release 2013.09.05

'media$player' is not defined for private videos.

README.md

@@ -113,7 +113,8 @@ which means you can modify it, redistribute it or use it however you like.
 
 ## Video Format Options:
     -f, --format FORMAT        video format code, specifiy the order of
-                               preference using slashes: "-f 22/17/18"
+                               preference using slashes: "-f 22/17/18". "-f mp4"
+                               and "-f flv" are also supported
     --all-formats              download all available video formats
     --prefer-free-formats      prefer free video formats unless a specific one
                                is requested

devscripts/bash-completion.in

@@ -4,8 +4,12 @@ __youtube-dl()
     COMPREPLY=()
     cur="${COMP_WORDS[COMP_CWORD]}"
     opts="{{flags}}"
+    keywords=":ytfavorites :ytrecommended :ytsubscriptions :ytwatchlater"
 
-    if [[ ${cur} == * ]] ; then
+    if [[ ${cur} =~ : ]]; then
+        COMPREPLY=( $(compgen -W "${keywords}" -- ${cur}) )
+        return 0
+    elif [[ ${cur} == * ]] ; then
         COMPREPLY=( $(compgen -W "${opts}" -- ${cur}) )
         return 0
     fi

devscripts/gh-pages/update-sites.py

@@ -0,0 +1,33 @@
+#!/usr/bin/env python3
+
+import sys
+import os
+import textwrap
+
+# We must be able to import youtube_dl
+sys.path.append(os.path.dirname(os.path.dirname(os.path.dirname(os.path.abspath(__file__)))))
+
+import youtube_dl
+
+def main():
+    with open('supportedsites.html.in', 'r', encoding='utf-8') as tmplf:
+        template = tmplf.read()
+
+    ie_htmls = []
+    for ie in sorted(youtube_dl.gen_extractors(), key=lambda i: i.IE_NAME):
+        ie_html = '<b>{}</b>'.format(ie.IE_NAME)
+        try:
+            ie_html += ': {}'.format(ie.IE_DESC)
+        except AttributeError:
+            pass
+        if ie.working() == False:
+            ie_html += ' (Currently broken)'
+        ie_htmls.append('<li>{}</li>'.format(ie_html))
+
+    template = template.replace('@SITES@', textwrap.indent('\n'.join(ie_htmls), '\t'))
+
+    with open('supportedsites.html', 'w', encoding='utf-8') as sitesf:
+        sitesf.write(template)
+
+if __name__ == '__main__':
+    main()

devscripts/release.sh

@@ -85,6 +85,7 @@ ROOT=$(pwd)
     "$ROOT/devscripts/gh-pages/sign-versions.py" < "$ROOT/updates_key.pem"
     "$ROOT/devscripts/gh-pages/generate-download.py"
     "$ROOT/devscripts/gh-pages/update-copyright.py"
+    "$ROOT/devscripts/gh-pages/update-sites.py"
     git add *.html *.html.in update
     git commit -m "release $version"
     git show HEAD

devscripts/youtube_genalgo.py

@@ -20,15 +20,15 @@ tests = [
     # 87
     ("qwertyuioplkjhgfdsazxcvbnm1234567890QWERTYUIOPLKJHGFDSAZXCVBNM!@#$^&*()_-+={[]}|:;?/>.<",
      "uioplkjhgfdsazxcvbnm1t34567890QWE2TYUIOPLKJHGFDSAZXCVeNM!@#$^&*()_-+={[]}|:;?/>.<"),
-    # 86 - vflh9ybst 2013/08/23
+    # 86 - vflHOr_nV 2013/08/30
     ("qwertyuioplkjhgfdsazxcvbnm1234567890QWERTYUIOPLKJHGFDSAZXCVBNM!@#$%^&*()_-+={[|};?/>.<",
-     "yuioplkjhgfdsazxcvbnm1234567890QWERrYUIOPLKqHGFDSAZXCVBNM!@#$%^&*()_-+={[|};?/>.<"),
+     "?;}|[{=+._)(*&^%$#@!MNBqCXZASDFGHJKLPOIUYTREWQ<987654321mnbvcxzasdfghjklpoiuytrew"),
     # 85
     ("qwertyuioplkjhgfdsazxcvbnm1234567890QWERTYUIOPLKJHGFDSAZXCVBNM!@#$%^&*()_-+={[};?/>.<",
      ".>/?;}[{=+-_)(*&^%$#@!MNBVCXZASDFGHJKLPOIUYTREWQ0q876543r1mnbvcx9asdfghjklpoiuyt2"),
-    # 84 - vflh9ybst 2013/08/23 (sporadic)
+    # 84 - vflg0g8PQ 2013/08/29 (sporadic)
     ("qwertyuioplkjhgfdsazxcvbnm1234567890QWERTYUIOPLKJHGFDSAZXCVBNM!@#$%^&*()_-+={[};?>.<",
-     "yuioplkjhgfdsazxcvbnm1234567890QWERrYUIOPLKqHGFDSAZXCVBNM!@#$%^&*()_-+={[};?>.<"),
+     ">?;}[{=+-_)(*&^%$#@!MNBVCXZASDFGHJKLPOIUYTREWq0987654321mnbvcxzasdfghjklpoiuytr"),
     # 83
     ("qwertyuioplkjhgfdsazxcvbnm1234567890QWERTYUIOPLKJHGFDSAZXCVBNM!#$%^&*()_+={[};?/>.<",
      ".>/?;}[{=+_)(*&^%<#!MNBVCXZASPFGHJKLwOIUYTREWQ0987654321mnbvcxzasdfghjklpoiuytreq"),

test/test_download.py

@@ -127,12 +127,11 @@ def generator(test_case):
                     info_dict = json.load(infof)
                 for (info_field, expected) in tc.get('info_dict', {}).items():
                     if isinstance(expected, compat_str) and expected.startswith('md5:'):
-                        self.assertEqual(expected, 'md5:' + md5(info_dict.get(info_field)))
+                        got = 'md5:' + md5(info_dict.get(info_field))
                     else:
                         got = info_dict.get(info_field)
-                        self.assertEqual(
-                            expected, got,
-                            u'invalid value for field %s, expected %r, got %r' % (info_field, expected, got))
+                    self.assertEqual(expected, got,
+                        u'invalid value for field %s, expected %r, got %r' % (info_field, expected, got))
 
                 # If checkable fields are missing from the test case, print the info_dict
                 test_info_dict = dict((key, value if not isinstance(value, compat_str) or len(value) < 250 else 'md5:' + md5(value))

youtube_dl/__init__.py

@@ -28,6 +28,7 @@ __authors__  = (
     'Axel Noack',
     'Albert Kim',
     'Pierre Rudloff',
+    'Huarong Huo',
 )
 
 __license__ = 'Public Domain'
@@ -192,7 +193,7 @@ def parseOpts(overrideArguments=None):
 
     video_format.add_option('-f', '--format',
             action='store', dest='format', metavar='FORMAT',
-            help='video format code, specifiy the order of preference using slashes: "-f 22/17/18"')
+            help='video format code, specifiy the order of preference using slashes: "-f 22/17/18". "-f mp4" and "-f flv" are also supported')
     video_format.add_option('--all-formats',
             action='store_const', dest='format', help='download all available video formats', const='all')
     video_format.add_option('--prefer-free-formats',

youtube_dl/extractor/__init__.py

@@ -18,9 +18,11 @@ from .condenast import CondeNastIE
 from .criterion import CriterionIE
 from .cspan import CSpanIE
 from .dailymotion import DailymotionIE, DailymotionPlaylistIE
+from .daum import DaumIE
 from .depositfiles import DepositFilesIE
 from .dotsub import DotsubIE
 from .dreisat import DreiSatIE
+from .defense import DefenseGouvFrIE
 from .ehow import EHowIE
 from .eighttracks import EightTracksIE
 from .escapist import EscapistIE
@@ -56,9 +58,11 @@ from .mtv import MTVIE
 from .muzu import MuzuTVIE
 from .myspass import MySpassIE
 from .myvideo import MyVideoIE
+from .naver import NaverIE
 from .nba import NBAIE
 from .nbc import NBCNewsIE
 from .ooyala import OoyalaIE
+from .orf import ORFIE
 from .pbs import PBSIE
 from .photobucket import PhotobucketIE
 from .pornotube import PornotubeIE
@@ -88,6 +92,7 @@ from .tutv import TutvIE
 from .unistra import UnistraIE
 from .ustream import UstreamIE
 from .vbox7 import Vbox7IE
+from .veehd import VeeHDIE
 from .veoh import VeohIE
 from .vevo import VevoIE
 from .videofyme import VideofyMeIE

youtube_dl/extractor/common.py

@@ -150,7 +150,7 @@ class InfoExtractor(object):
         if m:
             encoding = m.group(1)
         else:
-            m = re.search(br'<meta[^>]+charset="?([^"]+)[ /">]',
+            m = re.search(br'<meta[^>]+charset=[\'"]?([^\'")]+)[ /\'">]',
                           webpage_bytes[:1024])
             if m:
                 encoding = m.group(1).decode('ascii')

youtube_dl/extractor/dailymotion.py

@@ -55,7 +55,8 @@ class DailymotionIE(InfoExtractor):
         embed_url = 'http://www.dailymotion.com/embed/video/%s' % video_id
         embed_page = self._download_webpage(embed_url, video_id,
                                             u'Downloading embed page')
-        info = self._search_regex(r'var info = ({.*?}),', embed_page, 'video info')
+        info = self._search_regex(r'var info = ({.*?}),$', embed_page,
+            'video info', flags=re.MULTILINE)
         info = json.loads(info)
 
         # TODO: support choosing qualities

youtube_dl/extractor/daum.py

@@ -0,0 +1,74 @@
+# encoding: utf-8
+import re
+import xml.etree.ElementTree
+
+from .common import InfoExtractor
+from ..utils import (
+    compat_urllib_parse,
+    determine_ext,
+)
+
+
+class DaumIE(InfoExtractor):
+    _VALID_URL = r'https?://tvpot\.daum\.net/.*?clipid=(?P<id>\d+)'
+    IE_NAME = u'daum.net'
+
+    _TEST = {
+        u'url': u'http://tvpot.daum.net/clip/ClipView.do?clipid=52554690',
+        u'file': u'52554690.mp4',
+        u'info_dict': {
+            u'title': u'DOTA 2GETHER 시즌2 6회 - 2부',
+            u'description': u'DOTA 2GETHER 시즌2 6회 - 2부',
+            u'upload_date': u'20130831',
+            u'duration': 3868,
+        },
+    }
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        video_id = mobj.group(1)
+        canonical_url = 'http://tvpot.daum.net/v/%s' % video_id
+        webpage = self._download_webpage(canonical_url, video_id)
+        full_id = self._search_regex(r'<link rel="video_src" href=".+?vid=(.+?)"',
+            webpage, u'full id')
+        query = compat_urllib_parse.urlencode({'vid': full_id})
+        info_xml = self._download_webpage(
+            'http://tvpot.daum.net/clip/ClipInfoXml.do?' + query, video_id,
+            u'Downloading video info')
+        urls_xml = self._download_webpage(
+            'http://videofarm.daum.net/controller/api/open/v1_2/MovieData.apixml?' + query,
+            video_id, u'Downloading video formats info')
+        info = xml.etree.ElementTree.fromstring(info_xml.encode('utf-8'))
+        urls = xml.etree.ElementTree.fromstring(urls_xml.encode('utf-8'))
+
+        self.to_screen(u'%s: Getting video urls' % video_id)
+        formats = []
+        for format_el in urls.findall('result/output_list/output_list'):
+            profile = format_el.attrib['profile']
+            format_query = compat_urllib_parse.urlencode({
+                'vid': full_id,
+                'profile': profile,
+            })
+            url_xml = self._download_webpage(
+                'http://videofarm.daum.net/controller/api/open/v1_2/MovieLocation.apixml?' + format_query,
+                video_id, note=False)
+            url_doc = xml.etree.ElementTree.fromstring(url_xml.encode('utf-8'))
+            format_url = url_doc.find('result/url').text
+            formats.append({
+                'url': format_url,
+                'ext': determine_ext(format_url),
+                'format_id': profile,
+            })
+
+        info = {
+            'id': video_id,
+            'title': info.find('TITLE').text,
+            'formats': formats,
+            'thumbnail': self._og_search_thumbnail(webpage),
+            'description': info.find('CONTENTS').text,
+            'duration': int(info.find('DURATION').text),
+            'upload_date': info.find('REGDTTM').text[:8],
+        }
+        # TODO: Remove when #980 has been merged
+        info.update(formats[-1])
+        return info

youtube_dl/extractor/defense.py

@@ -0,0 +1,39 @@
+import re
+import json
+
+from .common import InfoExtractor
+
+
+class DefenseGouvFrIE(InfoExtractor):
+    _IE_NAME = 'defense.gouv.fr'
+    _VALID_URL = (r'http://.*?\.defense\.gouv\.fr/layout/set/'
+        r'ligthboxvideo/base-de-medias/webtv/(.*)')
+
+    _TEST = {
+        u'url': (u'http://www.defense.gouv.fr/layout/set/ligthboxvideo/'
+        u'base-de-medias/webtv/attaque-chimique-syrienne-du-21-aout-2013-1'),
+        u'file': u'11213.mp4',
+        u'md5': u'75bba6124da7e63d2d60b5244ec9430c',
+        "info_dict": {
+            "title": "attaque-chimique-syrienne-du-21-aout-2013-1"
+        }
+    }
+
+    def _real_extract(self, url):
+        title = re.match(self._VALID_URL, url).group(1)
+        webpage = self._download_webpage(url, title)
+        video_id = self._search_regex(
+            r"flashvars.pvg_id=\"(\d+)\";",
+            webpage, 'ID')
+        
+        json_url = ('http://static.videos.gouv.fr/brightcovehub/export/json/'
+            + video_id)
+        info = self._download_webpage(json_url, title,
+                                                  'Downloading JSON config')
+        video_url = json.loads(info)['renditions'][0]['url']
+        
+        return {'id': video_id,
+                'ext': 'mp4',
+                'url': video_url,
+                'title': title,
+                }

youtube_dl/extractor/generic.py

@@ -153,7 +153,7 @@ class GenericIE(InfoExtractor):
                 mobj = re.search(r'<meta.*?property="og:video".*?content="(.*?)"', webpage)
         if mobj is None:
             # HTML5 video
-            mobj = re.search(r'<video[^<]*>.*?<source .*?src="([^"]+)"', webpage, flags=re.DOTALL)
+            mobj = re.search(r'<video[^<]*(?:>.*?<source.*?)? src="([^"]+)"', webpage, flags=re.DOTALL)
         if mobj is None:
             raise ExtractorError(u'Invalid URL: %s' % url)
 
@@ -162,9 +162,9 @@ class GenericIE(InfoExtractor):
         if mobj.group(1) is None:
             raise ExtractorError(u'Invalid URL: %s' % url)
 
-        video_url = compat_urllib_parse.unquote(mobj.group(1))
+        video_url = mobj.group(1)
         video_url = compat_urlparse.urljoin(url, video_url)
-        video_id = os.path.basename(video_url)
+        video_id = compat_urllib_parse.unquote(os.path.basename(video_url))
 
         # here's a fun little line of code for you:
         video_extension = os.path.splitext(video_id)[1][1:]

youtube_dl/extractor/ign.py

@@ -13,7 +13,7 @@ class IGNIE(InfoExtractor):
     Some videos of it.ign.com are also supported
     """
 
-    _VALID_URL = r'https?://.+?\.ign\.com/(?:videos|show_videos)(/.+)?/(?P<name_or_id>.+)'
+    _VALID_URL = r'https?://.+?\.ign\.com/(?P<type>videos|show_videos|articles)(/.+)?/(?P<name_or_id>.+)'
     IE_NAME = u'ign.com'
 
     _CONFIG_URL_TEMPLATE = 'http://www.ign.com/videos/configs/id/%s.config'
@@ -41,7 +41,11 @@ class IGNIE(InfoExtractor):
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
         name_or_id = mobj.group('name_or_id')
+        page_type = mobj.group('type')
         webpage = self._download_webpage(url, name_or_id)
+        if page_type == 'articles':
+            video_url = self._search_regex(r'var videoUrl = "(.+?)"', webpage, u'video url')
+            return self.url_result(video_url, ie='IGN')
         video_id = self._find_video_id(webpage)
         result = self._get_video_info(video_id)
         description = self._html_search_regex(self._DESCRIPTION_RE,
@@ -68,7 +72,7 @@ class IGNIE(InfoExtractor):
 class OneUPIE(IGNIE):
     """Extractor for 1up.com, it uses the ign videos system."""
 
-    _VALID_URL = r'https?://gamevideos.1up.com/video/id/(?P<name_or_id>.+)'
+    _VALID_URL = r'https?://gamevideos.1up.com/(?P<type>video)/id/(?P<name_or_id>.+)'
     IE_NAME = '1up.com'
 
     _DESCRIPTION_RE = r'<div id="vid_summary">(.+?)</div>'

youtube_dl/extractor/metacafe.py

@@ -122,7 +122,7 @@ class MetacafeIE(InfoExtractor):
         video_title = self._html_search_regex(r'(?im)<title>(.*) - Video</title>', webpage, u'title')
         description = self._og_search_description(webpage)
         video_uploader = self._html_search_regex(
-                r'submitter=(.*?);|googletag\.pubads\(\)\.setTargeting\("channel","([^"]+)"\);',
+                r'submitter=(.*?);|googletag\.pubads\(\)\.setTargeting\("(?:channel|submiter)","([^"]+)"\);',
                 webpage, u'uploader nickname', fatal=False)
 
         return {

youtube_dl/extractor/mit.py

@@ -25,23 +25,21 @@ class TechTVMITIE(InfoExtractor):
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
         video_id = mobj.group('id')
-        webpage = self._download_webpage(
+        raw_page = self._download_webpage(
             'http://techtv.mit.edu/videos/%s' % video_id, video_id)
-        embed_page = self._download_webpage(
-            'http://techtv.mit.edu/embeds/%s/' % video_id, video_id,
-            note=u'Downloading embed page')
+        clean_page = re.compile(u'<!--.*?-->', re.S).sub(u'', raw_page)
 
         base_url = self._search_regex(r'ipadUrl: \'(.+?cloudfront.net/)',
-            embed_page, u'base url')
-        formats_json = self._search_regex(r'bitrates: (\[.+?\])', embed_page,
+            raw_page, u'base url')
+        formats_json = self._search_regex(r'bitrates: (\[.+?\])', raw_page,
             u'video formats')
         formats = json.loads(formats_json)
         formats = sorted(formats, key=lambda f: f['bitrate'])
 
-        title = get_element_by_id('edit-title', webpage)
-        description = clean_html(get_element_by_id('edit-description', webpage))
+        title = get_element_by_id('edit-title', clean_page)
+        description = clean_html(get_element_by_id('edit-description', clean_page))
         thumbnail = self._search_regex(r'playlist:.*?url: \'(.+?)\'',
-            embed_page, u'thumbnail', flags=re.DOTALL)
+            raw_page, u'thumbnail', flags=re.DOTALL)
 
         return {'id': video_id,
                 'title': title,

youtube_dl/extractor/naver.py

@@ -0,0 +1,73 @@
+# encoding: utf-8
+import re
+import xml.etree.ElementTree
+
+from .common import InfoExtractor
+from ..utils import (
+    compat_urllib_parse,
+    ExtractorError,
+)
+
+
+class NaverIE(InfoExtractor):
+    _VALID_URL = r'https?://tvcast\.naver\.com/v/(?P<id>\d+)'
+
+    _TEST = {
+        u'url': u'http://tvcast.naver.com/v/81652',
+        u'file': u'81652.mp4',
+        u'info_dict': {
+            u'title': u'[9월 모의고사 해설강의][수학_김상희] 수학 A형 16~20번',
+            u'description': u'합격불변의 법칙 메가스터디 | 메가스터디 수학 김상희 선생님이 9월 모의고사 수학A형 16번에서 20번까지 해설강의를 공개합니다.',
+            u'upload_date': u'20130903',
+        },
+    }
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        video_id = mobj.group(1)
+        webpage = self._download_webpage(url, video_id)
+        m_id = re.search(r'var rmcPlayer = new nhn.rmcnmv.RMCVideoPlayer\("(.+?)", "(.+?)"',
+            webpage)
+        if m_id is None:
+            raise ExtractorError(u'couldn\'t extract vid and key')
+        vid = m_id.group(1)
+        key = m_id.group(2)
+        query = compat_urllib_parse.urlencode({'vid': vid, 'inKey': key,})
+        query_urls = compat_urllib_parse.urlencode({
+            'masterVid': vid,
+            'protocol': 'p2p',
+            'inKey': key,
+        })
+        info_xml = self._download_webpage(
+            'http://serviceapi.rmcnmv.naver.com/flash/videoInfo.nhn?' + query,
+            video_id, u'Downloading video info')
+        urls_xml = self._download_webpage(
+            'http://serviceapi.rmcnmv.naver.com/flash/playableEncodingOption.nhn?' + query_urls,
+            video_id, u'Downloading video formats info')
+        info = xml.etree.ElementTree.fromstring(info_xml.encode('utf-8'))
+        urls = xml.etree.ElementTree.fromstring(urls_xml.encode('utf-8'))
+
+        formats = []
+        for format_el in urls.findall('EncodingOptions/EncodingOption'):
+            domain = format_el.find('Domain').text
+            if domain.startswith('rtmp'):
+                continue
+            formats.append({
+                'url': domain + format_el.find('uri').text,
+                'ext': 'mp4',
+                'width': int(format_el.find('width').text),
+                'height': int(format_el.find('height').text),
+            })
+
+        info = {
+            'id': video_id,
+            'title': info.find('Subject').text,
+            'formats': formats,
+            'description': self._og_search_description(webpage),
+            'thumbnail': self._og_search_thumbnail(webpage),
+            'upload_date': info.find('WriteDate').text.replace('.', ''),
+            'view_count': int(info.find('PlayCount').text),
+        }
+        # TODO: Remove when #980 has been merged
+        info.update(formats[-1])
+        return info

youtube_dl/extractor/orf.py

@@ -0,0 +1,54 @@
+# coding: utf-8
+
+import re
+import xml.etree.ElementTree
+import json
+
+from .common import InfoExtractor
+from ..utils import (
+    compat_urlparse,
+    ExtractorError,
+    find_xpath_attr,
+)
+
+class ORFIE(InfoExtractor):
+    _VALID_URL = r'https?://tvthek.orf.at/(programs/.+?/episodes|topics/.+?)/(?P<id>\d+)'
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        playlist_id = mobj.group('id')
+        webpage = self._download_webpage(url, playlist_id)
+
+        flash_xml = self._search_regex('ORF.flashXML = \'(.+?)\'', webpage, u'flash xml')
+        flash_xml = compat_urlparse.parse_qs('xml='+flash_xml)['xml'][0]
+        flash_config = xml.etree.ElementTree.fromstring(flash_xml.encode('utf-8'))
+        playlist_json = self._search_regex(r'playlist\': \'(\[.*?\])\'', webpage, u'playlist').replace(r'\"','"')
+        playlist = json.loads(playlist_json)
+
+        videos = []
+        ns = '{http://tempuri.org/XMLSchema.xsd}'
+        xpath = '%(ns)sPlaylist/%(ns)sItems/%(ns)sItem' % {'ns': ns}
+        webpage_description = self._og_search_description(webpage)
+        for (i, (item, info)) in enumerate(zip(flash_config.findall(xpath), playlist), 1):
+            # Get best quality url
+            rtmp_url = None
+            for q in ['Q6A', 'Q4A', 'Q1A']:
+                video_url = find_xpath_attr(item, '%sVideoUrl' % ns, 'quality', q)
+                if video_url is not None:
+                    rtmp_url = video_url.text
+                    break
+            if rtmp_url is None:
+                raise ExtractorError(u'Couldn\'t get video url: %s' % info['id'])
+            description = self._html_search_regex(
+                r'id="playlist_entry_%s".*?<p>(.*?)</p>' % i, webpage,
+                u'description', default=webpage_description, flags=re.DOTALL)
+            videos.append({
+                '_type': 'video',
+                'id': info['id'],
+                'title': info['title'],
+                'url': rtmp_url,
+                'ext': 'flv',
+                'description': description,
+                })
+
+        return videos

youtube_dl/extractor/unistra.py

@@ -11,7 +11,7 @@ class UnistraIE(InfoExtractor):
         u'md5': u'736f605cfdc96724d55bb543ab3ced24',
         u'info_dict': {
             u'title': u'M!ss Yella',
-            u'description': u'md5:75e8439a3e2981cd5d4b6db232e8fdfc',
+            u'description': u'md5:104892c71bd48e55d70b902736b81bbf',
         },
     }
 

youtube_dl/extractor/veehd.py

@@ -0,0 +1,56 @@
+import re
+import json
+
+from .common import InfoExtractor
+from ..utils import (
+    compat_urlparse,
+    get_element_by_id,
+    clean_html,
+)
+
+class VeeHDIE(InfoExtractor):
+    _VALID_URL = r'https?://veehd.com/video/(?P<id>\d+)'
+
+    _TEST = {
+        u'url': u'http://veehd.com/video/4686958',
+        u'file': u'4686958.mp4',
+        u'info_dict': {
+            u'title': u'Time Lapse View from Space ( ISS)',
+            u'uploader_id': u'spotted',
+            u'description': u'md5:f0094c4cf3a72e22bc4e4239ef767ad7',
+        },
+    }
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        video_id = mobj.group('id')
+
+        webpage = self._download_webpage(url, video_id)
+        player_path = self._search_regex(r'\$\("#playeriframe"\).attr\({src : "(.+?)"',
+            webpage, u'player path')
+        player_url = compat_urlparse.urljoin(url, player_path)
+        player_page = self._download_webpage(player_url, video_id,
+            u'Downloading player page')
+        config_json = self._search_regex(r'value=\'config=({.+?})\'',
+            player_page, u'config json')
+        config = json.loads(config_json)
+
+        video_url = compat_urlparse.unquote(config['clip']['url'])
+        title = clean_html(get_element_by_id('videoName', webpage).rpartition('|')[0])
+        uploader_id = self._html_search_regex(r'<a href="/profile/\d+">(.+?)</a>',
+            webpage, u'uploader')
+        thumbnail = self._search_regex(r'<img id="veehdpreview" src="(.+?)"',
+            webpage, u'thumbnail')
+        description = self._html_search_regex(r'<td class="infodropdown".*?<div>(.*?)<ul',
+            webpage, u'description', flags=re.DOTALL)
+
+        return {
+            '_type': 'video',
+            'id': video_id,
+            'title': title,
+            'url': video_url,
+            'ext': 'mp4',
+            'uploader_id': uploader_id,
+            'thumbnail': thumbnail,
+            'description': description,
+        }

youtube_dl/extractor/vimeo.py

@@ -44,6 +44,16 @@ class VimeoIE(InfoExtractor):
                 u'title': u'Andy Allan - Putting the Carto into OpenStreetMap Cartography',
             },
         },
+        {
+            u'url': u'http://player.vimeo.com/video/54469442',
+            u'file': u'54469442.mp4',
+            u'md5': u'619b811a4417aa4abe78dc653becf511',
+            u'note': u'Videos that embed the url in the player page',
+            u'info_dict': {
+                u'title': u'Kathy Sierra: Building the minimum Badass User, Business of Software',
+                u'uploader': u'The BLN & Business of Software',
+            },
+        },
     ]
 
     def _login(self):
@@ -112,7 +122,8 @@ class VimeoIE(InfoExtractor):
 
         # Extract the config JSON
         try:
-            config = webpage.split(' = {config:')[1].split(',assets:')[0]
+            config = self._search_regex([r' = {config:({.+?}),assets:', r'c=({.+?);'],
+                webpage, u'info section', flags=re.DOTALL)
             config = json.loads(config)
         except:
             if re.search('The creator of this video has not given you permission to embed it on this domain.', webpage):
@@ -132,12 +143,22 @@ class VimeoIE(InfoExtractor):
         video_uploader_id = config["video"]["owner"]["url"].split('/')[-1] if config["video"]["owner"]["url"] else None
 
         # Extract video thumbnail
-        video_thumbnail = config["video"]["thumbnail"]
+        video_thumbnail = config["video"].get("thumbnail")
+        if video_thumbnail is None:
+            _, video_thumbnail = sorted((int(width), t_url) for (width, t_url) in config["video"]["thumbs"].items())[-1]
 
         # Extract video description
-        video_description = get_element_by_attribute("itemprop", "description", webpage)
-        if video_description: video_description = clean_html(video_description)
-        else: video_description = u''
+        video_description = None
+        try:
+            video_description = get_element_by_attribute("itemprop", "description", webpage)
+            if video_description: video_description = clean_html(video_description)
+        except AssertionError as err:
+            # On some pages like (http://player.vimeo.com/video/54469442) the
+            # html tags are not closed, python 2.6 cannot handle it
+            if err.args[0] == 'we should not get here!':
+                pass
+            else:
+                raise
 
         # Extract upload date
         video_upload_date = None
@@ -154,14 +175,15 @@ class VimeoIE(InfoExtractor):
         # TODO bind to format param
         codecs = [('h264', 'mp4'), ('vp8', 'flv'), ('vp6', 'flv')]
         files = { 'hd': [], 'sd': [], 'other': []}
+        config_files = config["video"].get("files") or config["request"].get("files")
         for codec_name, codec_extension in codecs:
-            if codec_name in config["video"]["files"]:
-                if 'hd' in config["video"]["files"][codec_name]:
+            if codec_name in config_files:
+                if 'hd' in config_files[codec_name]:
                     files['hd'].append((codec_name, codec_extension, 'hd'))
-                elif 'sd' in config["video"]["files"][codec_name]:
+                elif 'sd' in config_files[codec_name]:
                     files['sd'].append((codec_name, codec_extension, 'sd'))
                 else:
-                    files['other'].append((codec_name, codec_extension, config["video"]["files"][codec_name][0]))
+                    files['other'].append((codec_name, codec_extension, config_files[codec_name][0]))
 
         for quality in ('hd', 'sd', 'other'):
             if len(files[quality]) > 0:
@@ -173,8 +195,12 @@ class VimeoIE(InfoExtractor):
         else:
             raise ExtractorError(u'No known codec found')
 
-        video_url = "http://player.vimeo.com/play_redirect?clip_id=%s&sig=%s&time=%s&quality=%s&codecs=%s&type=moogaloop_local&embed_location=" \
-                    %(video_id, sig, timestamp, video_quality, video_codec.upper())
+        video_url = None
+        if isinstance(config_files[video_codec], dict):
+            video_url = config_files[video_codec][video_quality].get("url")
+        if video_url is None:
+            video_url = "http://player.vimeo.com/play_redirect?clip_id=%s&sig=%s&time=%s&quality=%s&codecs=%s&type=moogaloop_local&embed_location=" \
+                        %(video_id, sig, timestamp, video_quality, video_codec.upper())
 
         return [{
             'id':       video_id,

youtube_dl/extractor/youtube.py

@@ -153,8 +153,9 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
                      $"""
     _NEXT_URL_RE = r'[\?&]next_url=([^&]+)'
     # Listed in order of quality
-    _available_formats = ['38', '37', '46', '22', '45', '35', '44', '34', '18', '43', '6', '5', '17', '13',
-                          '95', '94', '93', '92', '132', '151',
+    _available_formats = ['38', '37', '46', '22', '45', '35', '44', '34', '18', '43', '6', '5', '36', '17', '13',
+                          # Apple HTTP Live Streaming
+                          '96', '95', '94', '93', '92', '132', '151',
                           # 3D
                           '85', '84', '102', '83', '101', '82', '100',
                           # Dash video
@@ -163,8 +164,10 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
                           # Dash audio
                           '141', '172', '140', '171', '139',
                           ]
-    _available_formats_prefer_free = ['38', '46', '37', '45', '22', '44', '35', '43', '34', '18', '6', '5', '17', '13',
-                                      '95', '94', '93', '92', '132', '151',
+    _available_formats_prefer_free = ['38', '46', '37', '45', '22', '44', '35', '43', '34', '18', '6', '5', '36', '17', '13',
+                                      # Apple HTTP Live Streaming
+                                      '96', '95', '94', '93', '92', '132', '151',
+                                      # 3D
                                       '85', '102', '84', '101', '83', '100', '82',
                                       # Dash video
                                       '138', '248', '137', '247', '136', '246', '245',
@@ -172,11 +175,18 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
                                       # Dash audio
                                       '172', '141', '171', '140', '139',
                                       ]
+    _video_formats_map = {
+        'flv': ['35', '34', '6', '5'],
+        '3gp': ['36', '17', '13'],
+        'mp4': ['38', '37', '22', '18'],
+        'webm': ['46', '45', '44', '43'],
+    }
     _video_extensions = {
         '13': '3gp',
-        '17': 'mp4',
+        '17': '3gp',
         '18': 'mp4',
         '22': 'mp4',
+        '36': '3gp',
         '37': 'mp4',
         '38': 'mp4',
         '43': 'webm',
@@ -193,7 +203,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
         '101': 'webm',
         '102': 'webm',
 
-        # videos that use m3u8
+        # Apple HTTP Live Streaming
         '92': 'mp4',
         '93': 'mp4',
         '94': 'mp4',
@@ -234,6 +244,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
         '22': '720x1280',
         '34': '360x640',
         '35': '480x854',
+        '36': '240x320',
         '37': '1080x1920',
         '38': '3072x4096',
         '43': '360x640',
@@ -335,7 +346,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
             u"info_dict": {
                 u"upload_date": u"20120506",
                 u"title": u"Icona Pop - I Love It (feat. Charli XCX) [OFFICIAL VIDEO]",
-                u"description": u"md5:b085c9804f5ab69f4adea963a2dceb3c",
+                u"description": u"md5:3e2666e0a55044490499ea45fe9037b7",
                 u"uploader": u"Icona Pop",
                 u"uploader_id": u"IconaPop"
             }
@@ -423,11 +434,11 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
         elif len(s) == 87:
             return s[6:27] + s[4] + s[28:39] + s[27] + s[40:59] + s[2] + s[60:]
         elif len(s) == 86:
-            return s[5:40] + s[3] + s[41:48] + s[0] + s[49:86]
+            return s[81:73:-1] + s[84] + s[72:58:-1] + s[0] + s[57:35:-1] + s[85] + s[34:0:-1]
         elif len(s) == 85:
             return s[83:34:-1] + s[0] + s[33:27:-1] + s[3] + s[26:19:-1] + s[34] + s[18:3:-1] + s[27]
         elif len(s) == 84:
-            return s[5:40] + s[3] + s[41:48] + s[0] + s[49:84]
+            return s[81:36:-1] + s[0] + s[35:2:-1]
         elif len(s) == 83:
             return s[81:64:-1] + s[82] + s[63:52:-1] + s[45] + s[51:45:-1] + s[1] + s[44:1:-1] + s[0]
         elif len(s) == 82:
@@ -597,13 +608,25 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
             video_url_list = [(f, url_map[f]) for f in existing_formats] # All formats
         else:
             # Specific formats. We pick the first in a slash-delimeted sequence.
-            # For example, if '1/2/3/4' is requested and '2' and '4' are available, we pick '2'.
+            # Format can be specified as itag or 'mp4' or 'flv' etc. We pick the highest quality
+            # available in the specified format. For example,
+            # if '1/2/3/4' is requested and '2' and '4' are available, we pick '2'.
+            # if '1/mp4/3/4' is requested and '1' and '5' (is a mp4) are available, we pick '1'.
+            # if '1/mp4/3/4' is requested and '4' and '5' (is a mp4) are available, we pick '5'.
             req_formats = req_format.split('/')
             video_url_list = None
             for rf in req_formats:
                 if rf in url_map:
                     video_url_list = [(rf, url_map[rf])]
                     break
+                if rf in self._video_formats_map:
+                    for srf in self._video_formats_map[rf]:
+                        if srf in url_map:
+                            video_url_list = [(srf, url_map[srf])]
+                            break
+                    else:
+                        continue
+                    break
             if video_url_list is None:
                 raise ExtractorError(u'requested format not available')
         return video_url_list
@@ -920,8 +943,11 @@ class YoutubePlaylistIE(InfoExtractor):
 
             for entry in response['feed']['entry']:
                 index = entry['yt$position']['$t']
-                if 'media$group' in entry and 'media$player' in entry['media$group']:
-                    videos.append((index, entry['media$group']['media$player']['url']))
+                if 'media$group' in entry and 'yt$videoid' in entry['media$group']:
+                    videos.append((
+                        index,
+                        'https://www.youtube.com/watch?v=' + entry['media$group']['yt$videoid']['$t']
+                    ))
 
         videos = [v[1] for v in sorted(videos)]
 
@@ -1161,7 +1187,7 @@ class YoutubeWatchLaterIE(YoutubeFeedsInfoExtractor):
 class YoutubeFavouritesIE(YoutubeBaseInfoExtractor):
     IE_NAME = u'youtube:favorites'
     IE_DESC = u'YouTube.com favourite videos, "ytfav" keyword (requires authentication)'
-    _VALID_URL = r'https?://www\.youtube\.com/my_favorites|:ytfav(?:o?rites)?'
+    _VALID_URL = r'https?://www\.youtube\.com/my_favorites|:ytfav(?:ou?rites)?'
     _LOGIN_REQUIRED = True
 
     def _real_extract(self, url):

youtube_dl/utils.py

@@ -213,7 +213,7 @@ if sys.version_info >= (2,7):
     def find_xpath_attr(node, xpath, key, val):
         """ Find the xpath xpath[@key=val] """
         assert re.match(r'^[a-zA-Z]+$', key)
-        assert re.match(r'^[a-zA-Z@\s]*$', val)
+        assert re.match(r'^[a-zA-Z0-9@\s]*$', val)
         expr = xpath + u"[@%s='%s']" % (key, val)
         return node.find(expr)
 else:

youtube_dl/version.py

@@ -1,2 +1,2 @@
 
-__version__ = '2013.08.28.1'
+__version__ = '2013.09.05'