lang):

youtube_dl/extractor/cliphunter.py

@@ -1,11 +1,13 @@
 from __future__ import unicode_literals
 
+import json
 import re
 
 from .common import InfoExtractor
+from ..utils import int_or_none
 
 
-translation_table = {
+_translation_table = {
     'a': 'h', 'd': 'e', 'e': 'v', 'f': 'o', 'g': 'f', 'i': 'd', 'l': 'n',
     'm': 'a', 'n': 'm', 'p': 'u', 'q': 't', 'r': 's', 'v': 'p', 'x': 'r',
     'y': 'l', 'z': 'i',
@@ -13,6 +15,10 @@ translation_table = {
 }
 
 
+def _decode(s):
+    return ''.join(_translation_table.get(c, c) for c in s)
+
+
 class CliphunterIE(InfoExtractor):
     IE_NAME = 'cliphunter'
 
@@ -22,10 +28,14 @@ class CliphunterIE(InfoExtractor):
     '''
     _TEST = {
         'url': 'http://www.cliphunter.com/w/1012420/Fun_Jynx_Maze_solo',
-        'file': '1012420.flv',
-        'md5': '15e7740f30428abf70f4223478dc1225',
+        'md5': 'a2ba71eebf523859fe527a61018f723e',
         'info_dict': {
+            'id': '1012420',
+            'ext': 'mp4',
             'title': 'Fun Jynx Maze solo',
+            'thumbnail': 're:^https?://.*\.jpg$',
+            'age_limit': 18,
+            'duration': 1317,
         }
     }
 
@@ -35,22 +45,55 @@ class CliphunterIE(InfoExtractor):
 
         webpage = self._download_webpage(url, video_id)
 
+        video_title = self._search_regex(
+            r'mediaTitle = "([^"]+)"', webpage, 'title')
+
         pl_fiji = self._search_regex(
             r'pl_fiji = \'([^\']+)\'', webpage, 'video data')
         pl_c_qual = self._search_regex(
             r'pl_c_qual = "(.)"', webpage, 'video quality')
-        video_title = self._search_regex(
-            r'mediaTitle = "([^"]+)"', webpage, 'title')
-
-        video_url = ''.join(translation_table.get(c, c) for c in pl_fiji)
-
+        video_url = _decode(pl_fiji)
         formats = [{
             'url': video_url,
-            'format_id': pl_c_qual,
+            'format_id': 'default-%s' % pl_c_qual,
         }]
 
+        qualities_json = self._search_regex(
+            r'var pl_qualities\s*=\s*(.*?);\n', webpage, 'quality info')
+        qualities_data = json.loads(qualities_json)
+
+        for i, t in enumerate(
+                re.findall(r"pl_fiji_([a-z0-9]+)\s*=\s*'([^']+')", webpage)):
+            quality_id, crypted_url = t
+            video_url = _decode(crypted_url)
+            f = {
+                'format_id': quality_id,
+                'url': video_url,
+                'quality': i,
+            }
+            if quality_id in qualities_data:
+                qd = qualities_data[quality_id]
+                m = re.match(
+                    r'''(?x)<b>(?P<width>[0-9]+)x(?P<height>[0-9]+)<\\/b>
+                        \s*\(\s*(?P<tbr>[0-9]+)\s*kb\\/s''', qd)
+                if m:
+                    f['width'] = int(m.group('width'))
+                    f['height'] = int(m.group('height'))
+                    f['tbr'] = int(m.group('tbr'))
+            formats.append(f)
+        self._sort_formats(formats)
+
+        thumbnail = self._search_regex(
+            r"var\s+mov_thumb\s*=\s*'([^']+)';",
+            webpage, 'thumbnail', fatal=False)
+        duration = int_or_none(self._search_regex(
+            r'pl_dur\s*=\s*([0-9]+)', webpage, 'duration', fatal=False))
+
         return {
             'id': video_id,
             'title': video_title,
             'formats': formats,
+            'duration': duration,
+            'age_limit': self._rta_search(webpage),
+            'thumbnail': thumbnail,
         }

youtube_dl/extractor/dbtv.py

@@ -0,0 +1,74 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+import re
+
+from .common import InfoExtractor
+from ..utils import (
+    float_or_none,
+    int_or_none,
+    clean_html,
+)
+
+
+class DBTVIE(InfoExtractor):
+    _VALID_URL = r'http://dbtv\.no/(?P<id>[0-9]+)#(?P<display_id>.+)'
+    _TEST = {
+        'url': 'http://dbtv.no/3649835190001#Skulle_teste_ut_fornøyelsespark,_men_kollegaen_var_bare_opptatt_av_bikinikroppen',
+        'md5': 'b89953ed25dacb6edb3ef6c6f430f8bc',
+        'info_dict': {
+            'id': '33100',
+            'display_id': 'Skulle_teste_ut_fornøyelsespark,_men_kollegaen_var_bare_opptatt_av_bikinikroppen',
+            'ext': 'mp4',
+            'title': 'Skulle teste ut fornøyelsespark, men kollegaen var bare opptatt av bikinikroppen',
+            'description': 'md5:1504a54606c4dde3e4e61fc97aa857e0',
+            'thumbnail': 're:https?://.*\.jpg$',
+            'timestamp': 1404039863.438,
+            'upload_date': '20140629',
+            'duration': 69.544,
+            'view_count': int,
+            'categories': list,
+        }
+    }
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        video_id = mobj.group('id')
+        display_id = mobj.group('display_id')
+
+        data = self._download_json(
+            'http://api.dbtv.no/discovery/%s' % video_id, display_id)
+
+        video = data['playlist'][0]
+
+        formats = [{
+            'url': f['URL'],
+            'vcodec': f.get('container'),
+            'width': int_or_none(f.get('width')),
+            'height': int_or_none(f.get('height')),
+            'vbr': float_or_none(f.get('rate'), 1000),
+            'filesize': int_or_none(f.get('size')),
+        } for f in video['renditions'] if 'URL' in f]
+
+        if not formats:
+            for url_key, format_id in [('URL', 'mp4'), ('HLSURL', 'hls')]:
+                if url_key in video:
+                    formats.append({
+                        'url': video[url_key],
+                        'format_id': format_id,
+                    })
+
+        self._sort_formats(formats)
+
+        return {
+            'id': video['id'],
+            'display_id': display_id,
+            'title': video['title'],
+            'description': clean_html(video['desc']),
+            'thumbnail': video.get('splash') or video.get('thumb'),
+            'timestamp': float_or_none(video.get('publishedAt'), 1000),
+            'duration': float_or_none(video.get('length'), 1000),
+            'view_count': int_or_none(video.get('views')),
+            'categories': video.get('tags'),
+            'formats': formats,
+        }

youtube_dl/extractor/nba.py

@@ -3,18 +3,23 @@ from __future__ import unicode_literals
 import re
 
 from .common import InfoExtractor
+from ..utils import (
+    remove_end,
+    parse_duration,
+)
 
 
 class NBAIE(InfoExtractor):
     _VALID_URL = r'https?://(?:watch\.|www\.)?nba\.com/(?:nba/)?video(?P<id>/[^?]*?)(?:/index\.html)?(?:\?.*)?$'
     _TEST = {
         'url': 'http://www.nba.com/video/games/nets/2012/12/04/0021200253-okc-bkn-recap.nba/index.html',
-        'md5': u'c0edcfc37607344e2ff8f13c378c88a4',
+        'md5': 'c0edcfc37607344e2ff8f13c378c88a4',
         'info_dict': {
             'id': '0021200253-okc-bkn-recap.nba',
             'ext': 'mp4',
-            'description': 'Kevin Durant scores 32 points and dishes out six assists as the Thunder beat the Nets in Brooklyn.',
             'title': 'Thunder vs. Nets',
+            'description': 'Kevin Durant scores 32 points and dishes out six assists as the Thunder beat the Nets in Brooklyn.',
+            'duration': 181,
         },
     }
 
@@ -27,13 +32,18 @@ class NBAIE(InfoExtractor):
         video_url = 'http://ht-mobile.cdn.turner.com/nba/big' + video_id + '_nba_1280x720.mp4'
 
         shortened_video_id = video_id.rpartition('/')[2]
-        title = self._og_search_title(webpage, default=shortened_video_id).replace('NBA.com: ', '')
+        title = remove_end(
+            self._og_search_title(webpage, default=shortened_video_id), ' : NBA.com')
+
+        description = self._og_search_description(webpage)
+        duration = parse_duration(
+            self._html_search_meta('duration', webpage, 'duration', fatal=False))
 
-        description = self._html_search_regex(r'<meta name="description" (?:content|value)="(.*?)" />', webpage, 'description', fatal=False)
 
         return {
             'id': shortened_video_id,
             'url': video_url,
             'title': title,
             'description': description,
+            'duration': duration,
         }

youtube_dl/extractor/techtalks.py

@@ -1,3 +1,5 @@
+from __future__ import unicode_literals
+
 import re
 
 from .common import InfoExtractor
@@ -11,24 +13,30 @@ class TechTalksIE(InfoExtractor):
     _VALID_URL = r'https?://techtalks\.tv/talks/[^/]*/(?P<id>\d+)/'
 
     _TEST = {
-        u'url': u'http://techtalks.tv/talks/learning-topic-models-going-beyond-svd/57758/',
-        u'playlist': [
+        'url': 'http://techtalks.tv/talks/learning-topic-models-going-beyond-svd/57758/',
+        'info_dict': {
+            'id': '57758',
+            'title': 'Learning Topic Models --- Going beyond SVD',
+        },
+        'playlist': [
             {
-                u'file': u'57758.flv',
-                u'info_dict': {
-                    u'title': u'Learning Topic Models --- Going beyond SVD',
+                'info_dict': {
+                    'id': '57758',
+                    'ext': 'flv',
+                    'title': 'Learning Topic Models --- Going beyond SVD',
                 },
             },
             {
-                u'file': u'57758-slides.flv',
-                u'info_dict': {
-                    u'title': u'Learning Topic Models --- Going beyond SVD',
+                'info_dict': {
+                    'id': '57758-slides',
+                    'ext': 'flv',
+                    'title': 'Learning Topic Models --- Going beyond SVD',
                 },
             },
         ],
-        u'params': {
+        'params': {
             # rtmp download
-            u'skip_download': True,
+            'skip_download': True,
         },
     }
 
@@ -36,30 +44,36 @@ class TechTalksIE(InfoExtractor):
         mobj = re.match(self._VALID_URL, url)
         talk_id = mobj.group('id')
         webpage = self._download_webpage(url, talk_id)
-        rtmp_url = self._search_regex(r'netConnectionUrl: \'(.*?)\'', webpage,
-            u'rtmp url')
-        play_path = self._search_regex(r'href=\'(.*?)\' [^>]*id="flowplayer_presenter"',
-            webpage, u'presenter play path')
+        rtmp_url = self._search_regex(
+            r'netConnectionUrl: \'(.*?)\'', webpage, 'rtmp url')
+        play_path = self._search_regex(
+            r'href=\'(.*?)\' [^>]*id="flowplayer_presenter"',
+            webpage, 'presenter play path')
         title = clean_html(get_element_by_attribute('class', 'title', webpage))
         video_info = {
-                'id': talk_id,
-                'title': title,
-                'url': rtmp_url,
-                'play_path': play_path,
-                'ext': 'flv',
-            }
+            'id': talk_id,
+            'title': title,
+            'url': rtmp_url,
+            'play_path': play_path,
+            'ext': 'flv',
+        }
         m_slides = re.search(r'<a class="slides" href=\'(.*?)\'', webpage)
         if m_slides is None:
             return video_info
         else:
-            return [
-                video_info,
-                # The slides video
-                {
-                    'id': talk_id + '-slides',
-                    'title': title,
-                    'url': rtmp_url,
-                    'play_path': m_slides.group(1),
-                    'ext': 'flv',
-                },
-            ]
+            return {
+                '_type': 'playlist',
+                'id': talk_id,
+                'title': title,
+                'entries': [
+                    video_info,
+                    # The slides video
+                    {
+                        'id': talk_id + '-slides',
+                        'title': title,
+                        'url': rtmp_url,
+                        'play_path': m_slides.group(1),
+                        'ext': 'flv',
+                    },
+                ],
+            }

youtube_dl/extractor/toypics.py

@@ -47,7 +47,7 @@ class ToypicsUserIE(InfoExtractor):
         'info_dict': {
             'id': 'Mikey',
         },
-        'playlist_mincount': 9917,
+        'playlist_mincount': 19,
     }
 
     def _real_extract(self, url):

youtube_dl/extractor/unistra.py

@@ -1,32 +1,66 @@
+from __future__ import unicode_literals
+
 import re
 
 from .common import InfoExtractor
+from ..utils import qualities
+
 
 class UnistraIE(InfoExtractor):
-    _VALID_URL = r'http://utv\.unistra\.fr/(?:index|video)\.php\?id_video\=(\d+)'
+    _VALID_URL = r'http://utv\.unistra\.fr/(?:index|video)\.php\?id_video\=(?P<id>\d+)'
 
-    _TEST = {
-        u'url': u'http://utv.unistra.fr/video.php?id_video=154',
-        u'file': u'154.mp4',
-        u'md5': u'736f605cfdc96724d55bb543ab3ced24',
-        u'info_dict': {
-            u'title': u'M!ss Yella',
-            u'description': u'md5:104892c71bd48e55d70b902736b81bbf',
+    _TESTS = [
+        {
+            'url': 'http://utv.unistra.fr/video.php?id_video=154',
+            'md5': '736f605cfdc96724d55bb543ab3ced24',
+            'info_dict': {
+                'id': '154',
+                'ext': 'mp4',
+                'title': 'M!ss Yella',
+                'description': 'md5:104892c71bd48e55d70b902736b81bbf',
+            },
         },
-    }
+        {
+            'url': 'http://utv.unistra.fr/index.php?id_video=437',
+            'md5': '1ddddd6cccaae76f622ce29b8779636d',
+            'info_dict': {
+                'id': '437',
+                'ext': 'mp4',
+                'title': 'Prix Louise Weiss 2014',
+                'description': 'md5:cc3a8735f079f4fb6b0b570fc10c135a',
+            },
+        }
+    ]
 
     def _real_extract(self, url):
-        id = re.match(self._VALID_URL, url).group(1)
-        webpage = self._download_webpage(url, id)
-        file = re.search(r'file: "(.*?)",', webpage).group(1)
-        title = self._html_search_regex(r'<title>UTV - (.*?)</', webpage, u'title')
+        mobj = re.match(self._VALID_URL, url)
+        video_id = mobj.group('id')
 
-        video_url = 'http://vod-flash.u-strasbg.fr:8080/' + file
+        webpage = self._download_webpage(url, video_id)
 
-        return {'id': id,
-                'title': title,
-                'ext': 'mp4',
-                'url': video_url,
-                'description': self._html_search_regex(r'<meta name="Description" content="(.*?)"', webpage, u'description', flags=re.DOTALL),
-                'thumbnail': self._search_regex(r'image: "(.*?)"', webpage, u'thumbnail'),
-                }
+        files = set(re.findall(r'file\s*:\s*"([^"]+)"', webpage))
+
+        quality = qualities(['SD', 'HD'])
+        formats = []
+        for file_path in files:
+            format_id = 'HD' if file_path.endswith('-HD.mp4') else 'SD'
+            formats.append({
+                'url': 'http://vod-flash.u-strasbg.fr:8080%s' % file_path,
+                'format_id': format_id,
+                'quality': quality(format_id)
+            })
+
+        title = self._html_search_regex(
+            r'<title>UTV - (.*?)</', webpage, 'title')
+        description = self._html_search_regex(
+            r'<meta name="Description" content="(.*?)"', webpage, 'description', flags=re.DOTALL)
+        thumbnail = self._search_regex(
+            r'image: "(.*?)"', webpage, 'thumbnail')
+
+        return {
+            'id': video_id,
+            'title': title,
+            'description': description,
+            'thumbnail': thumbnail,
+            'formats': formats
+        }

youtube_dl/extractor/washingtonpost.py

@@ -13,6 +13,9 @@ class WashingtonPostIE(InfoExtractor):
     _VALID_URL = r'^https?://(?:www\.)?washingtonpost\.com/.*?/(?P<id>[^/]+)/(?:$|[?#])'
     _TEST = {
         'url': 'http://www.washingtonpost.com/sf/national/2014/03/22/sinkhole-of-bureaucracy/',
+        'info_dict': {
+            'title': 'Sinkhole of bureaucracy',
+        },
         'playlist': [{
             'md5': 'c3f4b4922ffa259243f68e928db2db8c',
             'info_dict': {

youtube_dl/version.py

@@ -1,2 +1,2 @@
 
-__version__ = '2014.09.04.1'
+__version__ = '2014.09.06'