release 2014.04.04.6

The re engine does not want to repeat an empty string, for fear that something like

    (.*)*

could be matching the tokens ...

    ""
    "" ""
    "" "" ""

Of course, that's harmless with a question mark, although still somewhat strange.

test/test_all_urls.py

@@ -157,5 +157,11 @@ class TestAllURLsMatching(unittest.TestCase):
             'http://thedailyshow.cc.com/guests/michael-lewis/3efna8/exclusive---michael-lewis-extended-interview-pt--3',
             ['ComedyCentralShows'])
 
+    def test_yahoo_https(self):
+        # https://github.com/rg3/youtube-dl/issues/2701
+        self.assertMatch(
+            'https://screen.yahoo.com/smartwatches-latest-wearable-gadgets-163745379-cbs.html',
+            ['Yahoo'])
+
 if __name__ == '__main__':
     unittest.main()

youtube_dl/__init__.py

@@ -242,7 +242,7 @@ def parseOpts(overrideArguments=None):
         help='Use the specified HTTP/HTTPS proxy. Pass in an empty string (--proxy "") for direct connection')
     general.add_option('--no-check-certificate', action='store_true', dest='no_check_certificate', default=False, help='Suppress HTTPS certificate validation.')
     general.add_option(
-        '--prefer-insecure', action='store_true', dest='prefer_insecure',
+        '--prefer-insecure', '--prefer-unsecure', action='store_true', dest='prefer_insecure',
         help='Use an unencrypted connection to retrieve information about the video. (Currently supported only for YouTube)')
     general.add_option(
         '--cache-dir', dest='cachedir', default=get_cachedir(), metavar='DIR',

youtube_dl/downloader/common.py

@@ -4,9 +4,10 @@ import sys
 import time
 
 from ..utils import (
+    compat_str,
     encodeFilename,
-    timeconvert,
     format_bytes,
+    timeconvert,
 )
 
 
@@ -173,7 +174,7 @@ class FileDownloader(object):
                 return
             os.rename(encodeFilename(old_filename), encodeFilename(new_filename))
         except (IOError, OSError) as err:
-            self.report_error(u'unable to rename file: %s' % str(err))
+            self.report_error(u'unable to rename file: %s' % compat_str(err))
 
     def try_utime(self, filename, last_modified_hdr):
         """Try to set the last-modified time of the given file."""

youtube_dl/extractor/breakcom.py

@@ -27,9 +27,10 @@ class BreakIE(InfoExtractor):
             webpage, 'info json', flags=re.DOTALL)
         info = json.loads(info_json)
         video_url = info['videoUri']
-        m_youtube = re.search(r'(https?://www\.youtube\.com/watch\?v=.*)', video_url)
-        if m_youtube is not None:
-            return self.url_result(m_youtube.group(1), 'Youtube')
+        youtube_id = info.get('youtubeId')
+        if youtube_id:
+            return self.url_result(youtube_id, 'Youtube')
+
         final_url = video_url + '?' + info['AuthToken']
         return {
             'id': video_id,

youtube_dl/extractor/dailymotion.py

@@ -8,7 +8,6 @@ from .subtitles import SubtitlesInfoExtractor
 from ..utils import (
     compat_urllib_request,
     compat_str,
-    get_element_by_attribute,
     get_element_by_id,
     orderedSet,
     str_to_int,

youtube_dl/extractor/keezmovies.py

@@ -1,3 +1,5 @@
+from __future__ import unicode_literals
+
 import os
 import re
 
@@ -11,22 +13,22 @@ from ..aes import (
     aes_decrypt_text
 )
 
+
 class KeezMoviesIE(InfoExtractor):
-    _VALID_URL = r'^(?:https?://)?(?:www\.)?(?P<url>keezmovies\.com/video/.+?(?P<videoid>[0-9]+))(?:[/?&]|$)'
+    _VALID_URL = r'^https?://(?:www\.)?keezmovies\.com/video/.+?(?P<videoid>[0-9]+)(?:[/?&]|$)'
     _TEST = {
-        u'url': u'http://www.keezmovies.com/video/petite-asian-lady-mai-playing-in-bathtub-1214711',
-        u'file': u'1214711.mp4',
-        u'md5': u'6e297b7e789329923fcf83abb67c9289',
-        u'info_dict': {
-            u"title": u"Petite Asian Lady Mai Playing In Bathtub",
-            u"age_limit": 18,
+        'url': 'http://www.keezmovies.com/video/petite-asian-lady-mai-playing-in-bathtub-1214711',
+        'file': '1214711.mp4',
+        'md5': '6e297b7e789329923fcf83abb67c9289',
+        'info_dict': {
+            'title': 'Petite Asian Lady Mai Playing In Bathtub',
+            'age_limit': 18,
         }
     }
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
         video_id = mobj.group('videoid')
-        url = 'http://www.' + mobj.group('url')
 
         req = compat_urllib_request.Request(url)
         req.add_header('Cookie', 'age_verified=1')
@@ -38,10 +40,10 @@ class KeezMoviesIE(InfoExtractor):
             embedded_url = mobj.group(1)
             return self.url_result(embedded_url)
 
-        video_title = self._html_search_regex(r'<h1 [^>]*>([^<]+)', webpage, u'title')
-        video_url = compat_urllib_parse.unquote(self._html_search_regex(r'video_url=(.+?)&amp;', webpage, u'video_url'))
-        if webpage.find('encrypted=true')!=-1:
-            password = self._html_search_regex(r'video_title=(.+?)&amp;', webpage, u'password')
+        video_title = self._html_search_regex(r'<h1 [^>]*>([^<]+)', webpage, 'title')
+        video_url = compat_urllib_parse.unquote(self._html_search_regex(r'video_url=(.+?)&amp;', webpage, 'video_url'))
+        if 'encrypted=true' in webpage:
+            password = self._html_search_regex(r'video_title=(.+?)&amp;', webpage, 'password')
             video_url = aes_decrypt_text(video_url, password, 32).decode('utf-8')
         path = compat_urllib_parse_urlparse(video_url).path
         extension = os.path.splitext(path)[1][1:]

youtube_dl/extractor/morningstar.py

@@ -1,17 +1,9 @@
 # coding: utf-8
 from __future__ import unicode_literals
 
-import hashlib
-import json
 import re
-import time
 
 from .common import InfoExtractor
-from ..utils import (
-    compat_parse_qs,
-    compat_str,
-    int_or_none,
-)
 
 
 class MorningstarIE(InfoExtractor):

youtube_dl/extractor/pornhd.py

@@ -1,44 +1,81 @@
 from __future__ import unicode_literals
 
 import re
+import json
 
 from .common import InfoExtractor
-from ..utils import compat_urllib_parse
+from ..utils import int_or_none
 
 
 class PornHdIE(InfoExtractor):
-    _VALID_URL = r'(?:http://)?(?:www\.)?pornhd\.com/(?:[a-z]{2,4}/)?videos/(?P<video_id>[0-9]+)/(?P<video_title>.+)'
+    _VALID_URL = r'http://(?:www\.)?pornhd\.com/(?:[a-z]{2,4}/)?videos/(?P<id>\d+)'
     _TEST = {
         'url': 'http://www.pornhd.com/videos/1962/sierra-day-gets-his-cum-all-over-herself-hd-porn-video',
-        'file': '1962.flv',
-        'md5': '35272469887dca97abd30abecc6cdf75',
+        'md5': '956b8ca569f7f4d8ec563e2c41598441',
         'info_dict': {
-            "title": "sierra-day-gets-his-cum-all-over-herself-hd-porn-video",
-            "age_limit": 18,
+            'id': '1962',
+            'ext': 'mp4',
+            'title': 'Sierra loves doing laundry',
+            'description': 'md5:8ff0523848ac2b8f9b065ba781ccf294',
+            'age_limit': 18,
         }
     }
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
-
-        video_id = mobj.group('video_id')
-        video_title = mobj.group('video_title')
+        video_id = mobj.group('id')
 
         webpage = self._download_webpage(url, video_id)
 
-        next_url = self._html_search_regex(
-            r'&hd=(http.+?)&', webpage, 'video URL')
-        next_url = compat_urllib_parse.unquote(next_url)
+        title = self._og_search_title(webpage)
+        TITLE_SUFFIX = ' porn HD Video | PornHD.com '
+        if title.endswith(TITLE_SUFFIX):
+            title = title[:-len(TITLE_SUFFIX)]
 
-        video_url = self._download_webpage(
-            next_url, video_id, note='Retrieving video URL',
-            errnote='Could not retrieve video URL')
-        age_limit = 18
+        description = self._html_search_regex(
+            r'<div class="description">([^<]+)</div>', webpage, 'description', fatal=False)
+        view_count = int_or_none(self._html_search_regex(
+            r'(\d+) views 	</span>', webpage, 'view count', fatal=False))
+
+        formats = [
+            {
+                'url': format_url,
+                'ext': format.lower(),
+                'format_id': '%s-%s' % (format.lower(), quality.lower()),
+                'quality': 1 if quality.lower() == 'high' else 0,
+            } for format, quality, format_url in re.findall(
+                r'var __video([\da-zA-Z]+?)(Low|High)StreamUrl = \'(http://.+?)\?noProxy=1\'', webpage)
+        ]
+
+        mobj = re.search(r'flashVars = (?P<flashvars>{.+?});', webpage)
+        if mobj:
+            flashvars = json.loads(mobj.group('flashvars'))
+            formats.extend([
+                {
+                    'url': flashvars['hashlink'].replace('?noProxy=1', ''),
+                    'ext': 'flv',
+                    'format_id': 'flv-low',
+                    'quality': 0,
+                },
+                {
+                    'url': flashvars['hd'].replace('?noProxy=1', ''),
+                    'ext': 'flv',
+                    'format_id': 'flv-high',
+                    'quality': 1,
+                }
+            ])
+            thumbnail = flashvars['urlWallpaper']
+        else:
+            thumbnail = self._og_search_thumbnail(webpage)
+
+        self._sort_formats(formats)
 
         return {
             'id': video_id,
-            'url': video_url,
-            'ext': 'flv',
-            'title': video_title,
-            'age_limit': age_limit,
+            'title': title,
+            'description': description,
+            'thumbnail': thumbnail,
+            'view_count': view_count,
+            'formats': formats,
+            'age_limit': 18,
         }

youtube_dl/extractor/rts.py

@@ -9,46 +9,133 @@ from ..utils import (
     parse_duration,
     parse_iso8601,
     unescapeHTML,
+    compat_str,
 )
 
 
 class RTSIE(InfoExtractor):
     IE_DESC = 'RTS.ch'
-    _VALID_URL = r'^https?://(?:www\.)?rts\.ch/archives/tv/[^/]+/(?P<id>[0-9]+)-.*?\.html'
+    _VALID_URL = r'^https?://(?:www\.)?rts\.ch/(?:[^/]+/){2,}(?P<id>[0-9]+)-.*?\.html'
 
-    _TEST = {
-        'url': 'http://www.rts.ch/archives/tv/divers/3449373-les-enfants-terribles.html',
-        'md5': '753b877968ad8afaeddccc374d4256a5',
-        'info_dict': {
-            'id': '3449373',
-            'ext': 'mp4',
-            'duration': 1488,
-            'title': 'Les Enfants Terribles',
-            'description': 'France Pommier et sa soeur Luce Feral, les deux filles de ce groupe de 5.',
-            'uploader': 'Divers',
-            'upload_date': '19680921',
-            'timestamp': -40280400,
-            'thumbnail': 're:^https?://.*\.image'
+    _TESTS = [
+        {
+            'url': 'http://www.rts.ch/archives/tv/divers/3449373-les-enfants-terribles.html',
+            'md5': '753b877968ad8afaeddccc374d4256a5',
+            'info_dict': {
+                'id': '3449373',
+                'ext': 'mp4',
+                'duration': 1488,
+                'title': 'Les Enfants Terribles',
+                'description': 'France Pommier et sa soeur Luce Feral, les deux filles de ce groupe de 5.',
+                'uploader': 'Divers',
+                'upload_date': '19680921',
+                'timestamp': -40280400,
+                'thumbnail': 're:^https?://.*\.image'
+            },
         },
-    }
+        {
+            'url': 'http://www.rts.ch/emissions/passe-moi-les-jumelles/5624067-entre-ciel-et-mer.html',
+            'md5': 'c197f0b2421995c63a64cc73d800f42e',
+            'info_dict': {
+                'id': '5738317',
+                'ext': 'mp4',
+                'duration': 55,
+                'title': 'Bande de lancement de Passe-moi les jumelles',
+                'description': '',
+                'uploader': 'Passe-moi les jumelles',
+                'upload_date': '20140404',
+                'timestamp': 1396635300,
+                'thumbnail': 're:^https?://.*\.image'
+            },
+        },
+        {
+            'url': 'http://www.rts.ch/video/sport/hockey/5745975-1-2-kloten-fribourg-5-2-second-but-pour-gotteron-par-kwiatowski.html',
+            'md5': 'b4326fecd3eb64a458ba73c73e91299d',
+            'info_dict': {
+                'id': '5745975',
+                'ext': 'mp4',
+                'duration': 48,
+                'title': '1/2, Kloten - Fribourg (5-2): second but pour Gottéron par Kwiatowski',
+                'description': 'Hockey - Playoff',
+                'uploader': 'Hockey',
+                'upload_date': '20140403',
+                'timestamp': 1396556882,
+                'thumbnail': 're:^https?://.*\.image'
+            },
+            'skip': 'Blocked outside Switzerland',
+        },
+        {
+            'url': 'http://www.rts.ch/video/info/journal-continu/5745356-londres-cachee-par-un-epais-smog.html',
+            'md5': '9bb06503773c07ce83d3cbd793cebb91',
+            'info_dict': {
+                'id': '5745356',
+                'ext': 'mp4',
+                'duration': 33,
+                'title': 'Londres cachée par un épais smog',
+                'description': 'Un important voile de smog recouvre Londres depuis mercredi, provoqué par la pollution et du sable du Sahara.',
+                'uploader': 'Le Journal en continu',
+                'upload_date': '20140403',
+                'timestamp': 1396537322,
+                'thumbnail': 're:^https?://.*\.image'
+            },
+        },
+        {
+            'url': 'http://www.rts.ch/audio/couleur3/programmes/la-belle-video-de-stephane-laurenceau/5706148-urban-hippie-de-damien-krisl-03-04-2014.html',
+            'md5': 'dd8ef6a22dff163d063e2a52bc8adcae',
+            'info_dict': {
+                'id': '5706148',
+                'ext': 'mp3',
+                'duration': 123,
+                'title': '"Urban Hippie", de Damien Krisl',
+                'description': 'Des Hippies super glam.',
+                'upload_date': '20140403',
+                'timestamp': 1396551600,
+            },
+        },
+    ]
 
     def _real_extract(self, url):
         m = re.match(self._VALID_URL, url)
         video_id = m.group('id')
 
-        all_info = self._download_json(
-            'http://www.rts.ch/a/%s.html?f=json/article' % video_id, video_id)
-        info = all_info['video']['JSONinfo']
+        def download_json(video_id):
+            return self._download_json(
+                'http://www.rts.ch/a/%s.html?f=json/article' % video_id, video_id)
+
+        all_info = download_json(video_id)
+
+        # video_id extracted out of URL is not always a real id
+        if 'video' not in all_info and 'audio' not in all_info:
+            page = self._download_webpage(url, video_id)
+            video_id = self._html_search_regex(r'<(?:video|audio) data-id="(\d+)"', page, 'video id')
+            all_info = download_json(video_id)
+
+        info = all_info['video']['JSONinfo'] if 'video' in all_info else all_info['audio']
 
         upload_timestamp = parse_iso8601(info.get('broadcast_date'))
-        duration = parse_duration(info.get('duration'))
+        duration = info.get('duration') or info.get('cutout') or info.get('cutduration')
+        if isinstance(duration, compat_str):
+            duration = parse_duration(duration)
+        view_count = info.get('plays')
         thumbnail = unescapeHTML(info.get('preview_image_url'))
+
+        def extract_bitrate(url):
+            return int_or_none(self._search_regex(
+                r'-([0-9]+)k\.', url, 'bitrate', default=None))
+
         formats = [{
             'format_id': fid,
             'url': furl,
-            'tbr': int_or_none(self._search_regex(
-                r'-([0-9]+)k\.', furl, 'bitrate', default=None)),
+            'tbr': extract_bitrate(furl),
         } for fid, furl in info['streams'].items()]
+
+        if 'media' in info:
+            formats.extend([{
+                'format_id': '%s-%sk' % (media['ext'], media['rate']),
+                'url': 'http://download-video.rts.ch/%s' % media['url'],
+                'tbr': media['rate'] or extract_bitrate(media['url']),
+            } for media in info['media'] if media.get('rate')])
+
         self._sort_formats(formats)
 
         return {
@@ -57,6 +144,7 @@ class RTSIE(InfoExtractor):
             'title': info['title'],
             'description': info.get('intro'),
             'duration': duration,
+            'view_count': view_count,
             'uploader': info.get('programName'),
             'timestamp': upload_timestamp,
             'thumbnail': thumbnail,

youtube_dl/extractor/teamcoco.py

@@ -9,8 +9,18 @@ from ..utils import (
 
 
 class TeamcocoIE(InfoExtractor):
-    _VALID_URL = r'http://teamcoco\.com/video/(?P<url_title>.*)'
-    _TEST = {
+    _VALID_URL = r'http://teamcoco\.com/video/(?P<video_id>[0-9]+)?/?(?P<url_title>.*)'
+    _TESTS = [
+    {
+        'url': 'http://teamcoco.com/video/80187/conan-becomes-a-mary-kay-beauty-consultant',
+        'file': '80187.mp4',
+        'md5': '3f7746aa0dc86de18df7539903d399ea',
+        'info_dict': {
+            'title': 'Conan Becomes A Mary Kay Beauty Consultant',
+            'description': 'Mary Kay is perhaps the most trusted name in female beauty, so of course Conan is a natural choice to sell their products.'
+        }
+    },
+    {
         'url': 'http://teamcoco.com/video/louis-ck-interview-george-w-bush',
         'file': '19705.mp4',
         'md5': 'cde9ba0fa3506f5f017ce11ead928f9a',
@@ -19,6 +29,7 @@ class TeamcocoIE(InfoExtractor):
             "title": "Louis C.K. Interview Pt. 1 11/3/11"
         }
     }
+    ]
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
@@ -26,11 +37,13 @@ class TeamcocoIE(InfoExtractor):
             raise ExtractorError('Invalid URL: %s' % url)
         url_title = mobj.group('url_title')
         webpage = self._download_webpage(url, url_title)
-
-        video_id = self._html_search_regex(
-            r'<article class="video" data-id="(\d+?)"',
-            webpage, 'video id')
-
+        
+        video_id = mobj.group("video_id")
+        if video_id == '':
+            video_id = self._html_search_regex(
+                r'<article class="video" data-id="(\d+?)"',
+                webpage, 'video id')
+        
         self.report_extraction(video_id)
 
         data_url = 'http://teamcoco.com/cvp/2.0/%s.xml' % video_id

youtube_dl/extractor/yahoo.py

@@ -15,22 +15,24 @@ from ..utils import (
 
 class YahooIE(InfoExtractor):
     IE_DESC = 'Yahoo screen'
-    _VALID_URL = r'http://screen\.yahoo\.com/.*?-(?P<id>\d*?)\.html'
+    _VALID_URL = r'https?://screen\.yahoo\.com/.*?-(?P<id>[0-9]+)(?:-[a-z]+)?\.html'
     _TESTS = [
         {
             'url': 'http://screen.yahoo.com/julian-smith-travis-legg-watch-214727115.html',
-            'file': '214727115.mp4',
             'md5': '4962b075c08be8690a922ee026d05e69',
             'info_dict': {
+                'id': '214727115',
+                'ext': 'mp4',
                 'title': 'Julian Smith & Travis Legg Watch Julian Smith',
                 'description': 'Julian and Travis watch Julian Smith',
             },
         },
         {
             'url': 'http://screen.yahoo.com/wired/codefellas-s1-ep12-cougar-lies-103000935.html',
-            'file': '103000935.mp4',
             'md5': 'd6e6fc6e1313c608f316ddad7b82b306',
             'info_dict': {
+                'id': '103000935',
+                'ext': 'mp4',
                 'title': 'Codefellas - The Cougar Lies with Spanish Moss',
                 'description': 'Agent Topple\'s mustache does its dirty work, and Nicole brokers a deal for peace. But why is the NSA collecting millions of Instagram brunch photos? And if your waffles have nothing to hide, what are they so worried about?',
             },
@@ -60,10 +62,9 @@ class YahooIE(InfoExtractor):
             'env': 'prod',
             'format': 'json',
         })
-        query_result_json = self._download_webpage(
+        query_result = self._download_json(
             'http://video.query.yahoo.com/v1/public/yql?' + data,
             video_id, 'Downloading video info')
-        query_result = json.loads(query_result_json)
         info = query_result['query']['results']['mediaObj'][0]
         meta = info['meta']
 
@@ -86,7 +87,6 @@ class YahooIE(InfoExtractor):
             else:
                 format_url = compat_urlparse.urljoin(host, path)
                 format_info['url'] = format_url
-                
             formats.append(format_info)
 
         self._sort_formats(formats)
@@ -134,27 +134,25 @@ class YahooSearchIE(SearchInfoExtractor):
 
     def _get_n_results(self, query, n):
         """Get a specified number of results for a query"""
-
-        res = {
-            '_type': 'playlist',
-            'id': query,
-            'entries': []
-        }
-        for pagenum in itertools.count(0): 
+        entries = []
+        for pagenum in itertools.count(0):
             result_url = 'http://video.search.yahoo.com/search/?p=%s&fr=screen&o=js&gs=0&b=%d' % (compat_urllib_parse.quote_plus(query), pagenum * 30)
-            webpage = self._download_webpage(result_url, query,
-                                             note='Downloading results page '+str(pagenum+1))
-            info = json.loads(webpage)
+            info = self._download_json(result_url, query,
+                note='Downloading results page '+str(pagenum+1))
             m = info['m']
             results = info['results']
 
             for (i, r) in enumerate(results):
-                if (pagenum * 30) +i >= n:
+                if (pagenum * 30) + i >= n:
                     break
                 mobj = re.search(r'(?P<url>screen\.yahoo\.com/.*?-\d*?\.html)"', r)
                 e = self.url_result('http://' + mobj.group('url'), 'Yahoo')
-                res['entries'].append(e)
-            if (pagenum * 30 +i >= n) or (m['last'] >= (m['total'] -1)):
+                entries.append(e)
+            if (pagenum * 30 + i >= n) or (m['last'] >= (m['total'] - 1)):
                 break
 
-        return res
+        return {
+            '_type': 'playlist',
+            'id': query,
+            'entries': entries,
+        }

youtube_dl/version.py

@@ -1,2 +1,2 @@
 
-__version__ = '2014.04.04.2'
+__version__ = '2014.04.04.6'