release 2014.09.04.3

Conflicts:
	youtube_dl/extractor/eporner.py
	youtube_dl/extractor/hornbunny.py

test/test_all_urls.py

@@ -143,32 +143,6 @@ class TestAllURLsMatching(unittest.TestCase):
         self.assertMatch('http://video.pbs.org/viralplayer/2365173446/', ['PBS'])
         self.assertMatch('http://video.pbs.org/widget/partnerplayer/980042464/', ['PBS'])
 
-    def test_ComedyCentralShows(self):
-        self.assertMatch(
-            'http://thedailyshow.cc.com/extended-interviews/xm3fnq/andrew-napolitano-extended-interview',
-            ['ComedyCentralShows'])
-        self.assertMatch(
-            'http://thecolbertreport.cc.com/videos/29w6fx/-realhumanpraise-for-fox-news',
-            ['ComedyCentralShows'])
-        self.assertMatch(
-            'http://thecolbertreport.cc.com/videos/gh6urb/neil-degrasse-tyson-pt--1?xrs=eml_col_031114',
-            ['ComedyCentralShows'])
-        self.assertMatch(
-            'http://thedailyshow.cc.com/guests/michael-lewis/3efna8/exclusive---michael-lewis-extended-interview-pt--3',
-            ['ComedyCentralShows'])
-        self.assertMatch(
-            'http://thedailyshow.cc.com/episodes/sy7yv0/april-8--2014---denis-leary',
-            ['ComedyCentralShows'])
-        self.assertMatch(
-            'http://thecolbertreport.cc.com/episodes/8ase07/april-8--2014---jane-goodall',
-            ['ComedyCentralShows'])
-        self.assertMatch(
-            'http://thedailyshow.cc.com/video-playlists/npde3s/the-daily-show-19088-highlights',
-            ['ComedyCentralShows'])
-        self.assertMatch(
-            'http://thedailyshow.cc.com/special-editions/2l8fdb/special-edition---a-look-back-at-food',
-            ['ComedyCentralShows'])
-
     def test_yahoo_https(self):
         # https://github.com/rg3/youtube-dl/issues/2701
         self.assertMatch(

test/test_cache.py

@@ -0,0 +1,59 @@
+#!/usr/bin/env python
+# coding: utf-8
+
+from __future__ import unicode_literals
+
+import shutil
+
+# Allow direct execution
+import os
+import sys
+import unittest
+sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
+
+
+from test.helper import FakeYDL
+from youtube_dl.cache import Cache
+
+
+def _is_empty(d):
+    return not bool(os.listdir(d))
+
+
+def _mkdir(d):
+    if not os.path.exists(d):
+        os.mkdir(d)
+
+
+class TestCache(unittest.TestCase):
+    def setUp(self):
+        TEST_DIR = os.path.dirname(os.path.abspath(__file__))
+        TESTDATA_DIR = os.path.join(TEST_DIR, 'testdata')
+        _mkdir(TESTDATA_DIR)
+        self.test_dir = os.path.join(TESTDATA_DIR, 'cache_test')
+        self.tearDown()
+
+    def tearDown(self):
+        if os.path.exists(self.test_dir):
+            shutil.rmtree(self.test_dir)
+
+    def test_cache(self):
+        ydl = FakeYDL({
+            'cachedir': self.test_dir,
+        })
+        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', 'k2'), None)
+        self.assertFalse(_is_empty(self.test_dir))
+        self.assertEqual(c.load('test_cache', 'k.'), obj)
+        self.assertEqual(c.load('test_cache', 'y'), 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)
+
+
+if __name__ == '__main__':
+    unittest.main()

test/test_download.py

@@ -167,21 +167,21 @@ def generator(test_case):
                 if not test_case.get('params', {}).get('skip_download', False):
                     self.assertTrue(os.path.exists(tc_filename), msg='Missing file ' + tc_filename)
                     self.assertTrue(tc_filename in finished_hook_called)
+                    expected_minsize = tc.get('file_minsize', 10000)
+                    if expected_minsize is not None:
+                        if params.get('test'):
+                            expected_minsize = max(expected_minsize, 10000)
+                        got_fsize = os.path.getsize(tc_filename)
+                        assertGreaterEqual(
+                            self, got_fsize, expected_minsize,
+                            'Expected %s to be at least %s, but it\'s only %s ' %
+                            (tc_filename, format_bytes(expected_minsize),
+                                format_bytes(got_fsize)))
+                    if 'md5' in tc:
+                        md5_for_file = _file_md5(tc_filename)
+                        self.assertEqual(md5_for_file, tc['md5'])
                 info_json_fn = os.path.splitext(tc_filename)[0] + '.info.json'
                 self.assertTrue(os.path.exists(info_json_fn))
-                if 'md5' in tc:
-                    md5_for_file = _file_md5(tc_filename)
-                    self.assertEqual(md5_for_file, tc['md5'])
-                expected_minsize = tc.get('file_minsize', 10000)
-                if expected_minsize is not None:
-                    if params.get('test'):
-                        expected_minsize = max(expected_minsize, 10000)
-                    got_fsize = os.path.getsize(tc_filename)
-                    assertGreaterEqual(
-                        self, got_fsize, expected_minsize,
-                        'Expected %s to be at least %s, but it\'s only %s ' %
-                        (tc_filename, format_bytes(expected_minsize),
-                            format_bytes(got_fsize)))
                 with io.open(info_json_fn, encoding='utf-8') as infof:
                     info_dict = json.load(infof)
 

test/test_utils.py

@@ -211,6 +211,9 @@ class TestUtil(unittest.TestCase):
         self.assertEqual(parse_duration('00:01:01'), 61)
         self.assertEqual(parse_duration('x:y'), None)
         self.assertEqual(parse_duration('3h11m53s'), 11513)
+        self.assertEqual(parse_duration('3h 11m 53s'), 11513)
+        self.assertEqual(parse_duration('3 hours 11 minutes 53 seconds'), 11513)
+        self.assertEqual(parse_duration('3 hours 11 mins 53 secs'), 11513)
         self.assertEqual(parse_duration('62m45s'), 3765)
         self.assertEqual(parse_duration('6m59s'), 419)
         self.assertEqual(parse_duration('49s'), 49)

youtube_dl/YoutubeDL.py

@@ -57,6 +57,7 @@ from .utils import (
     YoutubeDLHandler,
     prepend_extension,
 )
+from .cache import Cache
 from .extractor import get_info_extractor, gen_extractors
 from .downloader import get_suitable_downloader
 from .postprocessor import FFmpegMergerPP
@@ -133,7 +134,7 @@ class YoutubeDL(object):
     daterange:         A DateRange object, download only if the upload_date is in the range.
     skip_download:     Skip the actual download of the video file
     cachedir:          Location of the cache files in the filesystem.
-                       None to disable filesystem cache.
+                       False to disable filesystem cache.
     noplaylist:        Download single video instead of a playlist if in doubt.
     age_limit:         An integer representing the user's age in years.
                        Unsuitable videos for the given age are skipped.
@@ -195,6 +196,7 @@ class YoutubeDL(object):
         self._screen_file = [sys.stdout, sys.stderr][params.get('logtostderr', False)]
         self._err_file = sys.stderr
         self.params = params
+        self.cache = Cache(self)
 
         if params.get('bidi_workaround', False):
             try:

youtube_dl/__init__.py

@@ -84,7 +84,6 @@ import optparse
 import os
 import random
 import shlex
-import shutil
 import sys
 
 
@@ -96,7 +95,6 @@ from .utils import (
     decodeOption,
     get_term_width,
     DownloadError,
-    get_cachedir,
     MaxDownloadsReached,
     preferredencoding,
     read_batch_urls,
@@ -518,10 +516,10 @@ def parseOpts(overrideArguments=None):
     filesystem.add_option('--cookies',
             dest='cookiefile', metavar='FILE', help='file to read cookies from and dump cookie jar in')
     filesystem.add_option(
-        '--cache-dir', dest='cachedir', default=get_cachedir(), metavar='DIR',
+        '--cache-dir', dest='cachedir', default=None, metavar='DIR',
         help='Location in the filesystem where youtube-dl can store some downloaded information permanently. By default $XDG_CACHE_HOME/youtube-dl or ~/.cache/youtube-dl . At the moment, only YouTube player files (for videos with obfuscated signatures) are cached, but that may change.')
     filesystem.add_option(
-        '--no-cache-dir', action='store_const', const=None, dest='cachedir',
+        '--no-cache-dir', action='store_const', const=False, dest='cachedir',
         help='Disable filesystem caching')
     filesystem.add_option(
         '--rm-cache-dir', action='store_true', dest='rm_cachedir',
@@ -872,20 +870,7 @@ def _real_main(argv=None):
 
         # Remove cache dir
         if opts.rm_cachedir:
-            if opts.cachedir is None:
-                ydl.to_screen(u'No cache dir specified (Did you combine --no-cache-dir and --rm-cache-dir?)')
-            else:
-                if ('.cache' not in opts.cachedir) or ('youtube-dl' not in opts.cachedir):
-                    ydl.to_screen(u'Not removing directory %s - this does not look like a cache dir')
-                    retcode = 141
-                else:
-                    ydl.to_screen(
-                        u'Removing cache dir %s .' % opts.cachedir,
-                        skip_eol=True)
-                    if os.path.exists(opts.cachedir):
-                        ydl.to_screen(u'.', skip_eol=True)
-                        shutil.rmtree(opts.cachedir)
-                    ydl.to_screen(u'.')
+            ydl.cache.remove()
 
         # Maybe do nothing
         if (len(all_urls) < 1) and (opts.load_info_filename is None):

youtube_dl/cache.py

@@ -0,0 +1,94 @@
+from __future__ import unicode_literals
+
+import errno
+import io
+import json
+import os
+import re
+import shutil
+import traceback
+
+from .utils import (
+    write_json_file,
+)
+
+
+class Cache(object):
+    def __init__(self, ydl):
+        self._ydl = ydl
+
+    def _get_root_dir(self):
+        res = self._ydl.params.get('cachedir')
+        if res is None:
+            cache_root = os.environ.get('XDG_CACHE_HOME', '~/.cache')
+            res = os.path.join(cache_root, 'youtube-dl')
+        return os.path.expanduser(res)
+
+    def _get_cache_fn(self, section, key, dtype):
+        assert re.match(r'^[a-zA-Z0-9_.-]+$', section), \
+            'invalid section %r' % section
+        assert re.match(r'^[a-zA-Z0-9_.-]+$', key), 'invalid key %r' % key
+        return os.path.join(
+            self._get_root_dir(), section, '%s.%s' % (key, dtype))
+
+    @property
+    def enabled(self):
+        return self._ydl.params.get('cachedir') is not False
+
+    def store(self, section, key, data, dtype='json'):
+        assert dtype in ('json',)
+
+        if not self.enabled:
+            return
+
+        fn = self._get_cache_fn(section, key, dtype)
+        try:
+            try:
+                os.makedirs(os.path.dirname(fn))
+            except OSError as ose:
+                if ose.errno != errno.EEXIST:
+                    raise
+            write_json_file(data, fn)
+        except Exception:
+            tb = traceback.format_exc()
+            self._ydl.report_warning(
+                'Writing cache to %r failed: %s' % (fn, tb))
+
+    def load(self, section, key, dtype='json', default=None):
+        assert dtype in ('json',)
+
+        if not self.enabled:
+            return default
+
+        cache_fn = self._get_cache_fn(section, key, dtype)
+        try:
+            try:
+                with io.open(cache_fn, 'r', encoding='utf-8') as cachef:
+                    return json.load(cachef)
+            except ValueError:
+                try:
+                    file_size = os.path.getsize(cache_fn)
+                except (OSError, IOError) as oe:
+                    file_size = str(oe)
+                self._ydl.report_warning(
+                    'Cache retrieval from %s failed (%s)' % (cache_fn, file_size))
+        except IOError:
+            pass  # No cache available
+
+        return default
+
+    def remove(self):
+        if not self.enabled:
+            self._ydl.to_screen('Cache is disabled (Did you combine --no-cache-dir and --rm-cache-dir?)')
+            return
+
+        cachedir = self._get_root_dir()
+        if not any((term in cachedir) for term in ('cache', 'tmp')):
+            raise Exception('Not removing directory %s - this does not look like a cache dir' % cachedir)
+
+        self._ydl.to_screen(
+            'Removing cache dir %s .' % cachedir, skip_eol=True)
+        if os.path.exists(cachedir):
+            self._ydl.to_screen('.', skip_eol=True)
+            shutil.rmtree(cachedir)
+        self._ydl.to_screen('.')

youtube_dl/extractor/__init__.py

@@ -4,6 +4,7 @@ from .addanime import AddAnimeIE
 from .adultswim import AdultSwimIE
 from .aftonbladet import AftonbladetIE
 from .anitube import AnitubeIE
+from .anysex import AnySexIE
 from .aol import AolIE
 from .allocine import AllocineIE
 from .aparat import AparatIE
@@ -23,6 +24,7 @@ from .auengine import AUEngineIE
 from .bambuser import BambuserIE, BambuserChannelIE
 from .bandcamp import BandcampIE, BandcampAlbumIE
 from .bbccouk import BBCCoUkIE
+from .beeg import BeegIE
 from .bilibili import BiliBiliIE
 from .blinkx import BlinkxIE
 from .bliptv import BlipTVIE, BlipTVUserIE
@@ -68,6 +70,7 @@ from .daum import DaumIE
 from .dfb import DFBIE
 from .dotsub import DotsubIE
 from .dreisat import DreiSatIE
+from .drtuber import DrTuberIE
 from .drtv import DRTVIE
 from .dump import DumpIE
 from .defense import DefenseGouvFrIE
@@ -83,8 +86,9 @@ from .ellentv import (
     EllenTVClipsIE,
 )
 from .elpais import ElPaisIE
-from .empflix import EmpflixIE
+from .empflix import EMPFlixIE
 from .engadget import EngadgetIE
+from .eporner import EpornerIE
 from .escapist import EscapistIE
 from .everyonesmixtape import EveryonesMixtapeIE
 from .exfm import ExfmIE
@@ -134,6 +138,7 @@ from .grooveshark import GroovesharkIE
 from .hark import HarkIE
 from .helsinki import HelsinkiIE
 from .hentaistigma import HentaiStigmaIE
+from .hornbunny import HornBunnyIE
 from .hotnewhiphop import HotNewHipHopIE
 from .howcast import HowcastIE
 from .howstuffworks import HowStuffWorksIE
@@ -230,6 +235,7 @@ from .niconico import NiconicoIE
 from .ninegag import NineGagIE
 from .noco import NocoIE
 from .normalboots import NormalbootsIE
+from .nosvideo import NosVideoIE
 from .novamov import NovaMovIE
 from .nowness import NownessIE
 from .nowvideo import NowVideoIE
@@ -257,6 +263,7 @@ from .podomatic import PodomaticIE
 from .pornhd import PornHdIE
 from .pornhub import PornHubIE
 from .pornotube import PornotubeIE
+from .promptfile import PromptFileIE
 from .prosiebensat1 import ProSiebenSat1IE
 from .pyvideo import PyvideoIE
 from .radiofrance import RadioFranceIE
@@ -321,6 +328,7 @@ from .stanfordoc import StanfordOpenClassroomIE
 from .steam import SteamIE
 from .streamcloud import StreamcloudIE
 from .streamcz import StreamCZIE
+from .sunporno import SunPornoIE
 from .swrmediathek import SWRMediathekIE
 from .syfy import SyfyIE
 from .sztvhu import SztvHuIE
@@ -340,6 +348,7 @@ from .theplatform import ThePlatformIE
 from .thisav import ThisAVIE
 from .tinypic import TinyPicIE
 from .tlc import TlcIE, TlcDeIE
+from .tnaflix import TNAFlixIE
 from .toutv import TouTvIE
 from .toypics import ToypicsUserIE, ToypicsIE
 from .traileraddict import TrailerAddictIE
@@ -392,6 +401,7 @@ from .vine import (
 from .viki import VikiIE
 from .vk import VKIE
 from .vodlocker import VodlockerIE
+from .vporn import VpornIE
 from .vube import VubeIE
 from .vuclip import VuClipIE
 from .vulture import VultureIE

youtube_dl/extractor/anysex.py

@@ -0,0 +1,61 @@
+from __future__ import unicode_literals
+
+import re
+
+from .common import InfoExtractor
+from ..utils import (
+    parse_duration,
+    int_or_none,
+)
+
+
+class AnySexIE(InfoExtractor):
+    _VALID_URL = r'https?://(?:www\.)?anysex\.com/(?P<id>\d+)'
+    _TEST = {
+        'url': 'http://anysex.com/156592/',
+        'md5': '023e9fbb7f7987f5529a394c34ad3d3d',
+        'info_dict': {
+            'id': '156592',
+            'ext': 'mp4',
+            'title': 'Busty and sexy blondie in her bikini strips for you',
+            'description': 'md5:de9e418178e2931c10b62966474e1383',
+            'categories': ['Erotic'],
+            'duration': 270,
+            'age_limit': 18,
+        }
+    }
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        video_id = mobj.group('id')
+
+        webpage = self._download_webpage(url, video_id)
+
+        video_url = self._html_search_regex(r"video_url\s*:\s*'([^']+)'", webpage, 'video URL')
+
+        title = self._html_search_regex(r'<title>(.*?)</title>', webpage, 'title')
+        description = self._html_search_regex(
+            r'<div class="description">([^<]+)</div>', webpage, 'description', fatal=False)
+        thumbnail = self._html_search_regex(
+            r'preview_url\s*:\s*\'(.*?)\'', webpage, 'thumbnail', fatal=False)
+
+        categories = re.findall(
+            r'<a href="http://anysex\.com/categories/[^"]+" title="[^"]*">([^<]+)</a>', webpage)
+
+        duration = parse_duration(self._search_regex(
+            r'<b>Duration:</b> (\d+:\d+)', webpage, 'duration', fatal=False))
+        view_count = int_or_none(self._html_search_regex(
+            r'<b>Views:</b> (\d+)', webpage, 'view count', fatal=False))
+
+        return {
+            'id': video_id,
+            'url': video_url,
+            'ext': 'mp4',
+            'title': title,
+            'description': description,
+            'thumbnail': thumbnail,
+            'categories': categories,
+            'duration': duration,
+            'view_count': view_count,
+            'age_limit': 18,
+        }

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/beeg.py

@@ -0,0 +1,65 @@
+from __future__ import unicode_literals
+
+import re
+
+from .common import InfoExtractor
+
+
+class BeegIE(InfoExtractor):
+    _VALID_URL = r'https?://(?:www\.)?beeg\.com/(?P<id>\d+)'
+    _TEST = {
+        'url': 'http://beeg.com/5416503',
+        'md5': '634526ae978711f6b748fe0dd6c11f57',
+        'info_dict': {
+            'id': '5416503',
+            'ext': 'mp4',
+            'title': 'Sultry Striptease',
+            'description': 'md5:6db3c6177972822aaba18652ff59c773',
+            'categories': list,  # NSFW
+            'thumbnail': 're:https?://.*\.jpg$',
+            'age_limit': 18,
+        }
+    }
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        video_id = mobj.group('id')
+
+        webpage = self._download_webpage(url, video_id)
+
+        quality_arr = self._search_regex(
+            r'(?s)var\s+qualityArr\s*=\s*{\s*(.+?)\s*}', webpage, 'quality formats')
+
+        formats = [{
+            'url': fmt[1],
+            'format_id': fmt[0],
+            'height': int(fmt[0][:-1]),
+        } for fmt in re.findall(r"'([^']+)'\s*:\s*'([^']+)'", quality_arr)]
+
+        self._sort_formats(formats)
+
+        title = self._html_search_regex(
+            r'<title>([^<]+)\s*-\s*beeg\.?</title>', webpage, 'title')
+        
+        description = self._html_search_regex(
+            r'<meta name="description" content="([^"]*)"',
+            webpage, 'description', fatal=False)
+        thumbnail = self._html_search_regex(
+            r'\'previewer.url\'\s*:\s*"([^"]*)"',
+            webpage, 'thumbnail', fatal=False)
+
+        categories_str = self._html_search_regex(
+            r'<meta name="keywords" content="([^"]+)"', webpage, 'categories', fatal=False)
+        categories = (
+            None if categories_str is None
+            else categories_str.split(','))
+
+        return {
+            'id': video_id,
+            'title': title,
+            'description': description,
+            'thumbnail': thumbnail,
+            'categories': categories,
+            'formats': formats,
+            'age_limit': 18,
+        }

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/comedycentral.py

@@ -43,14 +43,14 @@ class ComedyCentralShowsIE(InfoExtractor):
                           (?P<showname>thedailyshow|thecolbertreport)\.(?:cc\.)?com/
                          ((?:full-)?episodes/(?:[0-9a-z]{6}/)?(?P<episode>.*)|
                           (?P<clip>
-                              (?:(?:guests/[^/]+|videos|video-playlists|special-editions)/[^/]+/(?P<videotitle>[^/?#]+))
+                              (?:(?:guests/[^/]+|videos|video-playlists|special-editions|news-team/[^/]+)/[^/]+/(?P<videotitle>[^/?#]+))
                               |(the-colbert-report-(videos|collections)/(?P<clipID>[0-9]+)/[^/]*/(?P<cntitle>.*?))
                               |(watch/(?P<date>[^/]*)/(?P<tdstitle>.*))
                           )|
                           (?P<interview>
                               extended-interviews/(?P<interID>[0-9a-z]+)/(?:playlist_tds_extended_)?(?P<interview_title>.*?)(/.*?)?)))
                      (?:[?#].*|$)'''
-    _TEST = {
+    _TESTS = [{
         'url': 'http://thedailyshow.cc.com/watch/thu-december-13-2012/kristen-stewart',
         'md5': '4e2f5cb088a83cd8cdb7756132f9739d',
         'info_dict': {
@@ -61,7 +61,34 @@ class ComedyCentralShowsIE(InfoExtractor):
             'uploader': 'thedailyshow',
             'title': 'thedailyshow kristen-stewart part 1',
         }
-    }
+    }, {
+        'url': 'http://thedailyshow.cc.com/extended-interviews/xm3fnq/andrew-napolitano-extended-interview',
+        'only_matching': True,
+    }, {
+        'url': 'http://thecolbertreport.cc.com/videos/29w6fx/-realhumanpraise-for-fox-news',
+        'only_matching': True,
+    }, {
+        'url': 'http://thecolbertreport.cc.com/videos/gh6urb/neil-degrasse-tyson-pt--1?xrs=eml_col_031114',
+        'only_matching': True,
+    }, {
+        'url': 'http://thedailyshow.cc.com/guests/michael-lewis/3efna8/exclusive---michael-lewis-extended-interview-pt--3',
+        'only_matching': True,
+    }, {
+        'url': 'http://thedailyshow.cc.com/episodes/sy7yv0/april-8--2014---denis-leary',
+        'only_matching': True,
+    }, {
+        'url': 'http://thecolbertreport.cc.com/episodes/8ase07/april-8--2014---jane-goodall',
+        'only_matching': True,
+    }, {
+        'url': 'http://thedailyshow.cc.com/video-playlists/npde3s/the-daily-show-19088-highlights',
+        'only_matching': True,
+    }, {
+        'url': 'http://thedailyshow.cc.com/special-editions/2l8fdb/special-edition---a-look-back-at-food',
+        'only_matching': True,
+    }, {
+        'url': 'http://thedailyshow.cc.com/news-team/michael-che/7wnfel/we-need-to-talk-about-israel',
+        'only_matching': True,
+    }]
 
     _available_formats = ['3500', '2200', '1700', '1200', '750', '400']
 
@@ -185,6 +212,9 @@ class ComedyCentralShowsIE(InfoExtractor):
                     'ext': self._video_extensions.get(format, 'mp4'),
                     'height': h,
                     'width': w,
+
+                    'format_note': 'HTTP 400 at the moment (patches welcome!)',
+                    'preference': -100,
                 })
                 formats.append({
                     'format_id': 'rtmp-%s' % format,

youtube_dl/extractor/common.py

@@ -677,9 +677,12 @@ class InfoExtractor(object):
                 }
                 codecs = last_info.get('CODECS')
                 if codecs:
-                    video, audio = codecs.split(',')
-                    f['vcodec'] = video.partition('.')[0]
-                    f['acodec'] = audio.partition('.')[0]
+                    # TODO: looks like video codec is not always necessarily goes first
+                    va_codecs = codecs.split(',')
+                    if va_codecs[0]:
+                        f['vcodec'] = va_codecs[0].partition('.')[0]
+                    if len(va_codecs) > 1 and va_codecs[1]:
+                        f['acodec'] = va_codecs[1].partition('.')[0]
                 resolution = last_info.get('RESOLUTION')
                 if resolution:
                     width_str, height_str = resolution.split('x')

youtube_dl/extractor/crunchyroll.py

@@ -5,6 +5,7 @@ import re
 import json
 import base64
 import zlib
+import xml.etree.ElementTree
 
 from hashlib import sha1
 from math import pow, sqrt, floor
@@ -17,6 +18,7 @@ from ..utils import (
     intlist_to_bytes,
     unified_strdate,
     clean_html,
+    urlencode_postdata,
 )
 from ..aes import (
     aes_cbc_decrypt,
@@ -51,6 +53,26 @@ class CrunchyrollIE(InfoExtractor):
         '1080': ('80', '108'),
     }
 
+    def _login(self):
+        (username, password) = self._get_login_info()
+        if username is None:
+            return
+        self.report_login()
+        login_url = 'https://www.crunchyroll.com/?a=formhandler'
+        data = urlencode_postdata({
+            'formname': 'RpcApiUser_Login',
+            'name': username,
+            'password': password,
+        })
+        login_request = compat_urllib_request.Request(login_url, data)
+        login_request.add_header('Content-Type', 'application/x-www-form-urlencoded')
+        self._download_webpage(login_request, None, False, 'Wrong login info')
+
+
+    def _real_initialize(self):
+        self._login()
+
+
     def _decrypt_subtitles(self, data, iv, id):
         data = bytes_to_intlist(data)
         iv = bytes_to_intlist(iv)
@@ -97,6 +119,75 @@ class CrunchyrollIE(InfoExtractor):
             output += '%d\n%s --> %s\n%s\n\n' % (i, start, end, text)
         return output
 
+    def _convert_subtitles_to_ass(self, subtitles):
+        output = ''
+
+        def ass_bool(strvalue):
+            assvalue = '0'
+            if strvalue == '1':
+                assvalue = '-1'
+            return assvalue
+
+        sub_root = xml.etree.ElementTree.fromstring(subtitles)
+        if not sub_root:
+            return output
+
+        output = '[Script Info]\n'
+        output += 'Title: %s\n' % sub_root.attrib["title"]
+        output += 'ScriptType: v4.00+\n'
+        output += 'WrapStyle: %s\n' % sub_root.attrib["wrap_style"]
+        output += 'PlayResX: %s\n' % sub_root.attrib["play_res_x"]
+        output += 'PlayResY: %s\n' % sub_root.attrib["play_res_y"]
+        output += """ScaledBorderAndShadow: yes
+
+[V4+ Styles]
+Format: Name, Fontname, Fontsize, PrimaryColour, SecondaryColour, OutlineColour, BackColour, Bold, Italic, Underline, StrikeOut, ScaleX, ScaleY, Spacing, Angle, BorderStyle, Outline, Shadow, Alignment, MarginL, MarginR, MarginV, Encoding
+"""
+        for style in sub_root.findall('./styles/style'):
+            output += 'Style: ' + style.attrib["name"]
+            output += ',' + style.attrib["font_name"]
+            output += ',' + style.attrib["font_size"]
+            output += ',' + style.attrib["primary_colour"]
+            output += ',' + style.attrib["secondary_colour"]
+            output += ',' + style.attrib["outline_colour"]
+            output += ',' + style.attrib["back_colour"]
+            output += ',' + ass_bool(style.attrib["bold"])
+            output += ',' + ass_bool(style.attrib["italic"])
+            output += ',' + ass_bool(style.attrib["underline"])
+            output += ',' + ass_bool(style.attrib["strikeout"])
+            output += ',' + style.attrib["scale_x"]
+            output += ',' + style.attrib["scale_y"]
+            output += ',' + style.attrib["spacing"]
+            output += ',' + style.attrib["angle"]
+            output += ',' + style.attrib["border_style"]
+            output += ',' + style.attrib["outline"]
+            output += ',' + style.attrib["shadow"]
+            output += ',' + style.attrib["alignment"]
+            output += ',' + style.attrib["margin_l"]
+            output += ',' + style.attrib["margin_r"]
+            output += ',' + style.attrib["margin_v"]
+            output += ',' + style.attrib["encoding"]
+            output += '\n'
+
+        output += """
+[Events]
+Format: Layer, Start, End, Style, Name, MarginL, MarginR, MarginV, Effect, Text
+"""
+        for event in sub_root.findall('./events/event'):
+            output += 'Dialogue: 0'
+            output += ',' + event.attrib["start"]
+            output += ',' + event.attrib["end"]
+            output += ',' + event.attrib["style"]
+            output += ',' + event.attrib["name"]
+            output += ',' + event.attrib["margin_l"]
+            output += ',' + event.attrib["margin_r"]
+            output += ',' + event.attrib["margin_v"]
+            output += ',' + event.attrib["effect"]
+            output += ',' + event.attrib["text"]
+            output += '\n'
+
+        return output
+
     def _real_extract(self,url):
         mobj = re.match(self._VALID_URL, url)
         video_id = mobj.group('video_id')
@@ -158,6 +249,7 @@ class CrunchyrollIE(InfoExtractor):
             })
 
         subtitles = {}
+        sub_format = self._downloader.params.get('subtitlesformat', 'srt')
         for sub_id, sub_name in re.findall(r'\?ssid=([0-9]+)" title="([^"]+)', webpage):
             sub_page = self._download_webpage('http://www.crunchyroll.com/xml/?req=RpcApiSubtitle_GetXml&subtitle_script_id='+sub_id,\
                                               video_id, note='Downloading subtitles for '+sub_name)
@@ -174,7 +266,10 @@ class CrunchyrollIE(InfoExtractor):
             lang_code = self._search_regex(r'lang_code=["\']([^"\']+)', subtitle, 'subtitle_lang_code', fatal=False)
             if not lang_code:
                 continue
-            subtitles[lang_code] = self._convert_subtitles_to_srt(subtitle)
+            if sub_format == 'ass':
+                subtitles[lang_code] = self._convert_subtitles_to_ass(subtitle)
+            else:
+                subtitles[lang_code] = self._convert_subtitles_to_srt(subtitle)
 
         return {
             'id':          video_id,

youtube_dl/extractor/drtuber.py

@@ -0,0 +1,70 @@
+from __future__ import unicode_literals
+
+import re
+
+from .common import InfoExtractor
+from ..utils import str_to_int
+
+
+class DrTuberIE(InfoExtractor):
+    _VALID_URL = r'https?://(?:www\.)?drtuber\.com/video/(?P<id>\d+)/(?P<display_id>[\w-]+)'
+    _TEST = {
+        'url': 'http://www.drtuber.com/video/1740434/hot-perky-blonde-naked-golf',
+        'md5': '93e680cf2536ad0dfb7e74d94a89facd',
+        'info_dict': {
+            'id': '1740434',
+            'display_id': 'hot-perky-blonde-naked-golf',
+            'ext': 'mp4',
+            'title': 'Hot Perky Blonde Naked Golf',
+            'like_count': int,
+            'dislike_count': int,
+            'comment_count': int,
+            'categories': list,  # NSFW
+            'thumbnail': 're:https?://.*\.jpg$',
+            'age_limit': 18,
+        }
+    }
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        video_id = mobj.group('id')
+        display_id = mobj.group('display_id')
+
+        webpage = self._download_webpage(url, display_id)
+
+        video_url = self._html_search_regex(
+            r'<source src="([^"]+)"', webpage, 'video URL')
+
+        title = self._html_search_regex(
+            r'<title>([^<]+)\s*-\s*Free', webpage, 'title')
+
+        thumbnail = self._html_search_regex(
+            r'poster="([^"]+)"',
+            webpage, 'thumbnail', fatal=False)
+
+        like_count = str_to_int(self._html_search_regex(
+            r'<span id="rate_likes">\s*<img[^>]+>\s*<span>([\d,\.]+)</span>',
+            webpage, 'like count', fatal=False))
+        dislike_count = str_to_int(self._html_search_regex(
+            r'<span id="rate_dislikes">\s*<img[^>]+>\s*<span>([\d,\.]+)</span>',
+            webpage, 'like count', fatal=False))
+        comment_count = str_to_int(self._html_search_regex(
+            r'<span class="comments_count">([\d,\.]+)</span>',
+            webpage, 'comment count', fatal=False))
+
+        cats_str = self._html_search_regex(
+            r'<meta name="keywords" content="([^"]+)"', webpage, 'categories', fatal=False)
+        categories = None if cats_str is None else cats_str.split(' ')
+
+        return {
+            'id': video_id,
+            'display_id': display_id,
+            'url': video_url,
+            'title': title,
+            'thumbnail': thumbnail,
+            'like_count': like_count,
+            'dislike_count': dislike_count,
+            'comment_count': comment_count,
+            'categories': categories,
+            'age_limit': self._rta_search(webpage),
+        }

youtube_dl/extractor/empflix.py

@@ -1,58 +1,25 @@
 from __future__ import unicode_literals
 
-import re
-
-from .common import InfoExtractor
-from ..utils import fix_xml_ampersands
+from .tnaflix import TNAFlixIE
 
 
-class EmpflixIE(InfoExtractor):
-    _VALID_URL = r'^https?://www\.empflix\.com/videos/.*?-(?P<id>[0-9]+)\.html'
+class EMPFlixIE(TNAFlixIE):
+    _VALID_URL = r'^https?://www\.empflix\.com/videos/(?P<display_id>[0-9a-zA-Z-]+)-(?P<id>[0-9]+)\.html'
+
+    _TITLE_REGEX = r'name="title" value="(?P<title>[^"]*)"'
+    _DESCRIPTION_REGEX = r'name="description" value="([^"]*)"'
+    _CONFIG_REGEX = r'flashvars\.config\s*=\s*escape\("([^"]+)"'
+
     _TEST = {
         'url': 'http://www.empflix.com/videos/Amateur-Finger-Fuck-33051.html',
         'md5': 'b1bc15b6412d33902d6e5952035fcabc',
         'info_dict': {
             'id': '33051',
+            'display_id': 'Amateur-Finger-Fuck',
             'ext': 'mp4',
             'title': 'Amateur Finger Fuck',
             'description': 'Amateur solo finger fucking.',
+            'thumbnail': 're:https?://.*\.jpg$',
             'age_limit': 18,
         }
     }
-
-    def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
-        video_id = mobj.group('id')
-
-        webpage = self._download_webpage(url, video_id)
-        age_limit = self._rta_search(webpage)
-
-        video_title = self._html_search_regex(
-            r'name="title" value="(?P<title>[^"]*)"', webpage, 'title')
-        video_description = self._html_search_regex(
-            r'name="description" value="([^"]*)"', webpage, 'description', fatal=False)
-
-        cfg_url = self._html_search_regex(
-            r'flashvars\.config = escape\("([^"]+)"',
-            webpage, 'flashvars.config')
-
-        cfg_xml = self._download_xml(
-            cfg_url, video_id, note='Downloading metadata',
-            transform_source=fix_xml_ampersands)
-
-        formats = [
-            {
-                'url': item.find('videoLink').text,
-                'format_id': item.find('res').text,
-            } for item in cfg_xml.findall('./quality/item')
-        ]
-        thumbnail = cfg_xml.find('./startThumb').text
-
-        return {
-            'id': video_id,
-            'title': video_title,
-            'description': video_description,
-            'thumbnail': thumbnail,
-            'formats': formats,
-            'age_limit': age_limit,
-        }

youtube_dl/extractor/eporner.py

@@ -0,0 +1,75 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+import re
+
+from .common import InfoExtractor
+from ..utils import (
+    parse_duration,
+    str_to_int,
+)
+
+
+class EpornerIE(InfoExtractor):
+    _VALID_URL = r'https?://(?:www\.)?eporner\.com/hd-porn/(?P<id>\d+)/(?P<display_id>[\w-]+)'
+    _TEST = {
+        'url': 'http://www.eporner.com/hd-porn/95008/Infamous-Tiffany-Teen-Strip-Tease-Video/',
+        'md5': '3b427ae4b9d60619106de3185c2987cd',
+        'info_dict': {
+            'id': '95008',
+            'display_id': 'Infamous-Tiffany-Teen-Strip-Tease-Video',
+            'ext': 'flv',
+            'title': 'Infamous Tiffany Teen Strip Tease Video',
+            'duration': 194,
+            'view_count': int,
+            'age_limit': 18,
+        }
+    }
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        video_id = mobj.group('id')
+        display_id = mobj.group('display_id')
+
+        webpage = self._download_webpage(url, display_id)
+        title = self._html_search_regex(
+            r'<title>(.*?) - EPORNER', webpage, 'title')
+
+        redirect_code = self._html_search_regex(
+            r'<script type="text/javascript" src="/config5/%s/([a-f\d]+)/">' % video_id,
+            webpage, 'redirect_code')
+        redirect_url = 'http://www.eporner.com/config5/%s/%s' % (video_id, redirect_code)
+        player_code = self._download_webpage(
+            redirect_url, display_id, note='Downloading player config')
+
+        sources = self._search_regex(
+            r'(?s)sources\s*:\s*\[\s*({.+?})\s*\]', player_code, 'sources')
+
+        formats = []
+        for video_url, format_id in re.findall(r'file\s*:\s*"([^"]+)",\s*label\s*:\s*"([^"]+)"', sources):
+            fmt = {
+                'url': video_url,
+                'format_id': format_id,
+            }
+            m = re.search(r'^(\d+)', format_id)
+            if m:
+                fmt['height'] = int(m.group(1))
+            formats.append(fmt)
+        self._sort_formats(formats)
+
+        duration = parse_duration(self._search_regex(
+            r'class="mbtim">([0-9:]+)</div>', webpage, 'duration',
+            fatal=False))
+        view_count = str_to_int(self._search_regex(
+            r'id="cinemaviews">\s*([0-9,]+)\s*<small>views',
+            webpage, 'view count', fatal=False))
+
+        return {
+            'id': video_id,
+            'display_id': display_id,
+            'title': title,
+            'duration': duration,
+            'view_count': view_count,
+            'formats': formats,
+            'age_limit': self._rta_search(webpage),
+        }

youtube_dl/extractor/facebook.py

@@ -79,7 +79,8 @@ class FacebookIE(InfoExtractor):
 
             check_form = {
                 'fb_dtsg': self._search_regex(r'name="fb_dtsg" value="(.+?)"', login_results, 'fb_dtsg'),
-                'h': self._search_regex(r'name="h" value="(\w*?)"', login_results, 'h'),
+                'h': self._search_regex(
+                    r'name="h"\s+(?:\w+="[^"]+"\s+)*?value="([^"]+)"', login_results, 'h'),
                 'name_action_selected': 'dont_save',
             }
             check_req = compat_urllib_request.Request(self._CHECKPOINT_URL, urlencode_postdata(check_form))

youtube_dl/extractor/generic.py

@@ -366,7 +366,22 @@ class GenericIE(InfoExtractor):
                 'extract_flat': False,
                 'skip_download': True,
             }
-        }
+        },
+        # MLB embed
+        {
+            'url': 'http://umpire-empire.com/index.php/topic/58125-laz-decides-no-thats-low/',
+            'md5': '96f09a37e44da40dd083e12d9a683327',
+            'info_dict': {
+                'id': '33322633',
+                'ext': 'mp4',
+                'title': 'Ump changes call to ball',
+                'description': 'md5:71c11215384298a172a6dcb4c2e20685',
+                'duration': 48,
+                'timestamp': 1401537900,
+                'upload_date': '20140531',
+                'thumbnail': 're:^https?://.*\.jpg$',
+            },
+        },
     ]
 
     def report_download_webpage(self, video_id):
@@ -809,6 +824,12 @@ class GenericIE(InfoExtractor):
         if mobj is not None:
             return self.url_result(mobj.group('url'), 'SBS')
 
+        mobj = re.search(
+            r'<iframe[^>]+?src=(["\'])(?P<url>https?://m\.mlb\.com/shared/video/embed/embed\.html\?.+?)\1',
+            webpage)
+        if mobj is not None:
+            return self.url_result(mobj.group('url'), 'MLB')
+
         # Start with something easy: JW Player in SWFObject
         found = re.findall(r'flashvars: [\'"](?:.*&)?file=(http[^\'"&]*)', webpage)
         if not found:

youtube_dl/extractor/hornbunny.py

@@ -0,0 +1,56 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+import re
+
+from .common import InfoExtractor
+from ..utils import (
+    int_or_none,
+    parse_duration,
+)
+
+
+class HornBunnyIE(InfoExtractor):
+    _VALID_URL = r'http?://(?:www\.)?hornbunny\.com/videos/(?P<title_dash>[a-z-]+)-(?P<id>\d+)\.html'
+    _TEST = {
+        'url': 'http://hornbunny.com/videos/panty-slut-jerk-off-instruction-5227.html',
+        'md5': '95e40865aedd08eff60272b704852ad7',
+        'info_dict': {
+            'id': '5227',
+            'ext': 'flv',
+            'title': 'panty slut jerk off instruction',
+            'duration': 550,
+            'age_limit': 18,
+        }
+    }
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        video_id = mobj.group('id')
+
+        webpage = self._download_webpage(
+            url, video_id, note='Downloading initial webpage')
+        title = self._html_search_regex(
+            r'class="title">(.*?)</h2>', webpage, 'title')
+        redirect_url = self._html_search_regex(
+            r'pg&settings=(.*?)\|0"\);', webpage, 'title')
+        webpage2 = self._download_webpage(redirect_url, video_id)
+        video_url = self._html_search_regex(
+            r'flvMask:(.*?);', webpage2, 'video_url')
+        
+        duration = parse_duration(self._search_regex(
+            r'<strong>Runtime:</strong>\s*([0-9:]+)</div>',
+            webpage, 'duration', fatal=False))
+        view_count = int_or_none(self._search_regex(
+            r'<strong>Views:</strong>\s*(\d+)</div>',
+            webpage, 'view count', fatal=False))
+
+        return {
+            'id': video_id,
+            'url': video_url,
+            'title': title,
+            'ext': 'flv',
+            'duration': duration,
+            'view_count': view_count,
+            'age_limit': 18,
+        }

youtube_dl/extractor/mlb.py

@@ -11,7 +11,7 @@ from ..utils import (
 
 
 class MLBIE(InfoExtractor):
-    _VALID_URL = r'https?://m\.mlb\.com/(?:.*?/)?video/(?:topic/[\da-z_-]+/)?v(?P<id>n?\d+)'
+    _VALID_URL = r'https?://m\.mlb\.com/(?:(?:.*?/)?video/(?:topic/[\da-z_-]+/)?v|shared/video/embed/embed\.html\?.*?\bcontent_id=)(?P<id>n?\d+)'
     _TESTS = [
         {
             'url': 'http://m.mlb.com/sea/video/topic/51231442/v34698933/nymsea-ackley-robs-a-home-run-with-an-amazing-catch/?c_id=sea',
@@ -69,6 +69,10 @@ class MLBIE(InfoExtractor):
                 'thumbnail': 're:^https?://.*\.jpg$',
             },
         },
+        {
+            'url': 'http://m.mlb.com/shared/video/embed/embed.html?content_id=35692085&topic_id=6479266&width=400&height=224&property=mlb',
+            'only_matching': True,
+        },
     ]
 
     def _real_extract(self, url):

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/nosvideo.py

@@ -0,0 +1,65 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+import re
+
+from .common import InfoExtractor
+from ..utils import (
+    compat_urllib_parse,
+    compat_urllib_request,
+    xpath_with_ns,
+)
+
+_x = lambda p: xpath_with_ns(p, {'xspf': 'http://xspf.org/ns/0/'})
+_find = lambda el, p: el.find(_x(p)).text.strip()
+
+
+class NosVideoIE(InfoExtractor):
+    _VALID_URL = r'https?://(?:www\.)?nosvideo\.com/' + \
+                 '(?:embed/|\?v=)(?P<id>[A-Za-z0-9]{12})/?'
+    _PLAYLIST_URL = 'http://nosvideo.com/xml/{xml_id:s}.xml'
+    _TEST = {
+        'url': 'http://nosvideo.com/?v=drlp6s40kg54',
+        'md5': '4b4ac54c6ad5d70ab88f2c2c6ccec71c',
+        'info_dict': {
+            'id': 'drlp6s40kg54',
+            'ext': 'mp4',
+            'title': 'big_buck_bunny_480p_surround-fix.avi.mp4',
+            'thumbnail': 're:^https?://.*\.jpg$',
+        }
+    }
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        video_id = mobj.group('id')
+
+        fields = {
+            'id': video_id,
+            'op': 'download1',
+            'method_free': 'Continue to Video',
+        }
+        post = compat_urllib_parse.urlencode(fields)
+        req = compat_urllib_request.Request(url, post)
+        req.add_header('Content-type', 'application/x-www-form-urlencoded')
+        webpage = self._download_webpage(req, video_id,
+                                         'Downloading download page')
+        xml_id = self._search_regex(r'php\|([^\|]+)\|', webpage, 'XML ID')
+        playlist_url = self._PLAYLIST_URL.format(xml_id=xml_id)
+        playlist = self._download_xml(playlist_url, video_id)
+
+        track = playlist.find(_x('.//xspf:track'))
+        title = _find(track, './xspf:title')
+        url = _find(track, './xspf:file')
+        thumbnail = _find(track, './xspf:image')
+
+        formats = [{
+            'format_id': 'sd',
+            'url': url,
+        }]
+
+        return {
+            'id': video_id,
+            'title': title,
+            'thumbnail': thumbnail,
+            'formats': formats,
+        }

youtube_dl/extractor/npo.py

@@ -5,6 +5,7 @@ import re
 from .common import InfoExtractor
 from ..utils import (
     unified_strdate,
+    qualities,
 )
 
 
@@ -17,7 +18,7 @@ class NPOIE(InfoExtractor):
         'md5': '4b3f9c429157ec4775f2c9cb7b911016',
         'info_dict': {
             'id': 'VPWON_1220719',
-            'ext': 'mp4',
+            'ext': 'm4v',
             'title': 'Nieuwsuur',
             'description': 'Dagelijks tussen tien en elf: nieuws, sport en achtergronden.',
             'upload_date': '20140622',
@@ -39,24 +40,32 @@ class NPOIE(InfoExtractor):
             video_id,
             note='Downloading token'
         )
-        token = self._search_regex(r'npoplayer.token = "(.+?)"', token_page, 'token')
-        streams_info = self._download_json(
-            'http://ida.omroep.nl/odi/?prid=%s&puboptions=h264_std&adaptive=yes&token=%s' % (video_id, token),
-            video_id
-        )
+        token = self._search_regex(r'npoplayer\.token = "(.+?)"', token_page, 'token')
 
-        stream_info = self._download_json(
-            streams_info['streams'][0] + '&type=json',
-            video_id,
-            'Downloading stream info'
-        )
+        formats = []
+        quality = qualities(['adaptive', 'h264_sb', 'h264_bb', 'h264_std'])
+        for format_id in metadata['pubopties']:
+            streams_info = self._download_json(
+                'http://ida.omroep.nl/odi/?prid=%s&puboptions=%s&adaptive=yes&token=%s' % (video_id, format_id, token),
+                video_id, 'Downloading %s streams info' % format_id)
+            stream_info = self._download_json(
+                streams_info['streams'][0] + '&type=json',
+                video_id, 'Downloading %s stream info' % format_id)
+            if format_id == 'adaptive':
+                formats.extend(self._extract_m3u8_formats(stream_info['url'], video_id))
+            else:
+                formats.append({
+                    'url': stream_info['url'],
+                    'format_id': format_id,
+                    'quality': quality(format_id),
+                })
+        self._sort_formats(formats)
 
         return {
             'id': video_id,
             'title': metadata['titel'],
-            'ext': 'mp4',
-            'url': stream_info['url'],
             'description': metadata['info'],
             'thumbnail': metadata['images'][-1]['url'],
             'upload_date': unified_strdate(metadata['gidsdatum']),
+            'formats': formats,
         }

youtube_dl/extractor/promptfile.py

@@ -0,0 +1,67 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+import re
+
+from .common import InfoExtractor
+from ..utils import (
+    ExtractorError,
+    determine_ext,
+    compat_urllib_parse,
+    compat_urllib_request,
+)
+
+
+class PromptFileIE(InfoExtractor):
+    _VALID_URL = r'https?://(?:www\.)?promptfile\.com/l/(?P<id>[0-9A-Z\-]+)'
+    _FILE_NOT_FOUND_REGEX = r'<div.+id="not_found_msg".+>.+</div>[^-]'
+    _TEST = {
+        'url': 'http://www.promptfile.com/l/D21B4746E9-F01462F0FF',
+        'md5': 'd1451b6302da7215485837aaea882c4c',
+        'info_dict': {
+            'id': 'D21B4746E9-F01462F0FF',
+            'ext': 'mp4',
+            'title': 'Birds.mp4',
+            'thumbnail': 're:^https?://.*\.jpg$',
+        }
+    }
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        video_id = mobj.group('id')
+        webpage = self._download_webpage(url, video_id)
+
+        if re.search(self._FILE_NOT_FOUND_REGEX, webpage) is not None:
+            raise ExtractorError('Video %s does not exist' % video_id,
+                                 expected=True)
+
+        fields = dict(re.findall(r'''(?x)type="hidden"\s+
+            name="(.+?)"\s+
+            value="(.*?)"
+            ''', webpage))
+        post = compat_urllib_parse.urlencode(fields)
+        req = compat_urllib_request.Request(url, post)
+        req.add_header('Content-type', 'application/x-www-form-urlencoded')
+        webpage = self._download_webpage(
+            req, video_id, 'Downloading video page')
+
+        url = self._html_search_regex(r'url:\s*\'([^\']+)\'', webpage, 'URL')
+        title = self._html_search_regex(
+            r'<span.+title="([^"]+)">', webpage, 'title')
+        thumbnail = self._html_search_regex(
+            r'<div id="player_overlay">.*button>.*?<img src="([^"]+)"',
+            webpage, 'thumbnail', fatal=False, flags=re.DOTALL)
+
+        formats = [{
+            'format_id': 'sd',
+            'url': url,
+            'ext': determine_ext(title),
+        }]
+        self._sort_formats(formats)
+
+        return {
+            'id': video_id,
+            'title': title,
+            'thumbnail': thumbnail,
+            'formats': formats,
+        }

youtube_dl/extractor/rtlnl.py

@@ -12,22 +12,16 @@ class RtlXlIE(InfoExtractor):
 
     _TEST = {
         'url': 'http://www.rtlxl.nl/#!/rtl-nieuws-132237/6e4203a6-0a5e-3596-8424-c599a59e0677',
+        'md5': 'cc16baa36a6c169391f0764fa6b16654',
         'info_dict': {
             'id': '6e4203a6-0a5e-3596-8424-c599a59e0677',
-            'ext': 'flv',
+            'ext': 'mp4',
             'title': 'RTL Nieuws - Laat',
-            'description': 'Dagelijks het laatste nieuws uit binnen- en '
-                'buitenland. Voor nog meer nieuws kunt u ook gebruikmaken van '
-                'onze mobiele apps.',
+            'description': 'md5:6b61f66510c8889923b11f2778c72dc5',
             'timestamp': 1408051800,
             'upload_date': '20140814',
             'duration': 576.880,
         },
-        'params': {
-            # We download the first bytes of the first fragment, it can't be
-            # processed by the f4m downloader beacuse it isn't complete
-            'skip_download': True,
-        },
     }
 
     def _real_extract(self, url):
@@ -41,14 +35,32 @@ class RtlXlIE(InfoExtractor):
         material = info['material'][0]
         episode_info = info['episodes'][0]
 
-        f4m_url = 'http://manifest.us.rtl.nl' + material['videopath']
         progname = info['abstracts'][0]['name']
         subtitle = material['title'] or info['episodes'][0]['name']
 
+        videopath = material['videopath']
+        f4m_url = 'http://manifest.us.rtl.nl' + videopath
+
+        formats = self._extract_f4m_formats(f4m_url, uuid)
+
+        video_urlpart = videopath.split('/flash/')[1][:-4]
+        PG_URL_TEMPLATE = 'http://pg.us.rtl.nl/rtlxl/network/%s/progressive/%s.mp4'
+
+        formats.extend([
+            {
+                'url': PG_URL_TEMPLATE % ('a2m', video_urlpart),
+                'format_id': 'pg-sd',
+            },
+            {
+                'url': PG_URL_TEMPLATE % ('a3m', video_urlpart),
+                'format_id': 'pg-hd',
+            }
+        ])
+
         return {
             'id': uuid,
             'title': '%s - %s' % (progname, subtitle),
-            'formats': self._extract_f4m_formats(f4m_url, uuid),
+            'formats': formats,
             'timestamp': material['original_date'],
             'description': episode_info['synopsis'],
             'duration': parse_duration(material.get('duration')),

youtube_dl/extractor/sunporno.py

@@ -0,0 +1,70 @@
+from __future__ import unicode_literals
+
+import re
+
+from .common import InfoExtractor
+from ..utils import (
+    parse_duration,
+    int_or_none,
+    qualities,
+    determine_ext,
+)
+
+
+class SunPornoIE(InfoExtractor):
+    _VALID_URL = r'https?://(?:www\.)?sunporno\.com/videos/(?P<id>\d+)'
+    _TEST = {
+        'url': 'http://www.sunporno.com/videos/807778/',
+        'md5': '6457d3c165fd6de062b99ef6c2ff4c86',
+        'info_dict': {
+            'id': '807778',
+            'ext': 'flv',
+            'title': 'md5:0a400058e8105d39e35c35e7c5184164',
+            'description': 'md5:a31241990e1bd3a64e72ae99afb325fb',
+            'thumbnail': 're:^https?://.*\.jpg$',
+            'duration': 302,
+            'age_limit': 18,
+        }
+    }
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        video_id = mobj.group('id')
+
+        webpage = self._download_webpage(url, video_id)
+
+        title = self._html_search_regex(r'<title>([^<]+)</title>', webpage, 'title')
+        description = self._html_search_meta('description', webpage, 'description')
+        thumbnail = self._html_search_regex(
+            r'poster="([^"]+)"', webpage, 'thumbnail', fatal=False)
+
+        duration = parse_duration(self._search_regex(
+            r'<span>Duration: (\d+:\d+)</span>', webpage, 'duration', fatal=False))
+
+        view_count = int_or_none(self._html_search_regex(
+            r'<span class="views">(\d+)</span>', webpage, 'view count', fatal=False))
+        comment_count = int_or_none(self._html_search_regex(
+            r'(\d+)</b> Comments?', webpage, 'comment count', fatal=False))
+
+        formats = []
+        quality = qualities(['mp4', 'flv'])
+        for video_url in re.findall(r'<source src="([^"]+)"', webpage):
+            video_ext = determine_ext(video_url)
+            formats.append({
+                'url': video_url,
+                'format_id': video_ext,
+                'quality': quality(video_ext),
+            })
+        self._sort_formats(formats)
+
+        return {
+            'id': video_id,
+            'title': title,
+            'description': description,
+            'thumbnail': thumbnail,
+            'duration': duration,
+            'view_count': view_count,
+            'comment_count': comment_count,
+            'formats': formats,
+            'age_limit': 18,
+        }

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/tnaflix.py

@@ -0,0 +1,84 @@
+from __future__ import unicode_literals
+
+import re
+
+from .common import InfoExtractor
+from ..utils import (
+    parse_duration,
+    fix_xml_ampersands,
+)
+
+
+class TNAFlixIE(InfoExtractor):
+    _VALID_URL = r'https?://(?:www\.)?tnaflix\.com/(?P<cat_id>[\w-]+)/(?P<display_id>[\w-]+)/video(?P<id>\d+)'
+
+    _TITLE_REGEX = None
+    _DESCRIPTION_REGEX = r'<h3 itemprop="description">([^<]+)</h3>'
+    _CONFIG_REGEX = r'flashvars\.config\s*=\s*escape\("([^"]+)"'
+
+    _TEST = {
+        'url': 'http://www.tnaflix.com/porn-stars/Carmella-Decesare-striptease/video553878',
+        'md5': 'ecf3498417d09216374fc5907f9c6ec0',
+        'info_dict': {
+            'id': '553878',
+            'display_id': 'Carmella-Decesare-striptease',
+            'ext': 'mp4',
+            'title': 'Carmella Decesare - striptease',
+            'description': '',
+            'thumbnail': 're:https?://.*\.jpg$',
+            'duration': 91,
+            'age_limit': 18,
+        }
+    }
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        video_id = mobj.group('id')
+        display_id = mobj.group('display_id')
+
+        webpage = self._download_webpage(url, display_id)
+
+        title = self._html_search_regex(
+            self._TITLE_REGEX, webpage, 'title') if self._TITLE_REGEX else self._og_search_title(webpage)
+        description = self._html_search_regex(
+            self._DESCRIPTION_REGEX, webpage, 'description', fatal=False, default='')
+
+        age_limit = self._rta_search(webpage)
+
+        duration = self._html_search_meta('duration', webpage, 'duration', default=None)
+        if duration:
+            duration = parse_duration(duration[1:])
+
+        cfg_url = self._html_search_regex(
+            self._CONFIG_REGEX, webpage, 'flashvars.config')
+
+        cfg_xml = self._download_xml(
+            cfg_url, display_id, note='Downloading metadata',
+            transform_source=fix_xml_ampersands)
+
+        thumbnail = cfg_xml.find('./startThumb').text
+
+        formats = []
+        for item in cfg_xml.findall('./quality/item'):
+            video_url = re.sub('speed=\d+', 'speed=', item.find('videoLink').text)
+            format_id = item.find('res').text
+            fmt = {
+                'url': video_url,
+                'format_id': format_id,
+            }
+            m = re.search(r'^(\d+)', format_id)
+            if m:
+                fmt['height'] = int(m.group(1))
+            formats.append(fmt)
+        self._sort_formats(formats)
+        
+        return {
+            'id': video_id,
+            'display_id': display_id,
+            'title': title,
+            'description': description,
+            'thumbnail': thumbnail,
+            'duration': duration,
+            'age_limit': age_limit,
+            'formats': formats,
+        }

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/tudou.py

@@ -1,5 +1,7 @@
 # coding: utf-8
 
+from __future__ import unicode_literals
+
 import re
 import json
 
@@ -9,22 +11,29 @@ from .common import InfoExtractor
 class TudouIE(InfoExtractor):
     _VALID_URL = r'(?:http://)?(?:www\.)?tudou\.com/(?:listplay|programs|albumplay)/(?:view|(.+?))/(?:([^/]+)|([^/]+))(?:\.html)?'
     _TESTS = [{
-        u'url': u'http://www.tudou.com/listplay/zzdE77v6Mmo/2xN2duXMxmw.html',
-        u'file': u'159448201.f4v',
-        u'md5': u'140a49ed444bd22f93330985d8475fcb',
-        u'info_dict': {
-            u"title": u"卡马乔国足开大脚长传冲吊集锦"
+        'url': 'http://www.tudou.com/listplay/zzdE77v6Mmo/2xN2duXMxmw.html',
+        'md5': '140a49ed444bd22f93330985d8475fcb',
+        'info_dict': {
+            'id': '159448201',
+            'ext': 'f4v',
+            'title': '卡马乔国足开大脚长传冲吊集锦',
+            'thumbnail': 're:^https?://.*\.jpg$',
         }
-    },
-    {
-        u'url': u'http://www.tudou.com/albumplay/TenTw_JgiPM/PzsAs5usU9A.html',
-        u'file': u'todo.mp4',
-        u'md5': u'todo.mp4',
-        u'info_dict': {
-            u'title': u'todo.mp4',
+    }, {
+        'url': 'http://www.tudou.com/programs/view/ajX3gyhL0pc/',
+        'info_dict': {
+            'id': '117049447',
+            'ext': 'f4v',
+            'title': 'La Sylphide-Bolshoi-Ekaterina Krysanova & Vyacheslav Lopatin 2012',
+            'thumbnail': 're:^https?://.*\.jpg$',
+        }
+    }, {
+        'url': 'http://www.tudou.com/albumplay/TenTw_JgiPM/PzsAs5usU9A.html',
+        'info_dict': {
+            'title': 'todo.mp4',
         },
-        u'add_ie': [u'Youku'],
-        u'skip': u'Only works from China'
+        'add_ie': ['Youku'],
+        'skip': 'Only works from China'
     }]
 
     def _url_for_id(self, id, quality = None):
@@ -44,20 +53,22 @@ class TudouIE(InfoExtractor):
         if m and m.group(1):
             return {
                 '_type': 'url',
-                'url': u'youku:' + m.group(1),
+                'url': 'youku:' + m.group(1),
                 'ie_key': 'Youku'
             }
 
         title = self._search_regex(
-            r",kw:\s*['\"](.+?)[\"']", webpage, u'title')
+            r",kw:\s*['\"](.+?)[\"']", webpage, 'title')
         thumbnail_url = self._search_regex(
-            r",pic:\s*[\"'](.+?)[\"']", webpage, u'thumbnail URL', fatal=False)
+            r",pic:\s*[\"'](.+?)[\"']", webpage, 'thumbnail URL', fatal=False)
 
         segs_json = self._search_regex(r'segs: \'(.*)\'', webpage, 'segments')
         segments = json.loads(segs_json)
         # It looks like the keys are the arguments that have to be passed as
         # the hd field in the request url, we pick the higher
-        quality = sorted(segments.keys())[-1]
+        # Also, filter non-number qualities (see issue #3643).
+        quality = sorted(filter(lambda k: k.isdigit(), segments.keys()),
+                         key=lambda k: int(k))[-1]
         parts = segments[quality]
         result = []
         len_parts = len(parts)
@@ -67,12 +78,13 @@ class TudouIE(InfoExtractor):
             part_id = part['k']
             final_url = self._url_for_id(part_id, quality)
             ext = (final_url.split('?')[0]).split('.')[-1]
-            part_info = {'id': part_id,
-                          'url': final_url,
-                          'ext': ext,
-                          'title': title,
-                          'thumbnail': thumbnail_url,
-                          }
+            part_info = {
+                'id': '%s' % part_id,
+                'url': final_url,
+                'ext': ext,
+                'title': title,
+                'thumbnail': thumbnail_url,
+            }
             result.append(part_info)
 
         return result

youtube_dl/extractor/tvigle.py

@@ -5,80 +5,82 @@ import re
 
 from .common import InfoExtractor
 from ..utils import (
-    unified_strdate,
-    clean_html,
-    int_or_none,
+    float_or_none,
+    str_to_int,
 )
 
 
 class TvigleIE(InfoExtractor):
     IE_NAME = 'tvigle'
     IE_DESC = 'Интернет-телевидение Tvigle.ru'
-    _VALID_URL = r'http://(?:www\.)?tvigle\.ru/category/.+?[\?&]v(?:ideo)?=(?P<id>\d+)'
+    _VALID_URL = r'http://(?:www\.)?tvigle\.ru/(?:[^/]+/)+(?P<display_id>[^/]+)/$'
 
     _TESTS = [
         {
-            'url': 'http://www.tvigle.ru/category/cinema/1608/?video=503081',
-            'md5': '09afba4616666249f087efc6dcf83cb3',
+            'url': 'http://www.tvigle.ru/video/brat-2/',
+            'md5': '72cb7eab33e54314e1790da402d3c9c3',
             'info_dict': {
-                'id': '503081',
-                'ext': 'flv',
+                'id': '5119390',
+                'display_id': 'brat-2',
+                'ext': 'mp4',
                 'title': 'Брат 2 ',
-                'description': 'md5:f5a42970f50648cee3d7ad740f3ae769',
-                'upload_date': '20110919',
+                'description': 'md5:5751f4fe345a58e1692585c361294bd8',
+                'duration': 7356.369,
+                'age_limit': 0,
             },
         },
         {
-            'url': 'http://www.tvigle.ru/category/men/vysotskiy_vospominaniya02/?flt=196&v=676433',
-            'md5': 'e7efe5350dd5011d0de6550b53c3ba7b',
+            'url': 'http://www.tvigle.ru/video/vladimir-vysotskii/vedushchii-teleprogrammy-60-minut-ssha-o-vladimire-vysotskom/',
+            'md5': 'd9012d7c7c598fe7a11d7fb46dc1f574',
             'info_dict': {
-                'id': '676433',
-                'ext': 'flv',
+                'id': '5142516',
+                'ext': 'mp4',
                 'title': 'Ведущий телепрограммы «60 минут» (США) о Владимире Высоцком',
                 'description': 'md5:027f7dc872948f14c96d19b4178428a4',
-                'upload_date': '20121218',
+                'duration': 186.080,
+                'age_limit': 0,
             },
         },
     ]
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
-        video_id = mobj.group('id')
+        display_id = mobj.group('display_id')
 
-        video_data = self._download_xml(
-            'http://www.tvigle.ru/xml/single.php?obj=%s' % video_id, video_id, 'Downloading video XML')
+        webpage = self._download_webpage(url, display_id)
 
-        video = video_data.find('./video')
+        video_id = self._html_search_regex(
+            r'<li class="video-preview current_playing" id="(\d+)">', webpage, 'video id')
 
-        title = video.get('name')
-        description = video.get('anons')
-        if description:
-            description = clean_html(description)
-        thumbnail = video_data.get('img')
-        upload_date = unified_strdate(video.get('date'))
-        like_count = int_or_none(video.get('vtp'))
+        video_data = self._download_json(
+            'http://cloud.tvigle.ru/api/play/video/%s/' % video_id, display_id)
+
+        item = video_data['playlist']['items'][0]
+
+        title = item['title']
+        description = item['description']
+        thumbnail = item['thumbnail']
+        duration = float_or_none(item['durationMilliseconds'], 1000)
+        age_limit = str_to_int(item['ageRestrictions'])
 
         formats = []
-        for num, (format_id, format_note) in enumerate([['low_file', 'SQ'], ['file', 'HQ'], ['hd', 'HD 720']]):
-            video_url = video.get(format_id)
-            if not video_url:
-                continue
-            formats.append({
-                'url': video_url,
-                'format_id': format_id,
-                'format_note': format_note,
-                'quality': num,
-            })
-
+        for vcodec, fmts in item['videos'].items():
+            for quality, video_url in fmts.items():
+                formats.append({
+                    'url': video_url,
+                    'format_id': '%s-%s' % (vcodec, quality),
+                    'vcodec': vcodec,
+                    'height': int(quality[:-1]),
+                })
         self._sort_formats(formats)
 
         return {
             'id': video_id,
+            'display_id': display_id,
             'title': title,
             'description': description,
             'thumbnail': thumbnail,
-            'upload_date': upload_date,
-            'like_count': like_count,
-            'age_limit': 18,
+            'duration': duration,
+            'age_limit': age_limit,
             'formats': formats,
         }

youtube_dl/extractor/vporn.py

@@ -0,0 +1,99 @@
+from __future__ import unicode_literals
+
+import re
+
+from .common import InfoExtractor
+from ..utils import (
+    parse_duration,
+    str_to_int,
+)
+
+
+class VpornIE(InfoExtractor):
+    _VALID_URL = r'https?://(?:www\.)?vporn\.com/[^/]+/(?P<display_id>[^/]+)/(?P<id>\d+)'
+    _TEST = {
+        'url': 'http://www.vporn.com/masturbation/violet-on-her-th-birthday/497944/',
+        'md5': 'facf37c1b86546fa0208058546842c55',
+        'info_dict': {
+            'id': '497944',
+            'display_id': 'violet-on-her-th-birthday',
+            'ext': 'mp4',
+            'title': 'Violet on her 19th birthday',
+            'description': 'Violet dances in front of the camera which is sure to get you horny.',
+            'thumbnail': 're:^https?://.*\.jpg$',
+            'uploader': 'kileyGrope',
+            'categories': ['Masturbation', 'Teen'],
+            'duration': 393,
+            'age_limit': 18,
+        }
+    }
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        video_id = mobj.group('id')
+        display_id = mobj.group('display_id')
+
+        webpage = self._download_webpage(url, display_id)
+
+        title = self._html_search_regex(
+            r'videoname\s*=\s*\'([^\']+)\'', webpage, 'title').strip()
+        description = self._html_search_regex(
+            r'<div class="description_txt">(.*?)</div>', webpage, 'description', fatal=False)
+        thumbnail = self._html_search_regex(
+            r'flashvars\.imageUrl\s*=\s*"([^"]+)"', webpage, 'description', fatal=False, default=None)
+        if thumbnail:
+            thumbnail = 'http://www.vporn.com' + thumbnail
+
+        uploader = self._html_search_regex(
+            r'(?s)UPLOADED BY.*?<a href="/user/[^"]+">([^<]+)</a>',
+            webpage, 'uploader', fatal=False)
+
+        categories = re.findall(r'<a href="/cat/[^"]+">([^<]+)</a>', webpage)
+
+        duration = parse_duration(self._search_regex(
+            r'duration (\d+ min \d+ sec)', webpage, 'duration', fatal=False))
+
+        view_count = str_to_int(self._html_search_regex(
+            r'<span>([\d,\.]+) VIEWS</span>', webpage, 'view count', fatal=False))
+        like_count = str_to_int(self._html_search_regex(
+            r'<span id="like" class="n">([\d,\.]+)</span>', webpage, 'like count', fatal=False))
+        dislike_count = str_to_int(self._html_search_regex(
+            r'<span id="dislike" class="n">([\d,\.]+)</span>', webpage, 'dislike count', fatal=False))
+        comment_count = str_to_int(self._html_search_regex(
+            r'<h4>Comments \(<b>([\d,\.]+)</b>\)</h4>', webpage, 'comment count', fatal=False))
+
+        formats = []
+
+        for video in re.findall(r'flashvars\.videoUrl([^=]+?)\s*=\s*"([^"]+)"', webpage):
+            video_url = video[1]
+            fmt = {
+                'url': video_url,
+                'format_id': video[0],
+            }
+            m = re.search(r'_(?P<width>\d+)x(?P<height>\d+)_(?P<vbr>\d+)k\.mp4$', video_url)
+            if m:
+                fmt.update({
+                    'width': int(m.group('width')),
+                    'height': int(m.group('height')),
+                    'vbr': int(m.group('vbr')),
+                })
+            formats.append(fmt)
+
+        self._sort_formats(formats)
+
+        return {
+            'id': video_id,
+            'display_id': display_id,
+            'title': title,
+            'description': description,
+            'thumbnail': thumbnail,
+            'uploader': uploader,
+            'categories': categories,
+            'duration': duration,
+            'view_count': view_count,
+            'like_count': like_count,
+            'dislike_count': dislike_count,
+            'comment_count': comment_count,
+            'age_limit': 18,
+            '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/extractor/youtube.py

@@ -1,7 +1,5 @@
 # coding: utf-8
 
-import errno
-import io
 import itertools
 import json
 import os.path
@@ -21,7 +19,6 @@ from ..utils import (
     compat_str,
 
     clean_html,
-    get_cachedir,
     get_element_by_id,
     get_element_by_attribute,
     ExtractorError,
@@ -30,7 +27,6 @@ from ..utils import (
     unescapeHTML,
     unified_strdate,
     orderedSet,
-    write_json_file,
     uppercase_escape,
 )
 
@@ -316,6 +312,8 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor):
                 u"upload_date": u"20121002",
                 u"description": u"test chars:  \"'/\\ä↭𝕐\ntest URL: https://github.com/rg3/youtube-dl/issues/1892\n\nThis is a test video for youtube-dl.\n\nFor more information, contact phihag@phihag.de .",
                 u"categories": [u'Science & Technology'],
+                'like_count': int,
+                'dislike_count': int,
             }
         },
         {
@@ -433,26 +431,10 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor):
         func_id = '%s_%s_%s' % (
             player_type, player_id, self._signature_cache_id(example_sig))
         assert os.path.basename(func_id) == func_id
-        cache_dir = get_cachedir(self._downloader.params)
 
-        cache_enabled = cache_dir is not None
-        if cache_enabled:
-            cache_fn = os.path.join(os.path.expanduser(cache_dir),
-                                    u'youtube-sigfuncs',
-                                    func_id + '.json')
-            try:
-                with io.open(cache_fn, 'r', encoding='utf-8') as cachef:
-                    cache_spec = json.load(cachef)
-                return lambda s: u''.join(s[i] for i in cache_spec)
-            except IOError:
-                pass  # No cache available
-            except ValueError:
-                try:
-                    file_size = os.path.getsize(cache_fn)
-                except (OSError, IOError) as oe:
-                    file_size = str(oe)
-                self._downloader.report_warning(
-                    u'Cache %s failed (%s)' % (cache_fn, file_size))
+        cache_spec = self._downloader.cache.load(u'youtube-sigfuncs', func_id)
+        if cache_spec is not None:
+            return lambda s: u''.join(s[i] for i in cache_spec)
 
         if player_type == 'js':
             code = self._download_webpage(
@@ -470,22 +452,12 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor):
         else:
             assert False, 'Invalid player type %r' % player_type
 
-        if cache_enabled:
-            try:
-                test_string = u''.join(map(compat_chr, range(len(example_sig))))
-                cache_res = res(test_string)
-                cache_spec = [ord(c) for c in cache_res]
-                try:
-                    os.makedirs(os.path.dirname(cache_fn))
-                except OSError as ose:
-                    if ose.errno != errno.EEXIST:
-                        raise
-                write_json_file(cache_spec, cache_fn)
-            except Exception:
-                tb = traceback.format_exc()
-                self._downloader.report_warning(
-                    u'Writing cache to %r failed: %s' % (cache_fn, tb))
+        if cache_spec is None:
+            test_string = u''.join(map(compat_chr, range(len(example_sig))))
+            cache_res = res(test_string)
+            cache_spec = [ord(c) for c in cache_res]
 
+        self._downloader.cache.store(u'youtube-sigfuncs', func_id, cache_spec)
         return res
 
     def _print_sig_code(self, func, example_sig):
@@ -784,7 +756,9 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor):
             upload_date = ' '.join(re.sub(r'[/,-]', r' ', mobj.group(1)).split())
             upload_date = unified_strdate(upload_date)
 
-        m_cat_container = get_element_by_id("eow-category", video_webpage)
+        m_cat_container = self._search_regex(
+            r'(?s)<h4[^>]*>\s*Category\s*</h4>\s*<ul[^>]*>(.*?)</ul>',
+            video_webpage, 'categories', fatal=False)
         if m_cat_container:
             category = self._html_search_regex(
                 r'(?s)<a[^<]+>(.*?)</a>', m_cat_container, 'category',
@@ -813,15 +787,15 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor):
             else:
                 video_description = u''
 
-        def _extract_count(klass):
+        def _extract_count(count_name):
             count = self._search_regex(
-                r'class="%s">([\d,]+)</span>' % re.escape(klass),
-                video_webpage, klass, default=None)
+                r'id="watch-%s"[^>]*>.*?([\d,]+)\s*</span>' % re.escape(count_name),
+                video_webpage, count_name, default=None)
             if count is not None:
                 return int(count.replace(',', ''))
             return None
-        like_count = _extract_count(u'likes-count')
-        dislike_count = _extract_count(u'dislikes-count')
+        like_count = _extract_count(u'like')
+        dislike_count = _extract_count(u'dislike')
 
         # subtitles
         video_subtitles = self.extract_subtitles(video_id, video_webpage)
@@ -1051,21 +1025,26 @@ class YoutubePlaylistIE(YoutubeBaseInfoExtractor):
         self._login()
 
     def _ids_to_results(self, ids):
-        return [self.url_result(vid_id, 'Youtube', video_id=vid_id)
-                       for vid_id in ids]
+        return [
+            self.url_result(vid_id, 'Youtube', video_id=vid_id)
+            for vid_id in ids]
 
     def _extract_mix(self, playlist_id):
         # The mixes are generated from a a single video
         # the id of the playlist is just 'RD' + video_id
         url = 'https://youtube.com/watch?v=%s&list=%s' % (playlist_id[-11:], playlist_id)
-        webpage = self._download_webpage(url, playlist_id, u'Downloading Youtube mix')
+        webpage = self._download_webpage(
+            url, playlist_id, u'Downloading Youtube mix')
         search_title = lambda class_name: get_element_by_attribute('class', class_name, webpage)
-        title_span = (search_title('playlist-title') or
-            search_title('title long-title') or search_title('title'))
+        title_span = (
+            search_title('playlist-title') or
+            search_title('title long-title') or
+            search_title('title'))
         title = clean_html(title_span)
-        video_re = r'''(?x)data-video-username=".*?".*?
-                       href="/watch\?v=([0-9A-Za-z_-]{11})&amp;[^"]*?list=%s''' % re.escape(playlist_id)
-        ids = orderedSet(re.findall(video_re, webpage, flags=re.DOTALL))
+        ids = orderedSet(re.findall(
+            r'''(?xs)data-video-username=".*?".*?
+                       href="/watch\?v=([0-9A-Za-z_-]{11})&amp;[^"]*?list=%s''' % re.escape(playlist_id),
+            webpage))
         url_results = self._ids_to_results(ids)
 
         return self.playlist_result(url_results, playlist_id, title)
@@ -1158,6 +1137,7 @@ class YoutubeTopListIE(YoutubePlaylistIE):
             msg = u'Downloading Youtube mix'
             if i > 0:
                 msg += ', retry #%d' % i
+
             webpage = self._download_webpage(url, title, msg)
             ids = orderedSet(re.findall(video_re, webpage))
             if ids:
@@ -1430,12 +1410,6 @@ class YoutubeFeedsInfoExtractor(YoutubeBaseInfoExtractor):
             paging = mobj.group('paging')
         return self.playlist_result(feed_entries, playlist_title=self._PLAYLIST_TITLE)
 
-class YoutubeSubscriptionsIE(YoutubeFeedsInfoExtractor):
-    IE_DESC = u'YouTube.com subscriptions feed, "ytsubs" keyword (requires authentication)'
-    _VALID_URL = r'https?://www\.youtube\.com/feed/subscriptions|:ytsubs(?:criptions)?'
-    _FEED_NAME = 'subscriptions'
-    _PLAYLIST_TITLE = u'Youtube Subscriptions'
-
 class YoutubeRecommendedIE(YoutubeFeedsInfoExtractor):
     IE_DESC = u'YouTube.com recommended videos, "ytrec" keyword (requires authentication)'
     _VALID_URL = r'https?://www\.youtube\.com/feed/recommended|:ytrec(?:ommended)?'
@@ -1468,6 +1442,43 @@ class YoutubeFavouritesIE(YoutubeBaseInfoExtractor):
         return self.url_result(playlist_id, 'YoutubePlaylist')
 
 
+class YoutubeSubscriptionsIE(YoutubePlaylistIE):
+    IE_NAME = u'youtube:subscriptions'
+    IE_DESC = u'YouTube.com subscriptions feed, "ytsubs" keyword (requires authentication)'
+    _VALID_URL = r'https?://www\.youtube\.com/feed/subscriptions|:ytsubs(?:criptions)?'
+
+    def _real_extract(self, url):
+        title = u'Youtube Subscriptions'
+        page = self._download_webpage('https://www.youtube.com/feed/subscriptions', title)
+
+        # The extraction process is the same as for playlists, but the regex
+        # for the video ids doesn't contain an index
+        ids = []
+        more_widget_html = content_html = page
+
+        for page_num in itertools.count(1):
+            matches = re.findall(r'href="\s*/watch\?v=([0-9A-Za-z_-]{11})', content_html)
+            new_ids = orderedSet(matches)
+            ids.extend(new_ids)
+
+            mobj = re.search(r'data-uix-load-more-href="/?(?P<more>[^"]+)"', more_widget_html)
+            if not mobj:
+                break
+
+            more = self._download_json(
+                'https://youtube.com/%s' % mobj.group('more'), title,
+                'Downloading page #%s' % page_num,
+                transform_source=uppercase_escape)
+            content_html = more['content_html']
+            more_widget_html = more['load_more_widget_html']
+
+        return {
+            '_type': 'playlist',
+            'title': title,
+            'entries': self._ids_to_results(ids),
+        }
+
+
 class YoutubeTruncatedURLIE(InfoExtractor):
     IE_NAME = 'youtube:truncated_url'
     IE_DESC = False  # Do not list

youtube_dl/utils.py

@@ -1076,12 +1076,6 @@ def intlist_to_bytes(xs):
         return bytes(xs)
 
 
-def get_cachedir(params={}):
-    cache_root = os.environ.get('XDG_CACHE_HOME',
-                                os.path.expanduser('~/.cache'))
-    return params.get('cachedir', os.path.join(cache_root, 'youtube-dl'))
-
-
 # Cross-platform file locking
 if sys.platform == 'win32':
     import ctypes.wintypes
@@ -1141,10 +1135,10 @@ else:
     import fcntl
 
     def _lock_file(f, exclusive):
-        fcntl.lockf(f, fcntl.LOCK_EX if exclusive else fcntl.LOCK_SH)
+        fcntl.flock(f, fcntl.LOCK_EX if exclusive else fcntl.LOCK_SH)
 
     def _unlock_file(f):
-        fcntl.lockf(f, fcntl.LOCK_UN)
+        fcntl.flock(f, fcntl.LOCK_UN)
 
 
 class locked_file(object):
@@ -1318,9 +1312,10 @@ def str_or_none(v, default=None):
 
 
 def str_to_int(int_str):
+    """ A more relaxed version of int_or_none """
     if int_str is None:
         return None
-    int_str = re.sub(r'[,\.]', u'', int_str)
+    int_str = re.sub(r'[,\.\+]', u'', int_str)
     return int(int_str)
 
 
@@ -1332,8 +1327,10 @@ def parse_duration(s):
     if s is None:
         return None
 
+    s = s.strip()
+
     m = re.match(
-        r'(?:(?:(?P<hours>[0-9]+)[:h])?(?P<mins>[0-9]+)[:m])?(?P<secs>[0-9]+)s?(?::[0-9]+)?(?P<ms>\.[0-9]+)?$', s)
+        r'(?i)(?:(?:(?P<hours>[0-9]+)\s*(?:[:h]|hours?)\s*)?(?P<mins>[0-9]+)\s*(?:[:m]|mins?|minutes?)\s*)?(?P<secs>[0-9]+)(?P<ms>\.[0-9]+)?\s*(?:s|secs?|seconds?)?$', s)
     if not m:
         return None
     res = int(m.group('secs'))

youtube_dl/version.py

@@ -1,2 +1,2 @@
 
-__version__ = '2014.08.28.1'
+__version__ = '2014.09.04.3'