Merge remote-tracking branch 'upstream/master'

test/test_InfoExtractor.py

@@ -35,10 +35,14 @@ class TestInfoExtractor(unittest.TestCase):
             <meta name="og:title" content='Foo'/>
             <meta content="Some video's description " name="og:description"/>
             <meta property='og:image' content='http://domain.com/pic.jpg?key1=val1&amp;key2=val2'/>
+            <meta content='application/x-shockwave-flash' property='og:video:type'>
+            <meta content='Foo' property=og:foobar>
             '''
         self.assertEqual(ie._og_search_title(html), 'Foo')
         self.assertEqual(ie._og_search_description(html), 'Some video\'s description ')
         self.assertEqual(ie._og_search_thumbnail(html), 'http://domain.com/pic.jpg?key1=val1&key2=val2')
+        self.assertEqual(ie._og_search_video_url(html, default=None), None)
+        self.assertEqual(ie._og_search_property('foobar', html), 'Foo')
 
     def test_html_search_meta(self):
         ie = self.ie

youtube_dl/extractor/__init__.py

@@ -319,7 +319,6 @@ from .macgamestore import MacGameStoreIE
 from .mailru import MailRuIE
 from .malemotion import MalemotionIE
 from .mdr import MDRIE
-from .megavideoz import MegaVideozIE
 from .metacafe import MetacafeIE
 from .metacritic import MetacriticIE
 from .mgoon import MgoonIE

youtube_dl/extractor/channel9.py

@@ -3,7 +3,11 @@ from __future__ import unicode_literals
 import re
 
 from .common import InfoExtractor
-from ..utils import ExtractorError
+from ..utils import (
+    ExtractorError,
+    parse_filesize,
+    qualities,
+)
 
 
 class Channel9IE(InfoExtractor):
@@ -28,7 +32,7 @@ class Channel9IE(InfoExtractor):
                 'title': 'Developer Kick-Off Session: Stuff We Love',
                 'description': 'md5:c08d72240b7c87fcecafe2692f80e35f',
                 'duration': 4576,
-                'thumbnail': 'http://video.ch9.ms/ch9/9d51/03902f2d-fc97-4d3c-b195-0bfe15a19d51/KOS002_220.jpg',
+                'thumbnail': 're:http://.*\.jpg',
                 'session_code': 'KOS002',
                 'session_day': 'Day 1',
                 'session_room': 'Arena 1A',
@@ -44,31 +48,29 @@ class Channel9IE(InfoExtractor):
                 'title': 'Self-service BI with Power BI - nuclear testing',
                 'description': 'md5:d1e6ecaafa7fb52a2cacdf9599829f5b',
                 'duration': 1540,
-                'thumbnail': 'http://video.ch9.ms/ch9/87e1/0300391f-a455-4c72-bec3-4422f19287e1/selfservicenuk_512.jpg',
+                'thumbnail': 're:http://.*\.jpg',
                 'authors': ['Mike Wilmot'],
             },
+        },
+        {
+            # low quality mp4 is best
+            'url': 'https://channel9.msdn.com/Events/CPP/CppCon-2015/Ranges-for-the-Standard-Library',
+            'info_dict': {
+                'id': 'Events/CPP/CppCon-2015/Ranges-for-the-Standard-Library',
+                'ext': 'mp4',
+                'title': 'Ranges for the Standard Library',
+                'description': 'md5:2e6b4917677af3728c5f6d63784c4c5d',
+                'duration': 5646,
+                'thumbnail': 're:http://.*\.jpg',
+            },
+            'params': {
+                'skip_download': True,
+            },
         }
     ]
 
     _RSS_URL = 'http://channel9.msdn.com/%s/RSS'
 
-    # Sorted by quality
-    _known_formats = ['MP3', 'MP4', 'Mid Quality WMV', 'Mid Quality MP4', 'High Quality WMV', 'High Quality MP4']
-
-    def _restore_bytes(self, formatted_size):
-        if not formatted_size:
-            return 0
-        m = re.match(r'^(?P<size>\d+(?:\.\d+)?)\s+(?P<units>[a-zA-Z]+)', formatted_size)
-        if not m:
-            return 0
-        units = m.group('units')
-        try:
-            exponent = ['B', 'KB', 'MB', 'GB', 'TB', 'PB', 'EB', 'ZB', 'YB'].index(units.upper())
-        except ValueError:
-            return 0
-        size = float(m.group('size'))
-        return int(size * (1024 ** exponent))
-
     def _formats_from_html(self, html):
         FORMAT_REGEX = r'''
             (?x)
@@ -78,16 +80,20 @@ class Channel9IE(InfoExtractor):
             <h3>File\s+size</h3>\s*(?P<filesize>.*?)\s*
             </div>)?                                                # File size part may be missing
         '''
-        # Extract known formats
+        quality = qualities((
+            'MP3', 'MP4',
+            'Low Quality WMV', 'Low Quality MP4',
+            'Mid Quality WMV', 'Mid Quality MP4',
+            'High Quality WMV', 'High Quality MP4'))
         formats = [{
             'url': x.group('url'),
             'format_id': x.group('quality'),
             'format_note': x.group('note'),
             'format': '%s (%s)' % (x.group('quality'), x.group('note')),
-            'filesize': self._restore_bytes(x.group('filesize')),  # File size is approximate
+            'filesize_approx': parse_filesize(x.group('filesize')),
-            'preference': self._known_formats.index(x.group('quality')),
+            'quality': quality(x.group('quality')),
             'vcodec': 'none' if x.group('note') == 'Audio only' else None,
-        } for x in list(re.finditer(FORMAT_REGEX, html)) if x.group('quality') in self._known_formats]
+        } for x in list(re.finditer(FORMAT_REGEX, html))]
 
         self._sort_formats(formats)
 

youtube_dl/extractor/common.py

@@ -646,7 +646,8 @@ class InfoExtractor(object):
     @staticmethod
     def _og_regexes(prop):
         content_re = r'content=(?:"([^>]+?)"|\'([^>]+?)\'|\s*([^\s"\'=<>`]+?))'
-        property_re = r'(?:name|property)=[\'"]?og:%s[\'"]?' % re.escape(prop)
+        property_re = (r'(?:name|property)=(?:\'og:%(prop)s\'|"og:%(prop)s"|\s*og:%(prop)s\b)'
+                       % {'prop': re.escape(prop)})
         template = r'<meta[^>]+?%s[^>]+?%s'
         return [
             template % (property_re, content_re),

youtube_dl/extractor/criterion.py

@@ -27,9 +27,7 @@ class CriterionIE(InfoExtractor):
         final_url = self._search_regex(
             r'so.addVariable\("videoURL", "(.+?)"\)\;', webpage, 'video url')
         title = self._og_search_title(webpage)
-        description = self._html_search_regex(
+        description = self._html_search_meta('description', webpage)
-            r'<meta name="description" content="(.+?)" />',
-            webpage, 'video description')
         thumbnail = self._search_regex(
             r'so.addVariable\("thumbnailURL", "(.+?)"\)\;',
             webpage, 'thumbnail url')

youtube_dl/extractor/fivemin.py

@@ -2,11 +2,15 @@ from __future__ import unicode_literals
 
 from .common import InfoExtractor
 from ..compat import (
-    compat_str,
     compat_urllib_parse,
+    compat_parse_qs,
+    compat_urllib_parse_urlparse,
+    compat_urlparse,
 )
 from ..utils import (
     ExtractorError,
+    parse_duration,
+    replace_extension,
 )
 
 
@@ -28,6 +32,7 @@ class FiveMinIE(InfoExtractor):
                 'id': '518013791',
                 'ext': 'mp4',
                 'title': 'iPad Mini with Retina Display Review',
+                'duration': 177,
             },
         },
         {
@@ -38,9 +43,52 @@ class FiveMinIE(InfoExtractor):
                 'id': '518086247',
                 'ext': 'mp4',
                 'title': 'How to Make a Next-Level Fruit Salad',
+                'duration': 184,
             },
         },
     ]
+    _ERRORS = {
+        'ErrorVideoNotExist': 'We\'re sorry, but the video you are trying to watch does not exist.',
+        'ErrorVideoNoLongerAvailable': 'We\'re sorry, but the video you are trying to watch is no longer available.',
+        'ErrorVideoRejected': 'We\'re sorry, but the video you are trying to watch has been removed.',
+        'ErrorVideoUserNotGeo': 'We\'re sorry, but the video you are trying to watch cannot be viewed from your current location.',
+        'ErrorVideoLibraryRestriction': 'We\'re sorry, but the video you are trying to watch is currently unavailable for viewing at this domain.',
+        'ErrorExposurePermission': 'We\'re sorry, but the video you are trying to watch is currently unavailable for viewing at this domain.',
+    }
+    _QUALITIES = {
+        1: {
+            'width': 640,
+            'height': 360,
+        },
+        2: {
+            'width': 854,
+            'height': 480,
+        },
+        4: {
+            'width': 1280,
+            'height': 720,
+        },
+        8: {
+            'width': 1920,
+            'height': 1080,
+        },
+        16: {
+            'width': 640,
+            'height': 360,
+        },
+        32: {
+            'width': 854,
+            'height': 480,
+        },
+        64: {
+            'width': 1280,
+            'height': 720,
+        },
+        128: {
+            'width': 640,
+            'height': 360,
+        },
+    }
 
     def _real_extract(self, url):
         video_id = self._match_id(url)
@@ -59,26 +107,36 @@ class FiveMinIE(InfoExtractor):
             'https://syn.5min.com/handlers/SenseHandler.ashx?' + query,
             video_id)
         if not response['success']:
-            err_msg = response['errorMessage']
+            raise ExtractorError(
-            if err_msg == 'ErrorVideoUserNotGeo':
+                '%s said: %s' % (
-                msg = 'Video not available from your location'
+                    self.IE_NAME,
-            else:
+                    self._ERRORS.get(response['errorMessage'], response['errorMessage'])),
-                msg = 'Aol said: %s' % err_msg
+                expected=True)
-            raise ExtractorError(msg, expected=True, video_id=video_id)
         info = response['binding'][0]
 
-        second_id = compat_str(int(video_id[:-2]) + 1)
         formats = []
-        for quality, height in [(1, 320), (2, 480), (4, 720), (8, 1080)]:
+        parsed_video_url = compat_urllib_parse_urlparse(compat_parse_qs(
-            if any(r['ID'] == quality for r in info['Renditions']):
+            compat_urllib_parse_urlparse(info['EmbededURL']).query)['videoUrl'][0])
+        for rendition in info['Renditions']:
+            if rendition['RenditionType'] == 'm3u8':
+                formats.extend(self._extract_m3u8_formats(rendition['Url'], video_id, m3u8_id='hls'))
+            elif rendition['RenditionType'] == 'aac':
+                continue
+            else:
+                rendition_url = compat_urlparse.urlunparse(parsed_video_url._replace(path=replace_extension(parsed_video_url.path.replace('//', '/%s/' % rendition['ID']), rendition['RenditionType'])))
+                quality = self._QUALITIES.get(rendition['ID'], {})
                 formats.append({
-                    'format_id': compat_str(quality),
+                    'format_id': '%s-%d' % (rendition['RenditionType'], rendition['ID']),
-                    'url': 'http://avideos.5min.com/%s/%s/%s_%s.mp4' % (second_id[-3:], second_id, video_id, quality),
+                    'url': rendition_url,
-                    'height': height,
+                    'width': quality.get('width'),
+                    'height': quality.get('height'),
                 })
+        self._sort_formats(formats)
 
         return {
             'id': video_id,
             'title': info['Title'],
+            'thumbnail': info.get('ThumbURL'),
+            'duration': parse_duration(info.get('Duration')),
             'formats': formats,
         }

youtube_dl/extractor/jeuxvideo.py

@@ -28,7 +28,7 @@ class JeuxVideoIE(InfoExtractor):
         mobj = re.match(self._VALID_URL, url)
         title = mobj.group(1)
         webpage = self._download_webpage(url, title)
-        title = self._html_search_meta('name', webpage)
+        title = self._html_search_meta('name', webpage) or self._og_search_title(webpage)
         config_url = self._html_search_regex(
             r'data-src="(/contenu/medias/video.php.*?)"',
             webpage, 'config URL')

youtube_dl/extractor/megavideoz.py

@@ -1,56 +0,0 @@
-# encoding: utf-8
-from __future__ import unicode_literals
-
-import re
-
-from .common import InfoExtractor
-from ..utils import (
-    ExtractorError,
-    float_or_none,
-    xpath_text,
-)
-
-
-class MegaVideozIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?megavideoz\.eu/video/(?P<id>[^/]+)(?:/(?P<display_id>[^/]+))?'
-    _TEST = {
-        'url': 'http://megavideoz.eu/video/WM6UB919XMXH/SMPTE-Universal-Film-Leader',
-        'info_dict': {
-            'id': '48723',
-            'display_id': 'SMPTE-Universal-Film-Leader',
-            'ext': 'mp4',
-            'title': 'SMPTE Universal Film Leader',
-            'thumbnail': 're:https?://.*?\.jpg',
-            'duration': 10.93,
-        }
-    }
-
-    def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
-        video_id = mobj.group('id')
-        display_id = mobj.group('display_id') or video_id
-
-        webpage = self._download_webpage(url, display_id)
-
-        if any(p in webpage for p in ('>Video Not Found<', '>404 Error<')):
-            raise ExtractorError('Video %s does not exist' % video_id, expected=True)
-
-        config = self._download_xml(
-            self._search_regex(
-                r"var\s+cnf\s*=\s*'([^']+)'", webpage, 'cnf url'),
-            display_id)
-
-        video_url = xpath_text(config, './file', 'video url', fatal=True)
-        title = xpath_text(config, './title', 'title', fatal=True)
-        thumbnail = xpath_text(config, './image', 'thumbnail')
-        duration = float_or_none(xpath_text(config, './duration', 'duration'))
-        video_id = xpath_text(config, './mediaid', 'video id') or video_id
-
-        return {
-            'id': video_id,
-            'display_id': display_id,
-            'url': video_url,
-            'title': title,
-            'thumbnail': thumbnail,
-            'duration': duration
-        }

youtube_dl/extractor/vimeo.py

@@ -212,7 +212,7 @@ class VimeoIE(VimeoBaseInfoExtractor):
             url = url.replace('http://', 'https://')
         password_request = compat_urllib_request.Request(url + '/password', data)
         password_request.add_header('Content-Type', 'application/x-www-form-urlencoded')
-        password_request.add_header('Cookie', 'clip_v=1; vuid=%s' % vuid)
+        password_request.add_header('Cookie', 'clip_test2=1; vuid=%s' % vuid)
         password_request.add_header('Referer', url)
         return self._download_webpage(
             password_request, video_id,

youtube_dl/extractor/yandexmusic.py

@@ -46,6 +46,12 @@ class YandexMusicTrackIE(InfoExtractor):
                 % (data['host'], key, data['ts'] + data['path'], storage[1]))
 
     def _get_track_info(self, track):
+        thumbnail = None
+        cover_uri = track.get('albums', [{}])[0].get('coverUri')
+        if cover_uri:
+            thumbnail = cover_uri.replace('%%', 'orig')
+            if not thumbnail.startswith('http'):
+                thumbnail = 'http://' + thumbnail
         return {
             'id': track['id'],
             'ext': 'mp3',
@@ -53,6 +59,7 @@ class YandexMusicTrackIE(InfoExtractor):
             'title': '%s - %s' % (track['artists'][0]['name'], track['title']),
             'filesize': int_or_none(track.get('fileSize')),
             'duration': float_or_none(track.get('durationMs'), 1000),
+            'thumbnail': thumbnail,
         }
 
     def _real_extract(self, url):

youtube_dl/utils.py

@@ -1371,7 +1371,12 @@ def int_or_none(v, scale=1, default=None, get_attr=None, invscale=1):
             v = getattr(v, get_attr, None)
     if v == '':
         v = None
-    return default if v is None else (int(v) * invscale // scale)
+    if v is None:
+        return default
+    try:
+        return int(v) * invscale // scale
+    except ValueError:
+        return default
 
 
 def str_or_none(v, default=None):
@@ -1387,7 +1392,12 @@ def str_to_int(int_str):
 
 
 def float_or_none(v, scale=1, invscale=1, default=None):
-    return default if v is None else (float(v) * invscale / scale)
+    if v is None:
+        return default
+    try:
+        return float(v) * invscale / scale
+    except ValueError:
+        return default
 
 
 def parse_duration(s):