release 2017.07.09

mymedia.yam.com is dead. An wikipedia user also pointed out that Yam's
blog service is no longer available. [1]

[1] https://zh.wikipedia.org/zh-tw/%E5%A4%A9%E7%A9%BA%E9%83%A8%E8%90%BD

.github/ISSUE_TEMPLATE.md

@@ -6,8 +6,8 @@
 
 ---
 
-### Make sure you are using the *latest* version: run `youtube-dl --version` and ensure your version is *2017.07.02*. If it's not read [this FAQ entry](https://github.com/rg3/youtube-dl/blob/master/README.md#how-do-i-update-youtube-dl) and update. Issues with outdated version will be rejected.
-- [ ] I've **verified** and **I assure** that I'm running youtube-dl **2017.07.02**
+### Make sure you are using the *latest* version: run `youtube-dl --version` and ensure your version is *2017.07.09*. If it's not read [this FAQ entry](https://github.com/rg3/youtube-dl/blob/master/README.md#how-do-i-update-youtube-dl) and update. Issues with outdated version will be rejected.
+- [ ] I've **verified** and **I assure** that I'm running youtube-dl **2017.07.09**
 
 ### Before submitting an *issue* make sure you have:
 - [ ] At least skimmed through [README](https://github.com/rg3/youtube-dl/blob/master/README.md) and **most notably** [FAQ](https://github.com/rg3/youtube-dl#faq) and [BUGS](https://github.com/rg3/youtube-dl#bugs) sections
@@ -35,7 +35,7 @@ $ youtube-dl -v <your command line>
 [debug] User config: []
 [debug] Command-line args: [u'-v', u'http://www.youtube.com/watch?v=BaW_jenozKcj']
 [debug] Encodings: locale cp1251, fs mbcs, out cp866, pref cp1251
-[debug] youtube-dl version 2017.07.02
+[debug] youtube-dl version 2017.07.09
 [debug] Python version 2.7.11 - Windows-2003Server-5.2.3790-SP2
 [debug] exe versions: ffmpeg N-75573-g1d0487f, ffprobe N-75573-g1d0487f, rtmpdump 2.4
 [debug] Proxy map: {}

AUTHORS

@@ -220,3 +220,6 @@ gritstub
 Adam Voss
 Mike Fährmann
 Jan Kundrát
+Giuseppe Fabiano
+Örn Guðjónsson
+Parmjit Virk

ChangeLog

@@ -1,3 +1,29 @@
+version 2017.07.09
+
+Core
++ [extractor/common] Add support for AMP tags in _parse_html5_media_entries
++ [utils] Support attributes with no values in get_elements_by_attribute
+
+Extractors
++ [dailymail] Add support for embeds
++ [joj] Add support for joj.sk (#13268)
+* [abc.net.au:iview] Extract more formats (#13492, #13489)
+* [egghead:course] Fix extraction (#6635, #13370)
++ [cjsw] Add support for cjsw.com (#13525)
++ [eagleplatform] Add support for referrer protected videos (#13557)
++ [eagleplatform] Add support for another embed pattern (#13557)
+* [veoh] Extend URL regular expression (#13601)
+* [npo:live] Fix live stream id extraction (#13568, #13605)
+* [googledrive] Fix height extraction (#13603)
++ [dailymotion] Add support for new layout (#13580)
+- [yam] Remove extractor
+* [xhamster] Extract all formats and fix duration extraction (#13593)
++ [xhamster] Add support for new URL schema (#13593)
+* [espn] Extend URL regular expression (#13244, #13549)
+* [kaltura] Fix typo in subtitles extraction (#13569)
+* [vier] Adapt extraction to redesign (#13575)
+
+
 version 2017.07.02
 
 Core

docs/supportedsites.md

@@ -154,6 +154,7 @@
  - **chirbit**
  - **chirbit:profile**
  - **Cinchcast**
+ - **CJSW**
  - **Clipfish**
  - **cliphunter**
  - **ClipRs**
@@ -369,6 +370,7 @@
  - **Jamendo**
  - **JamendoAlbum**
  - **JeuxVideo**
+ - **Joj**
  - **Jove**
  - **jpopsuki.tv**
  - **JWPlatform**
@@ -996,7 +998,6 @@
  - **XVideos**
  - **XXXYMovies**
  - **Yahoo**: Yahoo screen and movies
- - **Yam**: 蕃薯藤yam天空部落
  - **yandexmusic:album**: Яндекс.Музыка - Альбом
  - **yandexmusic:playlist**: Яндекс.Музыка - Плейлист
  - **yandexmusic:track**: Яндекс.Музыка - Трек

test/test_utils.py

@@ -98,6 +98,7 @@ from youtube_dl.compat import (
     compat_chr,
     compat_etree_fromstring,
     compat_getenv,
+    compat_os_name,
     compat_setenv,
     compat_urlparse,
     compat_parse_qs,
@@ -448,7 +449,9 @@ class TestUtil(unittest.TestCase):
 
     def test_shell_quote(self):
         args = ['ffmpeg', '-i', encodeFilename('ñ€ß\'.mp4')]
-        self.assertEqual(shell_quote(args), """ffmpeg -i 'ñ€ß'"'"'.mp4'""")
+        self.assertEqual(
+            shell_quote(args),
+            """ffmpeg -i 'ñ€ß'"'"'.mp4'""" if compat_os_name != 'nt' else '''ffmpeg -i "ñ€ß'.mp4"''')
 
     def test_str_to_int(self):
         self.assertEqual(str_to_int('123,456'), 123456)
@@ -932,7 +935,7 @@ class TestUtil(unittest.TestCase):
     def test_args_to_str(self):
         self.assertEqual(
             args_to_str(['foo', 'ba/r', '-baz', '2 be', '']),
-            'foo ba/r -baz \'2 be\' \'\''
+            'foo ba/r -baz \'2 be\' \'\'' if compat_os_name != 'nt' else 'foo ba/r -baz "2 be" ""'
         )
 
     def test_parse_filesize(self):
@@ -1228,6 +1231,12 @@ part 3</font></u>
         self.assertEqual(get_element_by_attribute('class', 'foo', html), None)
         self.assertEqual(get_element_by_attribute('class', 'no-such-foo', html), None)
 
+        html = '''
+            <div itemprop="author" itemscope>foo</div>
+        '''
+
+        self.assertEqual(get_element_by_attribute('itemprop', 'author', html), 'foo')
+
     def test_get_elements_by_class(self):
         html = '''
             <span class="foo bar">nice</span><span class="foo bar">also nice</span>

youtube_dl/YoutubeDL.py

@@ -1890,7 +1890,7 @@ class YoutubeDL(object):
                         info_dict.get('protocol') == 'm3u8' and
                         self.params.get('hls_prefer_native')):
                     if fixup_policy == 'warn':
-                        self.report_warning('%s: malformated aac bitstream.' % (
+                        self.report_warning('%s: malformed AAC bitstream detected.' % (
                             info_dict['id']))
                     elif fixup_policy == 'detect_or_warn':
                         fixup_pp = FFmpegFixupM3u8PP(self)
@@ -1899,7 +1899,7 @@ class YoutubeDL(object):
                             info_dict['__postprocessors'].append(fixup_pp)
                         else:
                             self.report_warning(
-                                '%s: malformated aac bitstream. %s'
+                                '%s: malformed AAC bitstream detected. %s'
                                 % (info_dict['id'], INSTALL_FFMPEG_MESSAGE))
                     else:
                         assert fixup_policy in ('ignore', 'never')

youtube_dl/extractor/abc.py

@@ -3,11 +3,13 @@ from __future__ import unicode_literals
 import re
 
 from .common import InfoExtractor
+from ..compat import compat_str
 from ..utils import (
     ExtractorError,
     js_to_json,
     int_or_none,
     parse_iso8601,
+    try_get,
 )
 
 
@@ -124,7 +126,20 @@ class ABCIViewIE(InfoExtractor):
         title = video_params.get('title') or video_params['seriesTitle']
         stream = next(s for s in video_params['playlist'] if s.get('type') == 'program')
 
-        formats = self._extract_akamai_formats(stream['hds-unmetered'], video_id)
+        format_urls = [
+            try_get(stream, lambda x: x['hds-unmetered'], compat_str)]
+
+        # May have higher quality video
+        sd_url = try_get(
+            stream, lambda x: x['streams']['hds']['sd'], compat_str)
+        if sd_url:
+            format_urls.append(sd_url.replace('metered', 'um'))
+
+        formats = []
+        for format_url in format_urls:
+            if format_url:
+                formats.extend(
+                    self._extract_akamai_formats(format_url, video_id))
         self._sort_formats(formats)
 
         subtitles = {}

youtube_dl/extractor/cjsw.py

@@ -0,0 +1,72 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+import re
+
+from .common import InfoExtractor
+from ..utils import (
+    determine_ext,
+    unescapeHTML,
+)
+
+
+class CJSWIE(InfoExtractor):
+    _VALID_URL = r'https?://(?:www\.)?cjsw\.com/program/(?P<program>[^/]+)/episode/(?P<id>\d+)'
+    _TESTS = [{
+        'url': 'http://cjsw.com/program/freshly-squeezed/episode/20170620',
+        'md5': 'cee14d40f1e9433632c56e3d14977120',
+        'info_dict': {
+            'id': '91d9f016-a2e7-46c5-8dcb-7cbcd7437c41',
+            'ext': 'mp3',
+            'title': 'Freshly Squeezed – Episode June 20, 2017',
+            'description': 'md5:c967d63366c3898a80d0c7b0ff337202',
+            'series': 'Freshly Squeezed',
+            'episode_id': '20170620',
+        },
+    }, {
+        # no description
+        'url': 'http://cjsw.com/program/road-pops/episode/20170707/',
+        'only_matching': True,
+    }]
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        program, episode_id = mobj.group('program', 'id')
+        audio_id = '%s/%s' % (program, episode_id)
+
+        webpage = self._download_webpage(url, episode_id)
+
+        title = unescapeHTML(self._search_regex(
+            (r'<h1[^>]+class=["\']episode-header__title["\'][^>]*>(?P<title>[^<]+)',
+             r'data-audio-title=(["\'])(?P<title>(?:(?!\1).)+)\1'),
+            webpage, 'title', group='title'))
+
+        audio_url = self._search_regex(
+            r'<button[^>]+data-audio-src=(["\'])(?P<url>(?:(?!\1).)+)\1',
+            webpage, 'audio url', group='url')
+
+        audio_id = self._search_regex(
+            r'/([\da-f]{8}-[\da-f]{4}-[\da-f]{4}-[\da-f]{4}-[\da-f]{12})\.mp3',
+            audio_url, 'audio id', default=audio_id)
+
+        formats = [{
+            'url': audio_url,
+            'ext': determine_ext(audio_url, 'mp3'),
+            'vcodec': 'none',
+        }]
+
+        description = self._html_search_regex(
+            r'<p>(?P<description>.+?)</p>', webpage, 'description',
+            default=None)
+        series = self._search_regex(
+            r'data-showname=(["\'])(?P<name>(?:(?!\1).)+)\1', webpage,
+            'series', default=program, group='name')
+
+        return {
+            'id': audio_id,
+            'title': title,
+            'description': description,
+            'formats': formats,
+            'series': series,
+            'episode_id': episode_id,
+        }

youtube_dl/extractor/common.py

@@ -2132,15 +2132,18 @@ class InfoExtractor(object):
             return is_plain_url, formats
 
         entries = []
+        # amp-video and amp-audio are very similar to their HTML5 counterparts
+        # so we wll include them right here (see
+        # https://www.ampproject.org/docs/reference/components/amp-video)
         media_tags = [(media_tag, media_type, '')
                       for media_tag, media_type
-                      in re.findall(r'(?s)(<(video|audio)[^>]*/>)', webpage)]
+                      in re.findall(r'(?s)(<(?:amp-)?(video|audio)[^>]*/>)', webpage)]
         media_tags.extend(re.findall(
             # We only allow video|audio followed by a whitespace or '>'.
             # Allowing more characters may end up in significant slow down (see
             # https://github.com/rg3/youtube-dl/issues/11979, example URL:
             # http://www.porntrex.com/maps/videositemap.xml).
-            r'(?s)(<(?P<tag>video|audio)(?:\s+[^>]*)?>)(.*?)</(?P=tag)>', webpage))
+            r'(?s)(<(?P<tag>(?:amp-)?(?:video|audio))(?:\s+[^>]*)?>)(.*?)</(?P=tag)>', webpage))
         for media_tag, media_type, media_content in media_tags:
             media_info = {
                 'formats': [],

youtube_dl/extractor/dailymail.py

@@ -1,6 +1,8 @@
 # coding: utf-8
 from __future__ import unicode_literals
 
+import re
+
 from .common import InfoExtractor
 from ..compat import compat_str
 from ..utils import (
@@ -12,8 +14,8 @@ from ..utils import (
 
 
 class DailyMailIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?dailymail\.co\.uk/video/[^/]+/video-(?P<id>[0-9]+)'
-    _TEST = {
+    _VALID_URL = r'https?://(?:www\.)?dailymail\.co\.uk/(?:video/[^/]+/video-|embed/video/)(?P<id>[0-9]+)'
+    _TESTS = [{
         'url': 'http://www.dailymail.co.uk/video/tvshowbiz/video-1295863/The-Mountain-appears-sparkling-water-ad-Heavy-Bubbles.html',
         'md5': 'f6129624562251f628296c3a9ffde124',
         'info_dict': {
@@ -22,7 +24,16 @@ class DailyMailIE(InfoExtractor):
             'title': 'The Mountain appears in sparkling water ad for \'Heavy Bubbles\'',
             'description': 'md5:a93d74b6da172dd5dc4d973e0b766a84',
         }
-    }
+    }, {
+        'url': 'http://www.dailymail.co.uk/embed/video/1295863.html',
+        'only_matching': True,
+    }]
+
+    @staticmethod
+    def _extract_urls(webpage):
+        return re.findall(
+            r'<iframe\b[^>]+\bsrc=["\'](?P<url>(?:https?:)?//(?:www\.)?dailymail\.co\.uk/embed/video/\d+\.html)',
+            webpage)
 
     def _real_extract(self, url):
         video_id = self._match_id(url)

youtube_dl/extractor/dailymotion.py

@@ -147,7 +147,7 @@ class DailymotionIE(DailymotionBaseInfoExtractor):
         view_count_str = self._search_regex(
             (r'<meta[^>]+itemprop="interactionCount"[^>]+content="UserPlays:([\s\d,.]+)"',
              r'video_views_count[^>]+>\s+([\s\d\,.]+)'),
-            webpage, 'view count', fatal=False)
+            webpage, 'view count', default=None)
         if view_count_str:
             view_count_str = re.sub(r'\s', '', view_count_str)
         view_count = str_to_int(view_count_str)
@@ -159,7 +159,9 @@ class DailymotionIE(DailymotionBaseInfoExtractor):
             [r'buildPlayer\(({.+?})\);\n',  # See https://github.com/rg3/youtube-dl/issues/7826
              r'playerV5\s*=\s*dmp\.create\([^,]+?,\s*({.+?})\);',
              r'buildPlayer\(({.+?})\);',
-             r'var\s+config\s*=\s*({.+?});'],
+             r'var\s+config\s*=\s*({.+?});',
+             # New layout regex (see https://github.com/rg3/youtube-dl/issues/13580)
+             r'__PLAYER_CONFIG__\s*=\s*({.+?});'],
             webpage, 'player v5', default=None)
         if player_v5:
             player = self._parse_json(player_v5, video_id)

youtube_dl/extractor/eagleplatform.py

@@ -11,6 +11,7 @@ from ..compat import (
 from ..utils import (
     ExtractorError,
     int_or_none,
+    unsmuggle_url,
 )
 
 
@@ -50,6 +51,10 @@ class EaglePlatformIE(InfoExtractor):
             'view_count': int,
         },
         'skip': 'Georestricted',
+    }, {
+        # referrer protected video (https://tvrain.ru/lite/teleshow/kak_vse_nachinalos/namin-418921/)
+        'url': 'tvrainru.media.eagleplatform.com:582306',
+        'only_matching': True,
     }]
 
     @staticmethod
@@ -60,16 +65,40 @@ class EaglePlatformIE(InfoExtractor):
             webpage)
         if mobj is not None:
             return mobj.group('url')
-        # Basic usage embedding (see http://dultonmedia.github.io/eplayer/)
+        PLAYER_JS_RE = r'''
+                        <script[^>]+
+                            src=(?P<qjs>["\'])(?:https?:)?//(?P<host>(?:(?!(?P=qjs)).)+\.media\.eagleplatform\.com)/player/player\.js(?P=qjs)
+                        .+?
+                    '''
+        # "Basic usage" embedding (see http://dultonmedia.github.io/eplayer/)
         mobj = re.search(
             r'''(?xs)
-                    <script[^>]+
-                        src=(?P<q1>["\'])(?:https?:)?//(?P<host>.+?\.media\.eagleplatform\.com)/player/player\.js(?P=q1)
-                    .+?
+                    %s
                     <div[^>]+
-                        class=(?P<q2>["\'])eagleplayer(?P=q2)[^>]+
+                        class=(?P<qclass>["\'])eagleplayer(?P=qclass)[^>]+
                         data-id=["\'](?P<id>\d+)
-            ''', webpage)
+            ''' % PLAYER_JS_RE, webpage)
+        if mobj is not None:
+            return 'eagleplatform:%(host)s:%(id)s' % mobj.groupdict()
+        # Generalization of "Javascript code usage", "Combined usage" and
+        # "Usage without attaching to DOM" embeddings (see
+        # http://dultonmedia.github.io/eplayer/)
+        mobj = re.search(
+            r'''(?xs)
+                    %s
+                    <script>
+                    .+?
+                    new\s+EaglePlayer\(
+                        (?:[^,]+\s*,\s*)?
+                        {
+                            .+?
+                            \bid\s*:\s*["\']?(?P<id>\d+)
+                            .+?
+                        }
+                    \s*\)
+                    .+?
+                    </script>
+            ''' % PLAYER_JS_RE, webpage)
         if mobj is not None:
             return 'eagleplatform:%(host)s:%(id)s' % mobj.groupdict()
 
@@ -79,9 +108,10 @@ class EaglePlatformIE(InfoExtractor):
         if status != 200:
             raise ExtractorError(' '.join(response['errors']), expected=True)
 
-    def _download_json(self, url_or_request, video_id, note='Downloading JSON metadata', *args, **kwargs):
+    def _download_json(self, url_or_request, video_id, *args, **kwargs):
         try:
-            response = super(EaglePlatformIE, self)._download_json(url_or_request, video_id, note)
+            response = super(EaglePlatformIE, self)._download_json(
+                url_or_request, video_id, *args, **kwargs)
         except ExtractorError as ee:
             if isinstance(ee.cause, compat_HTTPError):
                 response = self._parse_json(ee.cause.read().decode('utf-8'), video_id)
@@ -93,11 +123,24 @@ class EaglePlatformIE(InfoExtractor):
         return self._download_json(url_or_request, video_id, note)['data'][0]
 
     def _real_extract(self, url):
+        url, smuggled_data = unsmuggle_url(url, {})
+
         mobj = re.match(self._VALID_URL, url)
         host, video_id = mobj.group('custom_host') or mobj.group('host'), mobj.group('id')
 
+        headers = {}
+        query = {
+            'id': video_id,
+        }
+
+        referrer = smuggled_data.get('referrer')
+        if referrer:
+            headers['Referer'] = referrer
+            query['referrer'] = referrer
+
         player_data = self._download_json(
-            'http://%s/api/player_data?id=%s' % (host, video_id), video_id)
+            'http://%s/api/player_data' % host, video_id,
+            headers=headers, query=query)
 
         media = player_data['data']['playlist']['viewports'][0]['medialist'][0]
 

youtube_dl/extractor/egghead.py

@@ -1,15 +1,13 @@
 # coding: utf-8
 from __future__ import unicode_literals
 
-import re
-
 from .common import InfoExtractor
 
 
 class EggheadCourseIE(InfoExtractor):
     IE_DESC = 'egghead.io course'
     IE_NAME = 'egghead:course'
-    _VALID_URL = r'https://egghead\.io/courses/(?P<id>[a-zA-Z_0-9-]+)'
+    _VALID_URL = r'https://egghead\.io/courses/(?P<id>[^/?#&]+)'
     _TEST = {
         'url': 'https://egghead.io/courses/professor-frisby-introduces-composable-functional-javascript',
         'playlist_count': 29,
@@ -22,18 +20,16 @@ class EggheadCourseIE(InfoExtractor):
 
     def _real_extract(self, url):
         playlist_id = self._match_id(url)
-        webpage = self._download_webpage(url, playlist_id)
 
-        title = self._html_search_regex(r'<h1 class="title">([^<]+)</h1>', webpage, 'title')
-        ul = self._search_regex(r'(?s)<ul class="series-lessons-list">(.*?)</ul>', webpage, 'session list')
+        course = self._download_json(
+            'https://egghead.io/api/v1/series/%s' % playlist_id, playlist_id)
 
-        found = re.findall(r'(?s)<a class="[^"]*"\s*href="([^"]+)">\s*<li class="item', ul)
-        entries = [self.url_result(m) for m in found]
+        entries = [
+            self.url_result(
+                'wistia:%s' % lesson['wistia_id'], ie='Wistia',
+                video_id=lesson['wistia_id'], video_title=lesson.get('title'))
+            for lesson in course['lessons'] if lesson.get('wistia_id')]
 
-        return {
-            '_type': 'playlist',
-            'id': playlist_id,
-            'title': title,
-            'description': self._og_search_description(webpage),
-            'entries': entries,
-        }
+        return self.playlist_result(
+            entries, playlist_id, course.get('title'),
+            course.get('description'))

youtube_dl/extractor/espn.py

@@ -10,7 +10,25 @@ from ..utils import (
 
 
 class ESPNIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:espn\.go|(?:www\.)?espn)\.com/video/clip(?:\?.*?\bid=|/_/id/)(?P<id>\d+)'
+    _VALID_URL = r'''(?x)
+                    https?://
+                        (?:
+                            (?:(?:\w+\.)+)?espn\.go|
+                            (?:www\.)?espn
+                        )\.com/
+                        (?:
+                            (?:
+                                video/clip|
+                                watch/player
+                            )
+                            (?:
+                                \?.*?\bid=|
+                                /_/id/
+                            )
+                        )
+                        (?P<id>\d+)
+                    '''
+
     _TESTS = [{
         'url': 'http://espn.go.com/video/clip?id=10365079',
         'info_dict': {
@@ -25,20 +43,34 @@ class ESPNIE(InfoExtractor):
             'skip_download': True,
         },
     }, {
-        # intl video, from http://www.espnfc.us/video/mls-highlights/150/video/2743663/must-see-moments-best-of-the-mls-season
-        'url': 'http://espn.go.com/video/clip?id=2743663',
+        'url': 'https://broadband.espn.go.com/video/clip?id=18910086',
         'info_dict': {
-            'id': '2743663',
+            'id': '18910086',
             'ext': 'mp4',
-            'title': 'Must-See Moments: Best of the MLS season',
-            'description': 'md5:4c2d7232beaea572632bec41004f0aeb',
-            'timestamp': 1449446454,
-            'upload_date': '20151207',
+            'title': 'Kyrie spins around defender for two',
+            'description': 'md5:2b0f5bae9616d26fba8808350f0d2b9b',
+            'timestamp': 1489539155,
+            'upload_date': '20170315',
         },
         'params': {
             'skip_download': True,
         },
         'expected_warnings': ['Unable to download f4m manifest'],
+    }, {
+        'url': 'http://nonredline.sports.espn.go.com/video/clip?id=19744672',
+        'only_matching': True,
+    }, {
+        'url': 'https://cdn.espn.go.com/video/clip/_/id/19771774',
+        'only_matching': True,
+    }, {
+        'url': 'http://www.espn.com/watch/player?id=19141491',
+        'only_matching': True,
+    }, {
+        'url': 'http://www.espn.com/watch/player?bucketId=257&id=19505875',
+        'only_matching': True,
+    }, {
+        'url': 'http://www.espn.com/watch/player/_/id/19141491',
+        'only_matching': True,
     }, {
         'url': 'http://www.espn.com/video/clip?id=10365079',
         'only_matching': True,

youtube_dl/extractor/extractors.py

@@ -185,6 +185,7 @@ from .chirbit import (
     ChirbitProfileIE,
 )
 from .cinchcast import CinchcastIE
+from .cjsw import CJSWIE
 from .clipfish import ClipfishIE
 from .cliphunter import CliphunterIE
 from .cliprs import ClipRsIE
@@ -469,6 +470,7 @@ from .jamendo import (
 )
 from .jeuxvideo import JeuxVideoIE
 from .jove import JoveIE
+from .joj import JojIE
 from .jwplatform import JWPlatformIE
 from .jpopsukitv import JpopsukiIE
 from .kaltura import KalturaIE
@@ -1280,7 +1282,6 @@ from .yahoo import (
     YahooIE,
     YahooSearchIE,
 )
-from .yam import YamIE
 from .yandexmusic import (
     YandexMusicTrackIE,
     YandexMusicAlbumIE,

youtube_dl/extractor/generic.py

@@ -57,6 +57,7 @@ from .dailymotion import (
     DailymotionIE,
     DailymotionCloudIE,
 )
+from .dailymail import DailyMailIE
 from .onionstudios import OnionStudiosIE
 from .viewlift import ViewLiftEmbedIE
 from .mtv import MTVServicesEmbeddedIE
@@ -91,6 +92,7 @@ from .anvato import AnvatoIE
 from .washingtonpost import WashingtonPostIE
 from .wistia import WistiaIE
 from .mediaset import MediasetIE
+from .joj import JojIE
 
 
 class GenericIE(InfoExtractor):
@@ -759,6 +761,20 @@ class GenericIE(InfoExtractor):
             },
             'add_ie': ['Dailymotion'],
         },
+        # DailyMail embed
+        {
+            'url': 'http://www.bumm.sk/krimi/2017/07/05/biztonsagi-kamera-buktatta-le-az-agg-ferfit-utlegelo-apolot',
+            'info_dict': {
+                'id': '1495629',
+                'ext': 'mp4',
+                'title': 'Care worker punches elderly dementia patient in head 11 times',
+                'description': 'md5:3a743dee84e57e48ec68bf67113199a5',
+            },
+            'add_ie': ['DailyMail'],
+            'params': {
+                'skip_download': True,
+            },
+        },
         # YouTube embed
         {
             'url': 'http://www.badzine.de/ansicht/datum/2014/06/09/so-funktioniert-die-neue-englische-badminton-liga.html',
@@ -1185,7 +1201,7 @@ class GenericIE(InfoExtractor):
             },
             'add_ie': ['Kaltura'],
         },
-        # Eagle.Platform embed (generic URL)
+        # EaglePlatform embed (generic URL)
         {
             'url': 'http://lenta.ru/news/2015/03/06/navalny/',
             # Not checking MD5 as sometimes the direct HTTP link results in 404 and HLS is used
@@ -1199,8 +1215,26 @@ class GenericIE(InfoExtractor):
                 'view_count': int,
                 'age_limit': 0,
             },
+            'params': {
+                'skip_download': True,
+            },
         },
-        # ClipYou (Eagle.Platform) embed (custom URL)
+        # referrer protected EaglePlatform embed
+        {
+            'url': 'https://tvrain.ru/lite/teleshow/kak_vse_nachinalos/namin-418921/',
+            'info_dict': {
+                'id': '582306',
+                'ext': 'mp4',
+                'title': 'Стас Намин: «Мы нарушили девственность Кремля»',
+                'thumbnail': r're:^https?://.*\.jpg$',
+                'duration': 3382,
+                'view_count': int,
+            },
+            'params': {
+                'skip_download': True,
+            },
+        },
+        # ClipYou (EaglePlatform) embed (custom URL)
         {
             'url': 'http://muz-tv.ru/play/7129/',
             # Not checking MD5 as sometimes the direct HTTP link results in 404 and HLS is used
@@ -1212,6 +1246,9 @@ class GenericIE(InfoExtractor):
                 'duration': 216,
                 'view_count': int,
             },
+            'params': {
+                'skip_download': True,
+            },
         },
         # Pladform embed
         {
@@ -1749,6 +1786,26 @@ class GenericIE(InfoExtractor):
             },
             'add_ie': [MediasetIE.ie_key()],
         },
+        {
+            # JOJ.sk embeds
+            'url': 'https://www.noviny.sk/slovensko/238543-slovenskom-sa-prehnala-vlna-silnych-burok',
+            'info_dict': {
+                'id': '238543-slovenskom-sa-prehnala-vlna-silnych-burok',
+                'title': 'Slovenskom sa prehnala vlna silných búrok',
+            },
+            'playlist_mincount': 5,
+            'add_ie': [JojIE.ie_key()],
+        },
+        {
+            # AMP embed (see https://www.ampproject.org/docs/reference/components/amp-video)
+            'url': 'https://tvrain.ru/amp/418921/',
+            'md5': 'cc00413936695987e8de148b67d14f1d',
+            'info_dict': {
+                'id': '418921',
+                'ext': 'mp4',
+                'title': 'Стас Намин: «Мы нарушили девственность Кремля»',
+            },
+        },
         # {
         #     # TODO: find another test
         #     # http://schema.org/VideoObject
@@ -2148,6 +2205,12 @@ class GenericIE(InfoExtractor):
                 return self.playlist_from_matches(
                     playlists, video_id, video_title, lambda p: '//dailymotion.com/playlist/%s' % p)
 
+        # Look for DailyMail embeds
+        dailymail_urls = DailyMailIE._extract_urls(webpage)
+        if dailymail_urls:
+            return self.playlist_from_matches(
+                dailymail_urls, video_id, video_title, ie=DailyMailIE.ie_key())
+
         # Look for embedded Wistia player
         wistia_url = WistiaIE._extract_url(webpage)
         if wistia_url:
@@ -2443,12 +2506,12 @@ class GenericIE(InfoExtractor):
         if kaltura_url:
             return self.url_result(smuggle_url(kaltura_url, {'source_url': url}), KalturaIE.ie_key())
 
-        # Look for Eagle.Platform embeds
+        # Look for EaglePlatform embeds
         eagleplatform_url = EaglePlatformIE._extract_url(webpage)
         if eagleplatform_url:
-            return self.url_result(eagleplatform_url, EaglePlatformIE.ie_key())
+            return self.url_result(smuggle_url(eagleplatform_url, {'referrer': url}), EaglePlatformIE.ie_key())
 
-        # Look for ClipYou (uses Eagle.Platform) embeds
+        # Look for ClipYou (uses EaglePlatform) embeds
         mobj = re.search(
             r'<iframe[^>]+src="https?://(?P<host>media\.clipyou\.ru)/index/player\?.*\brecord_id=(?P<id>\d+).*"', webpage)
         if mobj is not None:
@@ -2691,6 +2754,12 @@ class GenericIE(InfoExtractor):
             return self.playlist_from_matches(
                 mediaset_urls, video_id, video_title, ie=MediasetIE.ie_key())
 
+        # Look for JOJ.sk embeds
+        joj_urls = JojIE._extract_urls(webpage)
+        if joj_urls:
+            return self.playlist_from_matches(
+                joj_urls, video_id, video_title, ie=JojIE.ie_key())
+
         def merge_dicts(dict1, dict2):
             merged = {}
             for k, v in dict1.items():

youtube_dl/extractor/googledrive.py

@@ -92,7 +92,7 @@ class GoogleDriveIE(InfoExtractor):
             if resolution:
                 f.update({
                     'width': resolution[0],
-                    'height': resolution[0],
+                    'height': resolution[1],
                 })
             formats.append(f)
         self._sort_formats(formats)

youtube_dl/extractor/joj.py

@@ -0,0 +1,100 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+import re
+
+from .common import InfoExtractor
+from ..compat import compat_str
+from ..utils import (
+    int_or_none,
+    js_to_json,
+    try_get,
+)
+
+
+class JojIE(InfoExtractor):
+    _VALID_URL = r'''(?x)
+                    (?:
+                        joj:|
+                        https?://media\.joj\.sk/embed/
+                    )
+                    (?P<id>[\da-f]{8}-[\da-f]{4}-[\da-f]{4}-[\da-f]{4}-[\da-f]{12})
+                '''
+    _TESTS = [{
+        'url': 'https://media.joj.sk/embed/a388ec4c-6019-4a4a-9312-b1bee194e932',
+        'info_dict': {
+            'id': 'a388ec4c-6019-4a4a-9312-b1bee194e932',
+            'ext': 'mp4',
+            'title': 'NOVÉ BÝVANIE',
+            'thumbnail': r're:^https?://.*\.jpg$',
+            'duration': 3118,
+        }
+    }, {
+        'url': 'joj:a388ec4c-6019-4a4a-9312-b1bee194e932',
+        'only_matching': True,
+    }]
+
+    @staticmethod
+    def _extract_urls(webpage):
+        return re.findall(
+            r'<iframe\b[^>]+\bsrc=["\'](?P<url>(?:https?:)?//media\.joj\.sk/embed/[\da-f]{8}-[\da-f]{4}-[\da-f]{4}-[\da-f]{4}-[\da-f]{12})',
+            webpage)
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+
+        webpage = self._download_webpage(
+            'https://media.joj.sk/embed/%s' % video_id, video_id)
+
+        title = self._search_regex(
+            (r'videoTitle\s*:\s*(["\'])(?P<title>(?:(?!\1).)+)\1',
+             r'<title>(?P<title>[^<]+)'), webpage, 'title',
+            default=None, group='title') or self._og_search_title(webpage)
+
+        bitrates = self._parse_json(
+            self._search_regex(
+                r'(?s)bitrates\s*=\s*({.+?});', webpage, 'bitrates',
+                default='{}'),
+            video_id, transform_source=js_to_json, fatal=False)
+
+        formats = []
+        for format_url in try_get(bitrates, lambda x: x['mp4'], list) or []:
+            if isinstance(format_url, compat_str):
+                height = self._search_regex(
+                    r'(\d+)[pP]\.', format_url, 'height', default=None)
+                formats.append({
+                    'url': format_url,
+                    'format_id': '%sp' % height if height else None,
+                    'height': int(height),
+                })
+        if not formats:
+            playlist = self._download_xml(
+                'https://media.joj.sk/services/Video.php?clip=%s' % video_id,
+                video_id)
+            for file_el in playlist.findall('./files/file'):
+                path = file_el.get('path')
+                if not path:
+                    continue
+                format_id = file_el.get('id') or file_el.get('label')
+                formats.append({
+                    'url': 'http://n16.joj.sk/storage/%s' % path.replace(
+                        'dat/', '', 1),
+                    'format_id': format_id,
+                    'height': int_or_none(self._search_regex(
+                        r'(\d+)[pP]', format_id or path, 'height',
+                        default=None)),
+                })
+        self._sort_formats(formats)
+
+        thumbnail = self._og_search_thumbnail(webpage)
+
+        duration = int_or_none(self._search_regex(
+            r'videoDuration\s*:\s*(\d+)', webpage, 'duration', fatal=False))
+
+        return {
+            'id': video_id,
+            'title': title,
+            'thumbnail': thumbnail,
+            'duration': duration,
+            'formats': formats,
+        }

youtube_dl/extractor/kaltura.py

@@ -324,7 +324,7 @@ class KalturaIE(InfoExtractor):
         if captions:
             for caption in captions.get('objects', []):
                 # Continue if caption is not ready
-                if f.get('status') != 2:
+                if caption.get('status') != 2:
                     continue
                 if not caption.get('id'):
                     continue

youtube_dl/extractor/npo.py

@@ -341,7 +341,7 @@ class NPOLiveIE(NPOBaseIE):
         webpage = self._download_webpage(url, display_id)
 
         live_id = self._search_regex(
-            r'data-prid="([^"]+)"', webpage, 'live id')
+            [r'media-id="([^"]+)"', r'data-prid="([^"]+)"'], webpage, 'live id')
 
         return {
             '_type': 'url_transparent',

youtube_dl/extractor/veoh.py

@@ -12,47 +12,46 @@ from ..utils import (
 
 
 class VeohIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?veoh\.com/(?:watch|iphone/#_Watch)/(?P<id>(?:v|yapi-)[\da-zA-Z]+)'
+    _VALID_URL = r'https?://(?:www\.)?veoh\.com/(?:watch|iphone/#_Watch)/(?P<id>(?:v|e|yapi-)[\da-zA-Z]+)'
 
-    _TESTS = [
-        {
-            'url': 'http://www.veoh.com/watch/v56314296nk7Zdmz3',
-            'md5': '620e68e6a3cff80086df3348426c9ca3',
-            'info_dict': {
-                'id': '56314296',
-                'ext': 'mp4',
-                'title': 'Straight Backs Are Stronger',
-                'uploader': 'LUMOback',
-                'description': 'At LUMOback, we believe straight backs are stronger.  The LUMOback Posture & Movement Sensor:  It gently vibrates when you slouch, inspiring improved posture and mobility.  Use the app to track your data and improve your posture over time. ',
-            },
+    _TESTS = [{
+        'url': 'http://www.veoh.com/watch/v56314296nk7Zdmz3',
+        'md5': '620e68e6a3cff80086df3348426c9ca3',
+        'info_dict': {
+            'id': '56314296',
+            'ext': 'mp4',
+            'title': 'Straight Backs Are Stronger',
+            'uploader': 'LUMOback',
+            'description': 'At LUMOback, we believe straight backs are stronger.  The LUMOback Posture & Movement Sensor:  It gently vibrates when you slouch, inspiring improved posture and mobility.  Use the app to track your data and improve your posture over time. ',
         },
-        {
-            'url': 'http://www.veoh.com/watch/v27701988pbTc4wzN?h1=Chile+workers+cover+up+to+avoid+skin+damage',
-            'md5': '4a6ff84b87d536a6a71e6aa6c0ad07fa',
-            'info_dict': {
-                'id': '27701988',
-                'ext': 'mp4',
-                'title': 'Chile workers cover up to avoid skin damage',
-                'description': 'md5:2bd151625a60a32822873efc246ba20d',
-                'uploader': 'afp-news',
-                'duration': 123,
-            },
-            'skip': 'This video has been deleted.',
+    }, {
+        'url': 'http://www.veoh.com/watch/v27701988pbTc4wzN?h1=Chile+workers+cover+up+to+avoid+skin+damage',
+        'md5': '4a6ff84b87d536a6a71e6aa6c0ad07fa',
+        'info_dict': {
+            'id': '27701988',
+            'ext': 'mp4',
+            'title': 'Chile workers cover up to avoid skin damage',
+            'description': 'md5:2bd151625a60a32822873efc246ba20d',
+            'uploader': 'afp-news',
+            'duration': 123,
         },
-        {
-            'url': 'http://www.veoh.com/watch/v69525809F6Nc4frX',
-            'md5': '4fde7b9e33577bab2f2f8f260e30e979',
-            'note': 'Embedded ooyala video',
-            'info_dict': {
-                'id': '69525809',
-                'ext': 'mp4',
-                'title': 'Doctors Alter Plan For Preteen\'s Weight Loss Surgery',
-                'description': 'md5:f5a11c51f8fb51d2315bca0937526891',
-                'uploader': 'newsy-videos',
-            },
-            'skip': 'This video has been deleted.',
+        'skip': 'This video has been deleted.',
+    }, {
+        'url': 'http://www.veoh.com/watch/v69525809F6Nc4frX',
+        'md5': '4fde7b9e33577bab2f2f8f260e30e979',
+        'note': 'Embedded ooyala video',
+        'info_dict': {
+            'id': '69525809',
+            'ext': 'mp4',
+            'title': 'Doctors Alter Plan For Preteen\'s Weight Loss Surgery',
+            'description': 'md5:f5a11c51f8fb51d2315bca0937526891',
+            'uploader': 'newsy-videos',
         },
-    ]
+        'skip': 'This video has been deleted.',
+    }, {
+        'url': 'http://www.veoh.com/watch/e152215AJxZktGS',
+        'only_matching': True,
+    }]
 
     def _extract_formats(self, source):
         formats = []

youtube_dl/extractor/vier.py

@@ -15,7 +15,21 @@ from ..utils import (
 class VierIE(InfoExtractor):
     IE_NAME = 'vier'
     IE_DESC = 'vier.be and vijf.be'
-    _VALID_URL = r'https?://(?:www\.)?(?P<site>vier|vijf)\.be/(?:[^/]+/videos/(?P<display_id>[^/]+)(?:/(?P<id>\d+))?|video/v3/embed/(?P<embed_id>\d+))'
+    _VALID_URL = r'''(?x)
+                    https?://
+                        (?:www\.)?(?P<site>vier|vijf)\.be/
+                        (?:
+                            (?:
+                                [^/]+/videos|
+                                video(?:/[^/]+)*
+                            )/
+                            (?P<display_id>[^/]+)(?:/(?P<id>\d+))?|
+                            (?:
+                                video/v3/embed|
+                                embed/video/public
+                            )/(?P<embed_id>\d+)
+                        )
+                    '''
     _NETRC_MACHINE = 'vier'
     _TESTS = [{
         'url': 'http://www.vier.be/planb/videos/het-wordt-warm-de-moestuin/16129',
@@ -83,6 +97,15 @@ class VierIE(InfoExtractor):
     }, {
         'url': 'http://www.vier.be/video/v3/embed/16129',
         'only_matching': True,
+    }, {
+        'url': 'https://www.vijf.be/embed/video/public/4093',
+        'only_matching': True,
+    }, {
+        'url': 'https://www.vier.be/video/blockbusters/in-juli-en-augustus-summer-classics',
+        'only_matching': True,
+    }, {
+        'url': 'https://www.vier.be/video/achter-de-rug/2017/achter-de-rug-seizoen-1-aflevering-6',
+        'only_matching': True,
     }]
 
     def _real_initialize(self):
@@ -133,14 +156,20 @@ class VierIE(InfoExtractor):
         video_id = self._search_regex(
             [r'data-nid="(\d+)"', r'"nid"\s*:\s*"(\d+)"'],
             webpage, 'video id', default=video_id or display_id)
-        application = self._search_regex(
-            [r'data-application="([^"]+)"', r'"application"\s*:\s*"([^"]+)"'],
-            webpage, 'application', default=site + '_vod')
-        filename = self._search_regex(
-            [r'data-filename="([^"]+)"', r'"filename"\s*:\s*"([^"]+)"'],
-            webpage, 'filename')
 
-        playlist_url = 'http://vod.streamcloud.be/%s/_definst_/mp4:%s.mp4/playlist.m3u8' % (application, filename)
+        playlist_url = self._search_regex(
+            r'data-file=(["\'])(?P<url>(?:https?:)?//[^/]+/.+?\.m3u8.*?)\1',
+            webpage, 'm3u8 url', default=None, group='url')
+
+        if not playlist_url:
+            application = self._search_regex(
+                [r'data-application="([^"]+)"', r'"application"\s*:\s*"([^"]+)"'],
+                webpage, 'application', default=site + '_vod')
+            filename = self._search_regex(
+                [r'data-filename="([^"]+)"', r'"filename"\s*:\s*"([^"]+)"'],
+                webpage, 'filename')
+            playlist_url = 'http://vod.streamcloud.be/%s/_definst_/mp4:%s.mp4/playlist.m3u8' % (application, filename)
+
         formats = self._extract_wowza_formats(
             playlist_url, display_id, skip_protocols=['dash'])
         self._sort_formats(formats)

youtube_dl/extractor/xhamster.py

@@ -3,6 +3,7 @@ from __future__ import unicode_literals
 import re
 
 from .common import InfoExtractor
+from ..compat import compat_str
 from ..utils import (
     clean_html,
     dict_get,
@@ -14,12 +15,21 @@ from ..utils import (
 
 
 class XHamsterIE(InfoExtractor):
-    _VALID_URL = r'(?P<proto>https?)://(?:.+?\.)?xhamster\.com/movies/(?P<id>[0-9]+)/(?P<seo>.*?)\.html(?:\?.*)?'
+    _VALID_URL = r'''(?x)
+                    https?://
+                        (?:.+?\.)?xhamster\.com/
+                        (?:
+                            movies/(?P<id>\d+)/(?P<display_id>[^/]*)\.html|
+                            videos/(?P<display_id_2>[^/]*)-(?P<id_2>\d+)
+                        )
+                    '''
+
     _TESTS = [{
         'url': 'http://xhamster.com/movies/1509445/femaleagent_shy_beauty_takes_the_bait.html',
         'md5': '8281348b8d3c53d39fffb377d24eac4e',
         'info_dict': {
             'id': '1509445',
+            'display_id': 'femaleagent_shy_beauty_takes_the_bait',
             'ext': 'mp4',
             'title': 'FemaleAgent Shy beauty takes the bait',
             'upload_date': '20121014',
@@ -32,6 +42,7 @@ class XHamsterIE(InfoExtractor):
         'url': 'http://xhamster.com/movies/2221348/britney_spears_sexy_booty.html?hd',
         'info_dict': {
             'id': '2221348',
+            'display_id': 'britney_spears_sexy_booty',
             'ext': 'mp4',
             'title': 'Britney Spears  Sexy Booty',
             'upload_date': '20130914',
@@ -66,26 +77,18 @@ class XHamsterIE(InfoExtractor):
         # This video is visible for marcoalfa123456's friends only
         'url': 'https://it.xhamster.com/movies/7263980/la_mia_vicina.html',
         'only_matching': True,
+    }, {
+        # new URL schema
+        'url': 'https://pt.xhamster.com/videos/euro-pedal-pumping-7937821',
+        'only_matching': True,
     }]
 
     def _real_extract(self, url):
-        def extract_video_url(webpage, name):
-            return self._search_regex(
-                [r'''file\s*:\s*(?P<q>["'])(?P<mp4>.+?)(?P=q)''',
-                 r'''<a\s+href=(?P<q>["'])(?P<mp4>.+?)(?P=q)\s+class=["']mp4Thumb''',
-                 r'''<video[^>]+file=(?P<q>["'])(?P<mp4>.+?)(?P=q)[^>]*>'''],
-                webpage, name, group='mp4')
-
-        def is_hd(webpage):
-            return '<div class=\'icon iconHD\'' in webpage
-
         mobj = re.match(self._VALID_URL, url)
+        video_id = mobj.group('id') or mobj.group('id_2')
+        display_id = mobj.group('display_id') or mobj.group('display_id_2')
 
-        video_id = mobj.group('id')
-        seo = mobj.group('seo')
-        proto = mobj.group('proto')
-        mrss_url = '%s://xhamster.com/movies/%s/%s.html' % (proto, video_id, seo)
-        webpage = self._download_webpage(mrss_url, video_id)
+        webpage = self._download_webpage(url, video_id)
 
         error = self._html_search_regex(
             r'<div[^>]+id=["\']videoClosed["\'][^>]*>(.+?)</div>',
@@ -99,6 +102,39 @@ class XHamsterIE(InfoExtractor):
              r'<title[^>]*>(.+?)(?:,\s*[^,]*?\s*Porn\s*[^,]*?:\s*xHamster[^<]*| - xHamster\.com)</title>'],
             webpage, 'title')
 
+        formats = []
+        format_urls = set()
+
+        sources = self._parse_json(
+            self._search_regex(
+                r'sources\s*:\s*({.+?})\s*,?\s*\n', webpage, 'sources',
+                default='{}'),
+            video_id, fatal=False)
+        for format_id, format_url in sources.items():
+            if not isinstance(format_url, compat_str):
+                continue
+            if format_url in format_urls:
+                continue
+            format_urls.add(format_url)
+            formats.append({
+                'format_id': format_id,
+                'url': format_url,
+                'height': int_or_none(self._search_regex(
+                    r'^(\d+)[pP]', format_id, 'height', default=None))
+            })
+
+        video_url = self._search_regex(
+            [r'''file\s*:\s*(?P<q>["'])(?P<mp4>.+?)(?P=q)''',
+             r'''<a\s+href=(?P<q>["'])(?P<mp4>.+?)(?P=q)\s+class=["']mp4Thumb''',
+             r'''<video[^>]+file=(?P<q>["'])(?P<mp4>.+?)(?P=q)[^>]*>'''],
+            webpage, 'video url', group='mp4', default=None)
+        if video_url and video_url not in format_urls:
+            formats.append({
+                'url': video_url,
+            })
+
+        self._sort_formats(formats)
+
         # Only a few videos have an description
         mobj = re.search(r'<span>Description: </span>([^<]+)', webpage)
         description = mobj.group(1) if mobj else None
@@ -117,7 +153,8 @@ class XHamsterIE(InfoExtractor):
             webpage, 'thumbnail', fatal=False, group='thumbnail')
 
         duration = parse_duration(self._search_regex(
-            r'Runtime:\s*</span>\s*([\d:]+)', webpage,
+            [r'<[^<]+\bitemprop=["\']duration["\'][^<]+\bcontent=["\'](.+?)["\']',
+             r'Runtime:\s*</span>\s*([\d:]+)'], webpage,
             'duration', fatal=False))
 
         view_count = int_or_none(self._search_regex(
@@ -132,30 +169,6 @@ class XHamsterIE(InfoExtractor):
 
         age_limit = self._rta_search(webpage)
 
-        hd = is_hd(webpage)
-
-        format_id = 'hd' if hd else 'sd'
-
-        video_url = extract_video_url(webpage, format_id)
-        formats = [{
-            'url': video_url,
-            'format_id': 'hd' if hd else 'sd',
-            'preference': 1,
-        }]
-
-        if not hd:
-            mrss_url = self._search_regex(r'<link rel="canonical" href="([^"]+)', webpage, 'mrss_url')
-            webpage = self._download_webpage(mrss_url + '?hd', video_id, note='Downloading HD webpage')
-            if is_hd(webpage):
-                video_url = extract_video_url(webpage, 'hd')
-                formats.append({
-                    'url': video_url,
-                    'format_id': 'hd',
-                    'preference': 2,
-                })
-
-        self._sort_formats(formats)
-
         categories_html = self._search_regex(
             r'(?s)<table.+?(<span>Categories:.+?)</table>', webpage,
             'categories', default=None)
@@ -164,6 +177,7 @@ class XHamsterIE(InfoExtractor):
 
         return {
             'id': video_id,
+            'display_id': display_id,
             'title': title,
             'description': description,
             'upload_date': upload_date,

youtube_dl/extractor/xuite.py

@@ -1,14 +1,13 @@
 # coding: utf-8
 from __future__ import unicode_literals
 
-import base64
-
 from .common import InfoExtractor
-from ..compat import compat_urllib_parse_unquote
 from ..utils import (
     ExtractorError,
+    float_or_none,
+    get_element_by_attribute,
     parse_iso8601,
-    parse_duration,
+    remove_end,
 )
 
 
@@ -24,6 +23,7 @@ class XuiteIE(InfoExtractor):
             'id': '3860914',
             'ext': 'mp3',
             'title': '孤單南半球-歐德陽',
+            'description': '孤單南半球-歐德陽',
             'thumbnail': r're:^https?://.*\.jpg$',
             'duration': 247.246,
             'timestamp': 1314932940,
@@ -44,7 +44,7 @@ class XuiteIE(InfoExtractor):
             'duration': 596.458,
             'timestamp': 1454242500,
             'upload_date': '20160131',
-            'uploader': 'yan12125',
+            'uploader': '屁姥',
             'uploader_id': '12158353',
             'categories': ['個人短片'],
             'description': 'http://download.blender.org/peach/bigbuckbunny_movies/BigBuckBunny_320x180.mp4',
@@ -72,10 +72,10 @@ class XuiteIE(InfoExtractor):
         # from http://forgetfulbc.blogspot.com/2016/06/date.html
         'url': 'http://vlog.xuite.net/embed/cE1xbENoLTI3NDQ3MzM2LmZsdg==?ar=0&as=0',
         'info_dict': {
-            'id': 'cE1xbENoLTI3NDQ3MzM2LmZsdg==',
+            'id': '27447336',
             'ext': 'mp4',
             'title': '男女平權只是口號？專家解釋約會時男生是否該幫女生付錢 (中字)',
-            'description': 'md5:f0abdcb69df300f522a5442ef3146f2a',
+            'description': 'md5:1223810fa123b179083a3aed53574706',
             'timestamp': 1466160960,
             'upload_date': '20160617',
             'uploader': 'B.C. & Lowy',
@@ -86,29 +86,9 @@ class XuiteIE(InfoExtractor):
         'only_matching': True,
     }]
 
-    @staticmethod
-    def base64_decode_utf8(data):
-        return base64.b64decode(data.encode('utf-8')).decode('utf-8')
-
-    @staticmethod
-    def base64_encode_utf8(data):
-        return base64.b64encode(data.encode('utf-8')).decode('utf-8')
-
-    def _extract_flv_config(self, encoded_media_id):
-        flv_config = self._download_xml(
-            'http://vlog.xuite.net/flash/player?media=%s' % encoded_media_id,
-            'flv config')
-        prop_dict = {}
-        for prop in flv_config.findall('./property'):
-            prop_id = self.base64_decode_utf8(prop.attrib['id'])
-            # CDATA may be empty in flv config
-            if not prop.text:
-                continue
-            encoded_content = self.base64_decode_utf8(prop.text)
-            prop_dict[prop_id] = compat_urllib_parse_unquote(encoded_content)
-        return prop_dict
-
     def _real_extract(self, url):
+        # /play/ URLs provide embedded video URL and more metadata
+        url = url.replace('/embed/', '/play/')
         video_id = self._match_id(url)
 
         webpage = self._download_webpage(url, video_id)
@@ -121,51 +101,53 @@ class XuiteIE(InfoExtractor):
                 '%s returned error: %s' % (self.IE_NAME, error_msg),
                 expected=True)
 
-        encoded_media_id = self._search_regex(
-            r'attributes\.name\s*=\s*"([^"]+)"', webpage,
-            'encoded media id', default=None)
-        if encoded_media_id is None:
-            video_id = self._html_search_regex(
-                r'data-mediaid="(\d+)"', webpage, 'media id')
-            encoded_media_id = self.base64_encode_utf8(video_id)
-        flv_config = self._extract_flv_config(encoded_media_id)
+        media_info = self._parse_json(self._search_regex(
+            r'var\s+mediaInfo\s*=\s*({.*});', webpage, 'media info'), video_id)
 
-        FORMATS = {
-            'audio': 'mp3',
-            'video': 'mp4',
-        }
+        video_id = media_info['MEDIA_ID']
 
         formats = []
-        for format_tag in ('src', 'hq_src'):
-            video_url = flv_config.get(format_tag)
+        for key in ('html5Url', 'html5HQUrl'):
+            video_url = media_info.get(key)
             if not video_url:
                 continue
             format_id = self._search_regex(
-                r'\bq=(.+?)\b', video_url, 'format id', default=format_tag)
+                r'\bq=(.+?)\b', video_url, 'format id', default=None)
             formats.append({
                 'url': video_url,
-                'ext': FORMATS.get(flv_config['type'], 'mp4'),
+                'ext': 'mp4' if format_id.isnumeric() else format_id,
                 'format_id': format_id,
                 'height': int(format_id) if format_id.isnumeric() else None,
             })
         self._sort_formats(formats)
 
-        timestamp = flv_config.get('publish_datetime')
+        timestamp = media_info.get('PUBLISH_DATETIME')
         if timestamp:
             timestamp = parse_iso8601(timestamp + ' +0800', ' ')
 
-        category = flv_config.get('category')
+        category = media_info.get('catName')
         categories = [category] if category else []
 
+        uploader = media_info.get('NICKNAME')
+        uploader_url = None
+
+        author_div = get_element_by_attribute('itemprop', 'author', webpage)
+        if author_div:
+            uploader = uploader or self._html_search_meta('name', author_div)
+            uploader_url = self._html_search_regex(
+                r'<link[^>]+itemprop="url"[^>]+href="([^"]+)"', author_div,
+                'uploader URL', fatal=False)
+
         return {
             'id': video_id,
-            'title': flv_config['title'],
-            'description': flv_config.get('description'),
-            'thumbnail': flv_config.get('thumb'),
+            'title': media_info['TITLE'],
+            'description': remove_end(media_info.get('metaDesc'), ' (Xuite 影音)'),
+            'thumbnail': media_info.get('ogImageUrl'),
             'timestamp': timestamp,
-            'uploader': flv_config.get('author_name'),
-            'uploader_id': flv_config.get('author_id'),
-            'duration': parse_duration(flv_config.get('duration')),
+            'uploader': uploader,
+            'uploader_id': media_info.get('MEMBER_ID'),
+            'uploader_url': uploader_url,
+            'duration': float_or_none(media_info.get('MEDIA_DURATION'), 1000000),
             'categories': categories,
             'formats': formats,
         }

youtube_dl/extractor/yam.py

@@ -1,123 +0,0 @@
-# coding: utf-8
-from __future__ import unicode_literals
-
-import re
-
-from .common import InfoExtractor
-from ..compat import compat_urlparse
-from ..utils import (
-    float_or_none,
-    month_by_abbreviation,
-    ExtractorError,
-    get_element_by_attribute,
-)
-
-
-class YamIE(InfoExtractor):
-    IE_DESC = '蕃薯藤yam天空部落'
-    _VALID_URL = r'https?://mymedia\.yam\.com/m/(?P<id>\d+)'
-
-    _TESTS = [{
-        # An audio hosted on Yam
-        'url': 'http://mymedia.yam.com/m/2283921',
-        'md5': 'c011b8e262a52d5473d9c2e3c9963b9c',
-        'info_dict': {
-            'id': '2283921',
-            'ext': 'mp3',
-            'title': '發現 - 趙薇 京華煙雲主題曲',
-            'description': '發現 - 趙薇 京華煙雲主題曲',
-            'uploader_id': 'princekt',
-            'upload_date': '20080807',
-            'duration': 313.0,
-        }
-    }, {
-        # An external video hosted on YouTube
-        'url': 'http://mymedia.yam.com/m/3599430',
-        'md5': '03127cf10d8f35d120a9e8e52e3b17c6',
-        'info_dict': {
-            'id': 'CNpEoQlrIgA',
-            'ext': 'mp4',
-            'upload_date': '20150306',
-            'uploader': '新莊社大瑜伽社',
-            'description': 'md5:11e2e405311633ace874f2e6226c8b17',
-            'uploader_id': '2323agoy',
-            'title': '20090412陽明山二子坪-1',
-        },
-        'skip': 'Video does not exist',
-    }, {
-        'url': 'http://mymedia.yam.com/m/3598173',
-        'info_dict': {
-            'id': '3598173',
-            'ext': 'mp4',
-        },
-        'skip': 'cause Yam system error',
-    }, {
-        'url': 'http://mymedia.yam.com/m/3599437',
-        'info_dict': {
-            'id': '3599437',
-            'ext': 'mp4',
-        },
-        'skip': 'invalid YouTube URL',
-    }, {
-        'url': 'http://mymedia.yam.com/m/2373534',
-        'md5': '7ff74b91b7a817269d83796f8c5890b1',
-        'info_dict': {
-            'id': '2373534',
-            'ext': 'mp3',
-            'title': '林俊傑&蔡卓妍-小酒窩',
-            'description': 'md5:904003395a0fcce6cfb25028ff468420',
-            'upload_date': '20080928',
-            'uploader_id': 'onliner2',
-        }
-    }]
-
-    def _real_extract(self, url):
-        video_id = self._match_id(url)
-        page = self._download_webpage(url, video_id)
-
-        # Check for errors
-        system_msg = self._html_search_regex(
-            r'系統訊息(?:<br>|\n|\r)*([^<>]+)<br>', page, 'system message',
-            default=None)
-        if system_msg:
-            raise ExtractorError(system_msg, expected=True)
-
-        # Is it hosted externally on YouTube?
-        youtube_url = self._html_search_regex(
-            r'<embed src="(http://www.youtube.com/[^"]+)"',
-            page, 'YouTube url', default=None)
-        if youtube_url:
-            return self.url_result(youtube_url, 'Youtube')
-
-        title = self._html_search_regex(
-            r'<h1[^>]+class="heading"[^>]*>\s*(.+)\s*</h1>', page, 'title')
-
-        api_page = self._download_webpage(
-            'http://mymedia.yam.com/api/a/?pID=' + video_id, video_id,
-            note='Downloading API page')
-        api_result_obj = compat_urlparse.parse_qs(api_page)
-
-        info_table = get_element_by_attribute('class', 'info', page)
-        uploader_id = self._html_search_regex(
-            r'<!-- 發表作者 -->：[\n ]+<a href="/([a-z0-9]+)"',
-            info_table, 'uploader id', fatal=False)
-        mobj = re.search(r'<!-- 發表於 -->(?P<mon>[A-Z][a-z]{2})\s+' +
-                         r'(?P<day>\d{1,2}), (?P<year>\d{4})', page)
-        if mobj:
-            upload_date = '%s%02d%02d' % (
-                mobj.group('year'),
-                month_by_abbreviation(mobj.group('mon')),
-                int(mobj.group('day')))
-        else:
-            upload_date = None
-        duration = float_or_none(api_result_obj['totaltime'][0], scale=1000)
-
-        return {
-            'id': video_id,
-            'url': api_result_obj['mp3file'][0],
-            'title': title,
-            'description': self._html_search_meta('description', page),
-            'duration': duration,
-            'uploader_id': uploader_id,
-            'upload_date': upload_date,
-        }

youtube_dl/postprocessor/ffmpeg.py

@@ -542,7 +542,7 @@ class FFmpegFixupM3u8PP(FFmpegPostProcessor):
             temp_filename = prepend_extension(filename, 'temp')
 
             options = ['-c', 'copy', '-f', 'mp4', '-bsf:a', 'aac_adtstoasc']
-            self._downloader.to_screen('[ffmpeg] Fixing malformated aac bitstream in "%s"' % filename)
+            self._downloader.to_screen('[ffmpeg] Fixing malformed AAC bitstream in "%s"' % filename)
             self.run_ffmpeg(filename, temp_filename, options)
 
             os.remove(encodeFilename(filename))

youtube_dl/utils.py

@@ -365,9 +365,9 @@ def get_elements_by_attribute(attribute, value, html, escape_value=True):
     retlist = []
     for m in re.finditer(r'''(?xs)
         <([a-zA-Z0-9:._-]+)
-         (?:\s+[a-zA-Z0-9:._-]+(?:=[a-zA-Z0-9:._-]*|="[^"]*"|='[^']*'))*?
+         (?:\s+[a-zA-Z0-9:._-]+(?:=[a-zA-Z0-9:._-]*|="[^"]*"|='[^']*'|))*?
          \s+%s=['"]?%s['"]?
-         (?:\s+[a-zA-Z0-9:._-]+(?:=[a-zA-Z0-9:._-]*|="[^"]*"|='[^']*'))*?
+         (?:\s+[a-zA-Z0-9:._-]+(?:=[a-zA-Z0-9:._-]*|="[^"]*"|='[^']*'|))*?
         \s*>
         (?P<content>.*?)
         </\1>

youtube_dl/version.py

@@ -1,3 +1,3 @@
 from __future__ import unicode_literals
 
-__version__ = '2017.07.02'
+__version__ = '2017.07.09'