Merge remote-tracking branch 'origin/master'

AUTHORS

@@ -110,3 +110,4 @@ Shaya Goldberg
 Paul Hartmann
 Frans de Jonge
 Robin de Rooij
+Ryan Schmidt

test/test_subtitles.py

@@ -138,7 +138,7 @@ def test_allsubtitles(self):
         self.DL.params['writesubtitles'] = True
         self.DL.params['allsubtitles'] = True
         subtitles = self.getSubtitles()
-        self.assertEqual(len(subtitles.keys()), 5)
+        self.assertTrue(len(subtitles.keys()) >= 6)
 
     def test_list_subtitles(self):
         self.DL.expect_warning('Automatic Captions not supported by this server')
@@ -247,7 +247,7 @@ def test_no_writesubtitles(self):
     def test_subtitles(self):
         self.DL.params['writesubtitles'] = True
         subtitles = self.getSubtitles()
-        self.assertEqual(md5(subtitles['en']), '26399116d23ae3cf2c087cea94bc43b4')
+        self.assertEqual(md5(subtitles['en']), '8062383cf4dec168fc40a088aa6d5888')
 
     def test_subtitles_lang(self):
         self.DL.params['writesubtitles'] = True
@@ -334,7 +334,7 @@ def test_allsubtitles(self):
         self.DL.params['allsubtitles'] = True
         subtitles = self.getSubtitles()
         self.assertEqual(set(subtitles.keys()), set(['cs']))
-        self.assertEqual(md5(subtitles['cs']), '9bf52d9549533c32c427e264bf0847d4')
+        self.assertTrue(len(subtitles['cs']) > 20000)
 
     def test_nosubtitles(self):
         self.DL.expect_warning('video doesn\'t have subtitles')

youtube_dl/extractor/__init__.py

@@ -189,6 +189,7 @@
 from .helsinki import HelsinkiIE
 from .hentaistigma import HentaiStigmaIE
 from .historicfilms import HistoricFilmsIE
+from .history import HistoryIE
 from .hitbox import HitboxIE, HitboxLiveIE
 from .hornbunny import HornBunnyIE
 from .hostingbulk import HostingBulkIE

youtube_dl/extractor/bambuser.py

@@ -50,7 +50,7 @@ def _real_extract(self, url):
             'duration': int(info['length']),
             'view_count': int(info['views_total']),
             'uploader': info['username'],
-            'uploader_id': info['uid'],
+            'uploader_id': info['owner']['uid'],
         }
 
 

youtube_dl/extractor/bbccouk.py

@@ -273,7 +273,7 @@ def _download_playlist(self, playlist_id):
                     formats, subtitles = self._download_media_selector(programme_id)
                 return programme_id, title, description, duration, formats, subtitles
         except ExtractorError as ee:
-            if not isinstance(ee.cause, compat_HTTPError) and ee.cause.code == 404:
+            if not (isinstance(ee.cause, compat_HTTPError) and ee.cause.code == 404):
                 raise
 
         # fallback to legacy playlist

youtube_dl/extractor/beeg.py

@@ -9,7 +9,7 @@ class BeegIE(InfoExtractor):
     _VALID_URL = r'https?://(?:www\.)?beeg\.com/(?P<id>\d+)'
     _TEST = {
         'url': 'http://beeg.com/5416503',
-        'md5': '634526ae978711f6b748fe0dd6c11f57',
+        'md5': '1bff67111adb785c51d1b42959ec10e5',
         'info_dict': {
             'id': '5416503',
             'ext': 'mp4',

youtube_dl/extractor/camdemy.py

@@ -16,7 +16,7 @@
 
 
 class CamdemyIE(InfoExtractor):
-    _VALID_URL = r'http://www.camdemy.com/media/(?P<id>\d+)'
+    _VALID_URL = r'http://(?:www\.)?camdemy\.com/media/(?P<id>\d+)'
     _TESTS = [{
         # single file
         'url': 'http://www.camdemy.com/media/5181/',

youtube_dl/extractor/common.py

@@ -665,7 +665,7 @@ def _media_rating_search(self, html):
         return RATING_TABLE.get(rating.lower(), None)
 
     def _family_friendly_search(self, html):
-        # See http://schema.org/VideoObj
+        # See http://schema.org/VideoObject
         family_friendly = self._html_search_meta('isFamilyFriendly', html)
 
         if not family_friendly:

youtube_dl/extractor/drtuber.py

@@ -15,7 +15,7 @@ class DrTuberIE(InfoExtractor):
             'id': '1740434',
             'display_id': 'hot-perky-blonde-naked-golf',
             'ext': 'mp4',
-            'title': 'Hot Perky Blonde Naked Golf',
+            'title': 'hot perky blonde naked golf',
             'like_count': int,
             'dislike_count': int,
             'comment_count': int,
@@ -36,7 +36,8 @@ def _real_extract(self, url):
             r'<source src="([^"]+)"', webpage, 'video URL')
 
         title = self._html_search_regex(
-            r'<title>([^<]+)\s*-\s*Free', webpage, 'title')
+            [r'class="hd_title" style="[^"]+">([^<]+)</h1>', r'<title>([^<]+) - \d+'],
+            webpage, 'title')
 
         thumbnail = self._html_search_regex(
             r'poster="([^"]+)"',

youtube_dl/extractor/firsttv.py

@@ -1,52 +1,71 @@
 # encoding: utf-8
 from __future__ import unicode_literals
 
-import re
-
 from .common import InfoExtractor
 from ..utils import int_or_none
 
 
 class FirstTVIE(InfoExtractor):
-    IE_NAME = 'firsttv'
+    IE_NAME = '1tv'
-    IE_DESC = 'Видеоархив - Первый канал'
+    IE_DESC = 'Первый канал'
-    _VALID_URL = r'http://(?:www\.)?1tv\.ru/videoarchive/(?P<id>\d+)'
+    _VALID_URL = r'http://(?:www\.)?1tv\.ru/(?:[^/]+/)+(?P<id>.+)'
 
-    _TEST = {
+    _TESTS = [{
         'url': 'http://www.1tv.ru/videoarchive/73390',
-        'md5': '3de6390cf0cca4a5eae1d1d83895e5ad',
+        'md5': '777f525feeec4806130f4f764bc18a4f',
         'info_dict': {
             'id': '73390',
             'ext': 'mp4',
             'title': 'Олимпийские канатные дороги',
-            'description': 'md5:cc730d2bf4215463e37fff6a1e277b13',
+            'description': 'md5:d41d8cd98f00b204e9800998ecf8427e',
-            'thumbnail': 'http://img1.1tv.ru/imgsize640x360/PR20140210114657.JPG',
+            'thumbnail': 're:^https?://.*\.(?:jpg|JPG)$',
             'duration': 149,
+            'like_count': int,
+            'dislike_count': int,
         },
         'skip': 'Only works from Russia',
-    }
+    }, {
+        'url': 'http://www.1tv.ru/prj/inprivate/vypusk/35930',
+        'md5': 'a1b6b60d530ebcf8daacf4565762bbaf',
+        'info_dict': {
+            'id': '35930',
+            'ext': 'mp4',
+            'title': 'Наедине со всеми. Людмила Сенчина',
+            'description': 'md5:89553aed1d641416001fe8d450f06cb9',
+            'thumbnail': 're:^https?://.*\.(?:jpg|JPG)$',
+            'duration': 2694,
+        },
+        'skip': 'Only works from Russia',
+    }]
 
     def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
+        video_id = self._match_id(url)
-        video_id = mobj.group('id')
 
         webpage = self._download_webpage(url, video_id, 'Downloading page')
 
         video_url = self._html_search_regex(
-            r'''(?s)jwplayer\('flashvideoportal_1'\)\.setup\({.*?'file': '([^']+)'.*?}\);''', webpage, 'video URL')
+            r'''(?s)(?:jwplayer\('flashvideoportal_1'\)\.setup\({|var\s+playlistObj\s*=).*?'file'\s*:\s*'([^']+)'.*?}\);''',
+            webpage, 'video URL')
 
         title = self._html_search_regex(
-            r'<div class="tv_translation">\s*<h1><a href="[^"]+">([^<]*)</a>', webpage, 'title')
+            [r'<div class="tv_translation">\s*<h1><a href="[^"]+">([^<]*)</a>',
+             r"'title'\s*:\s*'([^']+)'"], webpage, 'title')
         description = self._html_search_regex(
-            r'<div class="descr">\s*<div>&nbsp;</div>\s*<p>([^<]*)</p></div>', webpage, 'description', fatal=False)
+            r'<div class="descr">\s*<div>&nbsp;</div>\s*<p>([^<]*)</p></div>',
+            webpage, 'description', default=None) or self._html_search_meta(
+                'description', webpage, 'description')
 
         thumbnail = self._og_search_thumbnail(webpage)
-        duration = self._og_search_property('video:duration', webpage, 'video duration', fatal=False)
+        duration = self._og_search_property(
+            'video:duration', webpage,
+            'video duration', fatal=False)
 
-        like_count = self._html_search_regex(r'title="Понравилось".*?/></label> \[(\d+)\]',
+        like_count = self._html_search_regex(
-                                             webpage, 'like count', fatal=False)
+            r'title="Понравилось".*?/></label> \[(\d+)\]',
-        dislike_count = self._html_search_regex(r'title="Не понравилось".*?/></label> \[(\d+)\]',
+            webpage, 'like count', default=None)
-                                                webpage, 'dislike count', fatal=False)
+        dislike_count = self._html_search_regex(
+            r'title="Не понравилось".*?/></label> \[(\d+)\]',
+            webpage, 'dislike count', default=None)
 
         return {
             'id': video_id,

youtube_dl/extractor/history.py

@@ -0,0 +1,31 @@
+from __future__ import unicode_literals
+
+from .common import InfoExtractor
+from ..utils import smuggle_url
+
+
+class HistoryIE(InfoExtractor):
+    _VALID_URL = r'https?://(?:www\.)?history\.com/(?:[^/]+/)+(?P<id>[^/]+?)(?:$|[?#])'
+
+    _TESTS = [{
+        'url': 'http://www.history.com/topics/valentines-day/history-of-valentines-day/videos/bet-you-didnt-know-valentines-day?m=528e394da93ae&s=undefined&f=1&free=false',
+        'md5': '6fe632d033c92aa10b8d4a9be047a7c5',
+        'info_dict': {
+            'id': 'bLx5Dv5Aka1G',
+            'ext': 'mp4',
+            'title': "Bet You Didn't Know: Valentine's Day",
+            'description': 'md5:7b57ea4829b391995b405fa60bd7b5f7',
+        },
+        'add_ie': ['ThePlatform'],
+    }]
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+
+        webpage = self._download_webpage(url, video_id)
+
+        video_url = self._search_regex(
+            r'data-href="[^"]*/%s"[^>]+data-release-url="([^"]+)"' % video_id,
+            webpage, 'video url')
+
+        return self.url_result(smuggle_url(video_url, {'sig': {'key': 'crazyjava', 'secret': 's3cr3t'}}))

youtube_dl/extractor/nbc.py

@@ -1,7 +1,6 @@
 from __future__ import unicode_literals
 
 import re
-import json
 
 from .common import InfoExtractor
 from ..compat import (
@@ -52,9 +51,9 @@ def _real_extract(self, url):
 
 
 class NBCNewsIE(InfoExtractor):
-    _VALID_URL = r'''(?x)https?://www\.nbcnews\.com/
+    _VALID_URL = r'''(?x)https?://(?:www\.)?nbcnews\.com/
-        ((video/.+?/(?P<id>\d+))|
+        (?:video/.+?/(?P<id>\d+)|
-        (feature/[^/]+/(?P<title>.+)))
+        (?:feature|nightly-news)/[^/]+/(?P<title>.+))
         '''
 
     _TESTS = [
@@ -89,6 +88,16 @@ class NBCNewsIE(InfoExtractor):
                 'description': 'md5:757988edbaae9d7be1d585eb5d55cc04',
             },
         },
+        {
+            'url': 'http://www.nbcnews.com/nightly-news/video/nightly-news-with-brian-williams-full-broadcast-february-4-394064451844',
+            'md5': 'b5dda8cddd8650baa0dcb616dd2cf60d',
+            'info_dict': {
+                'id': 'sekXqyTVnmN3',
+                'ext': 'mp4',
+                'title': 'Nightly News with Brian Williams Full Broadcast (February 4)',
+                'description': 'md5:1c10c1eccbe84a26e5debb4381e2d3c5',
+            },
+        },
     ]
 
     def _real_extract(self, url):
@@ -107,13 +116,13 @@ def _real_extract(self, url):
                 'thumbnail': find_xpath_attr(info, 'media', 'type', 'thumbnail').text,
             }
         else:
-            # "feature" pages use theplatform.com
+            # "feature" and "nightly-news" pages use theplatform.com
             title = mobj.group('title')
             webpage = self._download_webpage(url, title)
             bootstrap_json = self._search_regex(
-                r'var bootstrapJson = ({.+})\s*$', webpage, 'bootstrap json',
+                r'var\s+(?:bootstrapJson|playlistData)\s*=\s*({.+});?\s*$',
-                flags=re.MULTILINE)
+                webpage, 'bootstrap json', flags=re.MULTILINE)
-            bootstrap = json.loads(bootstrap_json)
+            bootstrap = self._parse_json(bootstrap_json, video_id)
             info = bootstrap['results'][0]['video']
             mpxid = info['mpxId']
 

youtube_dl/extractor/streamcz.py

@@ -1,14 +1,30 @@
 # -*- coding: utf-8 -*-
 from __future__ import unicode_literals
 
+import hashlib
+import time
+
 from .common import InfoExtractor
+from ..compat import (
+    compat_urllib_request,
+)
 from ..utils import (
     int_or_none,
 )
 
 
+def _get_api_key(api_path):
+    if api_path.endswith('?'):
+        api_path = api_path[:-1]
+
+    api_key = 'fb5f58a820353bd7095de526253c14fd'
+    a = '{0:}{1:}{2:}'.format(api_key, api_path, int(round(time.time() / 24 / 3600)))
+    return hashlib.md5(a.encode('ascii')).hexdigest()
+
+
 class StreamCZIE(InfoExtractor):
     _VALID_URL = r'https?://(?:www\.)?stream\.cz/.+/(?P<id>[0-9]+)'
+    _API_URL = 'http://www.stream.cz/API'
 
     _TESTS = [{
         'url': 'http://www.stream.cz/peklonataliri/765767-ecka-pro-deti',
@@ -36,8 +52,11 @@ class StreamCZIE(InfoExtractor):
 
     def _real_extract(self, url):
         video_id = self._match_id(url)
-        data = self._download_json(
+        api_path = '/episode/%s' % video_id
-            'http://www.stream.cz/API/episode/%s' % video_id, video_id)
+
+        req = compat_urllib_request.Request(self._API_URL + api_path)
+        req.add_header('Api-Password', _get_api_key(api_path))
+        data = self._download_json(req, video_id)
 
         formats = []
         for quality, video in enumerate(data['video_qualities']):

youtube_dl/extractor/sunporno.py

@@ -52,7 +52,7 @@ def _real_extract(self, url):
 
         formats = []
         quality = qualities(['mp4', 'flv'])
-        for video_url in re.findall(r'<source src="([^"]+)"', webpage):
+        for video_url in re.findall(r'<(?:source|video) src="([^"]+)"', webpage):
             video_ext = determine_ext(video_url)
             formats.append({
                 'url': video_url,

youtube_dl/extractor/theplatform.py

@@ -2,6 +2,11 @@
 
 import re
 import json
+import time
+import hmac
+import binascii
+import hashlib
+
 
 from .subtitles import SubtitlesInfoExtractor
 from ..compat import (
@@ -11,6 +16,7 @@
     determine_ext,
     ExtractorError,
     xpath_with_ns,
+    unsmuggle_url,
 )
 
 _x = lambda p: xpath_with_ns(p, {'smil': 'http://www.w3.org/2005/SMIL21/Language'})
@@ -18,7 +24,7 @@
 
 class ThePlatformIE(SubtitlesInfoExtractor):
     _VALID_URL = r'''(?x)
-        (?:https?://(?:link|player)\.theplatform\.com/[sp]/[^/]+/
+        (?:https?://(?:link|player)\.theplatform\.com/[sp]/(?P<provider_id>[^/]+)/
            (?P<config>(?:[^/\?]+/(?:swf|config)|onsite)/select/)?
          |theplatform:)(?P<id>[^/\?&]+)'''
 
@@ -38,9 +44,33 @@ class ThePlatformIE(SubtitlesInfoExtractor):
         },
     }
 
+    @staticmethod
+    def _sign_url(url, sig_key, sig_secret, life=600, include_qs=False):
+        flags = '10' if include_qs else '00'
+        expiration_date = '%x' % (int(time.time()) + life)
+
+        def str_to_hex(str):
+            return binascii.b2a_hex(str.encode('ascii')).decode('ascii')
+
+        def hex_to_str(hex):
+            return binascii.a2b_hex(hex)
+
+        relative_path = url.split('http://link.theplatform.com/s/')[1].split('?')[0]
+        clear_text = hex_to_str(flags + expiration_date + str_to_hex(relative_path))
+        checksum = hmac.new(sig_key.encode('ascii'), clear_text, hashlib.sha1).hexdigest()
+        sig = flags + expiration_date + checksum + str_to_hex(sig_secret)
+        return '%s&sig=%s' % (url, sig)
+
     def _real_extract(self, url):
+        url, smuggled_data = unsmuggle_url(url, {})
+
         mobj = re.match(self._VALID_URL, url)
+        provider_id = mobj.group('provider_id')
         video_id = mobj.group('id')
+
+        if not provider_id:
+            provider_id = 'dJ5BDC'
+
         if mobj.group('config'):
             config_url = url + '&form=json'
             config_url = config_url.replace('swf/', 'config/')
@@ -48,8 +78,12 @@ def _real_extract(self, url):
             config = self._download_json(config_url, video_id, 'Downloading config')
             smil_url = config['releaseUrl'] + '&format=SMIL&formats=MPEG4&manifest=f4m'
         else:
-            smil_url = ('http://link.theplatform.com/s/dJ5BDC/{0}/meta.smil?'
+            smil_url = ('http://link.theplatform.com/s/{0}/{1}/meta.smil?'
-                        'format=smil&mbr=true'.format(video_id))
+                        'format=smil&mbr=true'.format(provider_id, video_id))
+
+        sig = smuggled_data.get('sig')
+        if sig:
+            smil_url = self._sign_url(smil_url, sig['key'], sig['secret'])
 
         meta = self._download_xml(smil_url, video_id)
         try:
@@ -62,7 +96,7 @@ def _real_extract(self, url):
         else:
             raise ExtractorError(error_msg, expected=True)
 
-        info_url = 'http://link.theplatform.com/s/dJ5BDC/{0}?format=preview'.format(video_id)
+        info_url = 'http://link.theplatform.com/s/{0}/{1}?format=preview'.format(provider_id, video_id)
         info_json = self._download_webpage(info_url, video_id)
         info = json.loads(info_json)
 

youtube_dl/postprocessor/ffmpeg.py

@@ -138,7 +138,7 @@ def run_ffmpeg_multiple_files(self, input_paths, out_path, opts):
 
         if self._downloader.params.get('verbose', False):
             self._downloader.to_screen('[debug] ffmpeg command line: %s' % shell_quote(cmd))
-        p = subprocess.Popen(cmd, stdout=subprocess.PIPE, stderr=subprocess.PIPE)
+        p = subprocess.Popen(cmd, stdout=subprocess.PIPE, stderr=subprocess.PIPE, stdin=subprocess.PIPE)
         stdout, stderr = p.communicate()
         if p.returncode != 0:
             stderr = stderr.decode('utf-8', 'replace')
@@ -178,8 +178,8 @@ def get_audio_codec(self, path):
                 encodeArgument('-show_streams'),
                 encodeFilename(self._ffmpeg_filename_argument(path), True)]
             if self._downloader.params.get('verbose', False):
-                self._downloader.to_screen('[debug] ffprobe command line: %s' % shell_quote(cmd))
+                self._downloader.to_screen('[debug] %s command line: %s' % (self.basename, shell_quote(cmd)))
-            handle = subprocess.Popen(cmd, stderr=compat_subprocess_get_DEVNULL(), stdout=subprocess.PIPE)
+            handle = subprocess.Popen(cmd, stderr=compat_subprocess_get_DEVNULL(), stdout=subprocess.PIPE, stdin=subprocess.PIPE)
             output = handle.communicate()[0]
             if handle.wait() != 0:
                 return None