release 2014.09.01.1

Add login support to Crunchyroll extractor

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/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
@@ -85,6 +87,7 @@ from .ellentv import (
 from .elpais import ElPaisIE
 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 +137,7 @@ 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
@@ -257,6 +261,7 @@ 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 +326,7 @@ 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
@@ -392,6 +398,7 @@ 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 +1,60 @@
+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,
+        }
+    }
+
+    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,
+        }

youtube_dl/extractor/beeg.py

@@ -0,0 +1,53 @@
+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$',
+        }
+    }
+
+    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"'480p'\s*:\s*'([^']+)'", webpage, 'video URL')
+
+        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 = categories_str.split(',')
+
+        return {
+            'id': video_id,
+            'url': video_url,
+            'title': title,
+            'description': description,
+            'thumbnail': thumbnail,
+            'categories': categories,
+        }

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/eporner.py

@@ -0,0 +1,55 @@
+# 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<title_dash>[\w-]+)/?'
+    _TEST = {
+        'url': 'http://www.eporner.com/hd-porn/95008/Infamous-Tiffany-Teen-Strip-Tease-Video/',
+        'md5': '3b427ae4b9d60619106de3185c2987cd',
+        'info_dict': {
+            'id': '95008',
+            'ext': 'flv',
+            'title': 'Infamous Tiffany Teen Strip Tease Video',
+            'duration': 194,
+            'view_count': int,
+        }
+    }
+
+    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>(.*?) - 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)
+        webpage2 = self._download_webpage(redirect_url, video_id)
+        video_url = self._html_search_regex(
+            r'file: "(.*?)",', webpage2, 'video_url')
+
+        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,
+            'url': video_url,
+            'title': title,
+            'duration': duration,
+            'view_count': view_count,
+        }

youtube_dl/extractor/hornbunny.py

@@ -0,0 +1,44 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+import re
+
+from .common import InfoExtractor
+from ..utils import int_or_none
+
+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
+        }
+    }
+
+    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'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')
+        
+        mobj = re.search(r'<strong>Runtime:</strong> (?P<minutes>\d+):(?P<seconds>\d+)</div>', webpage)
+        duration = int(mobj.group('minutes')) * 60 + int(mobj.group('seconds')) if mobj else None
+
+        view_count = self._html_search_regex(r'<strong>Views:</strong>  (\d+)</div>', webpage, 'view count', fatal=False)
+
+        return {
+            'id': video_id,
+            'url': video_url,
+            'title': title,
+            'ext': 'flv',
+            'duration': duration,
+            'view_count': int_or_none(view_count),
+        }

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 +1,68 @@
+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,
+        }
+    }
+
+    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,
+        }

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/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/youtube.py

@@ -316,6 +316,8 @@ 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,
             }
         },
         {
@@ -784,7 +786,9 @@ 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 +817,15 @@ 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)
@@ -1430,12 +1434,6 @@ 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 +1466,43 @@ 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

@@ -1318,6 +1318,7 @@ 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)
@@ -1332,8 +1333,10 @@ 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]+)?$', s)
+        r'(?:(?:(?P<hours>[0-9]+)\s*(?:[:h]|hours?)\s*)?(?P<mins>[0-9]+)\s*(?:[:m]|mins?|minutes?)\s*)?(?P<secs>[0-9]+)(?P<ms>\.[0-9]+)?\s*(?:s|secs?|seconds?)?$', 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__ = '2014.09.01.1'