X-Git-Url: http://git.bitcoin.ninja/index.cgi?a=blobdiff_plain;f=youtube_dl%2Fextractor%2Fyoutube.py;h=ec12e313cddbdcebdeb91a26fd97abc8ad97e13f;hb=9cafc3fd8b54b9b91a145cddf9e4db0bd59e1b5f;hp=ca40de522bc5e341f2ac269db997a0e73914c127;hpb=4f06c1c9fcbfbc74b81b5fa89a616914b5ce5aad;p=youtube-dl

diff --git a/youtube_dl/extractor/youtube.py b/youtube_dl/extractor/youtube.py
index ca40de522..ec12e313c 100644
--- a/youtube_dl/extractor/youtube.py
+++ b/youtube_dl/extractor/youtube.py
@@ -317,60 +317,60 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
         '151': {'ext': 'mp4', 'height': 72, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 24, 'vcodec': 'h264', 'preference': -10},
 
         # DASH mp4 video
-        '133': {'ext': 'mp4', 'height': 240, 'format_note': 'DASH video', 'vcodec': 'h264', 'preference': -40},
-        '134': {'ext': 'mp4', 'height': 360, 'format_note': 'DASH video', 'vcodec': 'h264', 'preference': -40},
-        '135': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'h264', 'preference': -40},
-        '136': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'h264', 'preference': -40},
-        '137': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'h264', 'preference': -40},
-        '138': {'ext': 'mp4', 'format_note': 'DASH video', 'vcodec': 'h264', 'preference': -40},  # Height can vary (https://github.com/rg3/youtube-dl/issues/4559)
-        '160': {'ext': 'mp4', 'height': 144, 'format_note': 'DASH video', 'vcodec': 'h264', 'preference': -40},
-        '212': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'h264', 'preference': -40},
-        '264': {'ext': 'mp4', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'h264', 'preference': -40},
-        '298': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'h264', 'fps': 60, 'preference': -40},
-        '299': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'h264', 'fps': 60, 'preference': -40},
-        '266': {'ext': 'mp4', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'h264', 'preference': -40},
+        '133': {'ext': 'mp4', 'height': 240, 'format_note': 'DASH video', 'vcodec': 'h264'},
+        '134': {'ext': 'mp4', 'height': 360, 'format_note': 'DASH video', 'vcodec': 'h264'},
+        '135': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'h264'},
+        '136': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'h264'},
+        '137': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'h264'},
+        '138': {'ext': 'mp4', 'format_note': 'DASH video', 'vcodec': 'h264'},  # Height can vary (https://github.com/rg3/youtube-dl/issues/4559)
+        '160': {'ext': 'mp4', 'height': 144, 'format_note': 'DASH video', 'vcodec': 'h264'},
+        '212': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'h264'},
+        '264': {'ext': 'mp4', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'h264'},
+        '298': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'h264', 'fps': 60},
+        '299': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'h264', 'fps': 60},
+        '266': {'ext': 'mp4', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'h264'},
 
         # Dash mp4 audio
-        '139': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 48, 'preference': -50, 'container': 'm4a_dash'},
-        '140': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 128, 'preference': -50, 'container': 'm4a_dash'},
-        '141': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 256, 'preference': -50, 'container': 'm4a_dash'},
-        '256': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'preference': -50, 'container': 'm4a_dash'},
-        '258': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'preference': -50, 'container': 'm4a_dash'},
-        '325': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'dtse', 'preference': -50, 'container': 'm4a_dash'},
-        '328': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'ec-3', 'preference': -50, 'container': 'm4a_dash'},
+        '139': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 48, 'container': 'm4a_dash'},
+        '140': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 128, 'container': 'm4a_dash'},
+        '141': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 256, 'container': 'm4a_dash'},
+        '256': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'container': 'm4a_dash'},
+        '258': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'container': 'm4a_dash'},
+        '325': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'dtse', 'container': 'm4a_dash'},
+        '328': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'ec-3', 'container': 'm4a_dash'},
 
         # Dash webm
-        '167': {'ext': 'webm', 'height': 360, 'width': 640, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8', 'preference': -40},
-        '168': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8', 'preference': -40},
-        '169': {'ext': 'webm', 'height': 720, 'width': 1280, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8', 'preference': -40},
-        '170': {'ext': 'webm', 'height': 1080, 'width': 1920, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8', 'preference': -40},
-        '218': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8', 'preference': -40},
-        '219': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8', 'preference': -40},
-        '278': {'ext': 'webm', 'height': 144, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp9', 'preference': -40},
-        '242': {'ext': 'webm', 'height': 240, 'format_note': 'DASH video', 'vcodec': 'vp9', 'preference': -40},
-        '243': {'ext': 'webm', 'height': 360, 'format_note': 'DASH video', 'vcodec': 'vp9', 'preference': -40},
-        '244': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9', 'preference': -40},
-        '245': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9', 'preference': -40},
-        '246': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9', 'preference': -40},
-        '247': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'vp9', 'preference': -40},
-        '248': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'vp9', 'preference': -40},
-        '271': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'vp9', 'preference': -40},
+        '167': {'ext': 'webm', 'height': 360, 'width': 640, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
+        '168': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
+        '169': {'ext': 'webm', 'height': 720, 'width': 1280, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
+        '170': {'ext': 'webm', 'height': 1080, 'width': 1920, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
+        '218': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
+        '219': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
+        '278': {'ext': 'webm', 'height': 144, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp9'},
+        '242': {'ext': 'webm', 'height': 240, 'format_note': 'DASH video', 'vcodec': 'vp9'},
+        '243': {'ext': 'webm', 'height': 360, 'format_note': 'DASH video', 'vcodec': 'vp9'},
+        '244': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9'},
+        '245': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9'},
+        '246': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9'},
+        '247': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'vp9'},
+        '248': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'vp9'},
+        '271': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'vp9'},
         # itag 272 videos are either 3840x2160 (e.g. RtoitU2A-3E) or 7680x4320 (sLprVF6d7Ug)
-        '272': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9', 'preference': -40},
-        '302': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60, 'preference': -40},
-        '303': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60, 'preference': -40},
-        '308': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60, 'preference': -40},
-        '313': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9', 'preference': -40},
-        '315': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60, 'preference': -40},
+        '272': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9'},
+        '302': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
+        '303': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
+        '308': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
+        '313': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9'},
+        '315': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
 
         # Dash webm audio
-        '171': {'ext': 'webm', 'acodec': 'vorbis', 'format_note': 'DASH audio', 'abr': 128, 'preference': -50},
-        '172': {'ext': 'webm', 'acodec': 'vorbis', 'format_note': 'DASH audio', 'abr': 256, 'preference': -50},
+        '171': {'ext': 'webm', 'acodec': 'vorbis', 'format_note': 'DASH audio', 'abr': 128},
+        '172': {'ext': 'webm', 'acodec': 'vorbis', 'format_note': 'DASH audio', 'abr': 256},
 
         # Dash webm audio with opus inside
-        '249': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 50, 'preference': -50},
-        '250': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 70, 'preference': -50},
-        '251': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 160, 'preference': -50},
+        '249': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 50},
+        '250': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 70},
+        '251': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 160},
 
         # RTMP (unnamed)
         '_rtmp': {'protocol': 'rtmp'},
@@ -963,7 +963,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
 
     def _extract_signature_function(self, video_id, player_url, example_sig):
         id_m = re.match(
-            r'.*?-(?P<id>[a-zA-Z0-9_-]+)(?:/watch_as3|/html5player(?:-new)?|/base)?\.(?P<ext>[a-z]+)$',
+            r'.*?-(?P<id>[a-zA-Z0-9_-]+)(?:/watch_as3|/html5player(?:-new)?|(?:/[a-z]{2}_[A-Z]{2})?/base)?\.(?P<ext>[a-z]+)$',
             player_url)
         if not id_m:
             raise ExtractorError('Cannot identify player %r' % player_url)
@@ -1253,25 +1253,39 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
         video_id = mobj.group(2)
         return video_id
 
-    def _extract_from_m3u8(self, manifest_url, video_id):
-        url_map = {}
-
-        def _get_urls(_manifest):
-            lines = _manifest.split('\n')
-            urls = filter(lambda l: l and not l.startswith('#'),
-                          lines)
-            return urls
-        manifest = self._download_webpage(manifest_url, video_id, 'Downloading formats manifest')
-        formats_urls = _get_urls(manifest)
-        for format_url in formats_urls:
-            itag = self._search_regex(r'itag/(\d+?)/', format_url, 'itag')
-            url_map[itag] = format_url
-        return url_map
-
     def _extract_annotations(self, video_id):
         url = 'https://www.youtube.com/annotations_invideo?features=1&legacy=1&video_id=%s' % video_id
         return self._download_webpage(url, video_id, note='Searching for annotations.', errnote='Unable to download video annotations.')
 
+    @staticmethod
+    def _extract_chapters(description, duration):
+        if not description:
+            return None
+        chapter_lines = re.findall(
+            r'(?:^|<br\s*/>)([^<]*<a[^>]+onclick=["\']yt\.www\.watch\.player\.seekTo[^>]+>(\d{1,2}:\d{1,2}(?::\d{1,2})?)</a>[^>]*)(?=$|<br\s*/>)',
+            description)
+        if not chapter_lines:
+            return None
+        chapters = []
+        for next_num, (chapter_line, time_point) in enumerate(
+                chapter_lines, start=1):
+            start_time = parse_duration(time_point)
+            if start_time is None:
+                continue
+            end_time = (duration if next_num == len(chapter_lines)
+                        else parse_duration(chapter_lines[next_num][1]))
+            if end_time is None:
+                continue
+            chapter_title = re.sub(
+                r'<a[^>]+>[^<]+</a>', '', chapter_line).strip(' \t-')
+            chapter_title = re.sub(r'\s+', ' ', chapter_title)
+            chapters.append({
+                'start_time': start_time,
+                'end_time': end_time,
+                'title': chapter_title,
+            })
+        return chapters
+
     def _real_extract(self, url):
         url, smuggled_data = unsmuggle_url(url, {})
 
@@ -1414,9 +1428,9 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
             video_title = '_'
 
         # description
-        video_description = get_element_by_id("eow-description", video_webpage)
+        description_original = video_description = get_element_by_id("eow-description", video_webpage)
         if video_description:
-            video_description = re.sub(r'''(?x)
+            description_original = video_description = re.sub(r'''(?x)
                 <a\s+
                     (?:[a-zA-Z-]+="[^"]*"\s+)*?
                     (?:title|href)="([^"]+)"\s+
@@ -1573,18 +1587,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
         if self._downloader.params.get('writeannotations', False):
             video_annotations = self._extract_annotations(video_id)
 
-        def _map_to_format_list(urlmap):
-            formats = []
-            for itag, video_real_url in urlmap.items():
-                dct = {
-                    'format_id': itag,
-                    'url': video_real_url,
-                    'player_url': player_url,
-                }
-                if itag in self._formats:
-                    dct.update(self._formats[itag])
-                formats.append(dct)
-            return formats
+        chapters = self._extract_chapters(description_original, video_duration)
 
         if 'conn' in video_info and video_info['conn'][0].startswith('rtmp'):
             self.report_rtmp_download()
@@ -1657,7 +1660,8 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
                                 player_desc = 'flash player %s' % player_version
                             else:
                                 player_version = self._search_regex(
-                                    [r'html5player-([^/]+?)(?:/html5player(?:-new)?)?\.js', r'(?:www|player)-([^/]+)/base\.js'],
+                                    [r'html5player-([^/]+?)(?:/html5player(?:-new)?)?\.js',
+                                     r'(?:www|player)-([^/]+)(?:/[a-z]{2}_[A-Z]{2})?/base\.js'],
                                     player_url,
                                     'html5 player', fatal=False)
                                 player_desc = 'html5 player %s' % player_version
@@ -1718,11 +1722,22 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
                 formats.append(dct)
         elif video_info.get('hlsvp'):
             manifest_url = video_info['hlsvp'][0]
-            url_map = self._extract_from_m3u8(manifest_url, video_id)
-            formats = _map_to_format_list(url_map)
-            # Accept-Encoding header causes failures in live streams on Youtube and Youtube Gaming
-            for a_format in formats:
+            formats = []
+            m3u8_formats = self._extract_m3u8_formats(
+                manifest_url, video_id, 'mp4', fatal=False)
+            for a_format in m3u8_formats:
+                itag = self._search_regex(
+                    r'/itag/(\d+)/', a_format['url'], 'itag', default=None)
+                if itag:
+                    a_format['format_id'] = itag
+                    if itag in self._formats:
+                        dct = self._formats[itag].copy()
+                        dct.update(a_format)
+                        a_format = dct
+                a_format['player_url'] = player_url
+                # Accept-Encoding header causes failures in live streams on Youtube and Youtube Gaming
                 a_format.setdefault('http_headers', {})['Youtubedl-no-compression'] = 'True'
+                formats.append(a_format)
         else:
             unavailable_message = self._html_search_regex(
                 r'(?s)<h1[^>]+id="unavailable-message"[^>]*>(.+?)</h1>',
@@ -1806,6 +1821,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
             'duration': video_duration,
             'age_limit': 18 if age_gate else 0,
             'annotations': video_annotations,
+            'chapters': chapters,
             'webpage_url': proto + '://www.youtube.com/watch?v=%s' % video_id,
             'view_count': view_count,
             'like_count': like_count,