Merge branch 'master' into loom

youtube_dl/extractor/ccma.py

@@ -1,6 +1,7 @@
 # coding: utf-8
 from __future__ import unicode_literals
 
+import datetime
 import re
 
 from .common import InfoExtractor
@@ -8,8 +9,8 @@ from ..utils import (
     clean_html,
     int_or_none,
     parse_duration,
-    parse_iso8601,
     parse_resolution,
+    try_get,
     url_or_none,
 )
 
@@ -24,8 +25,9 @@ class CCMAIE(InfoExtractor):
             'ext': 'mp4',
             'title': 'L\'espot de La Marató de TV3',
             'description': 'md5:f12987f320e2f6e988e9908e4fe97765',
-            'timestamp': 1470918540,
-            'upload_date': '20160811',
+            'timestamp': 1478608140,
+            'upload_date': '20161108',
+            'age_limit': 0,
         }
     }, {
         'url': 'http://www.ccma.cat/catradio/alacarta/programa/el-consell-de-savis-analitza-el-derbi/audio/943685/',
@@ -35,8 +37,24 @@ class CCMAIE(InfoExtractor):
             'ext': 'mp3',
             'title': 'El Consell de Savis analitza el derbi',
             'description': 'md5:e2a3648145f3241cb9c6b4b624033e53',
-            'upload_date': '20171205',
-            'timestamp': 1512507300,
+            'upload_date': '20170512',
+            'timestamp': 1494622500,
+            'vcodec': 'none',
+            'categories': ['Esports'],
+        }
+    }, {
+        'url': 'http://www.ccma.cat/tv3/alacarta/crims/crims-josep-tallada-lespereu-me-capitol-1/video/6031387/',
+        'md5': 'b43c3d3486f430f3032b5b160d80cbc3',
+        'info_dict': {
+            'id': '6031387',
+            'ext': 'mp4',
+            'title': 'Crims - Josep Talleda, l\'"Espereu-me" (capítol 1)',
+            'description': 'md5:7cbdafb640da9d0d2c0f62bad1e74e60',
+            'timestamp': 1582577700,
+            'upload_date': '20200224',
+            'subtitles': 'mincount:4',
+            'age_limit': 16,
+            'series': 'Crims',
         }
     }]
 
@@ -72,17 +90,27 @@ class CCMAIE(InfoExtractor):
 
         informacio = media['informacio']
         title = informacio['titol']
-        durada = informacio.get('durada', {})
+        durada = informacio.get('durada') or {}
         duration = int_or_none(durada.get('milisegons'), 1000) or parse_duration(durada.get('text'))
-        timestamp = parse_iso8601(informacio.get('data_emissio', {}).get('utc'))
+        tematica = try_get(informacio, lambda x: x['tematica']['text'])
+
+        timestamp = None
+        data_utc = try_get(informacio, lambda x: x['data_emissio']['utc'])
+        try:
+            timestamp = datetime.datetime.strptime(
+                data_utc, '%Y-%d-%mT%H:%M:%S%z').timestamp()
+        except TypeError:
+            pass
 
         subtitles = {}
-        subtitols = media.get('subtitols', {})
-        if subtitols:
-            sub_url = subtitols.get('url')
+        subtitols = media.get('subtitols') or []
+        if isinstance(subtitols, dict):
+            subtitols = [subtitols]
+        for st in subtitols:
+            sub_url = st.get('url')
             if sub_url:
                 subtitles.setdefault(
-                    subtitols.get('iso') or subtitols.get('text') or 'ca', []).append({
+                    st.get('iso') or st.get('text') or 'ca', []).append({
                         'url': sub_url,
                     })
 
@@ -97,6 +125,16 @@ class CCMAIE(InfoExtractor):
                     'height': int_or_none(imatges.get('alcada')),
                 }]
 
+        age_limit = None
+        codi_etic = try_get(informacio, lambda x: x['codi_etic']['id'])
+        if codi_etic:
+            codi_etic_s = codi_etic.split('_')
+            if len(codi_etic_s) == 2:
+                if codi_etic_s[1] == 'TP':
+                    age_limit = 0
+                else:
+                    age_limit = int_or_none(codi_etic_s[1])
+
         return {
             'id': media_id,
             'title': title,
@@ -106,4 +144,9 @@ class CCMAIE(InfoExtractor):
             'thumbnails': thumbnails,
             'subtitles': subtitles,
             'formats': formats,
+            'age_limit': age_limit,
+            'alt_title': informacio.get('titol_complet'),
+            'episode_number': int_or_none(informacio.get('capitol')),
+            'categories': [tematica] if tematica else None,
+            'series': informacio.get('programa'),
         }

youtube_dl/extractor/common.py

@@ -2064,7 +2064,7 @@ class InfoExtractor(object):
             })
         return entries
 
-    def _extract_mpd_formats(self, mpd_url, video_id, mpd_id=None, note=None, errnote=None, fatal=True, formats_dict={}, data=None, headers={}, query={}):
+    def _extract_mpd_formats(self, mpd_url, video_id, mpd_id=None, note=None, errnote=None, fatal=True, data=None, headers={}, query={}):
         res = self._download_xml_handle(
             mpd_url, video_id,
             note=note or 'Downloading MPD manifest',
@@ -2078,10 +2078,9 @@ class InfoExtractor(object):
         mpd_base_url = base_url(urlh.geturl())
 
         return self._parse_mpd_formats(
-            mpd_doc, mpd_id=mpd_id, mpd_base_url=mpd_base_url,
-            formats_dict=formats_dict, mpd_url=mpd_url)
+            mpd_doc, mpd_id, mpd_base_url, mpd_url)
 
-    def _parse_mpd_formats(self, mpd_doc, mpd_id=None, mpd_base_url='', formats_dict={}, mpd_url=None):
+    def _parse_mpd_formats(self, mpd_doc, mpd_id=None, mpd_base_url='', mpd_url=None):
         """
         Parse formats from MPD manifest.
         References:
@@ -2359,15 +2358,7 @@ class InfoExtractor(object):
                         else:
                             # Assuming direct URL to unfragmented media.
                             f['url'] = base_url
-
-                        # According to [1, 5.3.5.2, Table 7, page 35] @id of Representation
-                        # is not necessarily unique within a Period thus formats with
-                        # the same `format_id` are quite possible. There are numerous examples
-                        # of such manifests (see https://github.com/ytdl-org/youtube-dl/issues/15111,
-                        # https://github.com/ytdl-org/youtube-dl/issues/13919)
-                        full_info = formats_dict.get(representation_id, {}).copy()
-                        full_info.update(f)
-                        formats.append(full_info)
+                        formats.append(f)
                     else:
                         self.report_warning('Unknown MIME type %s in DASH manifest' % mime_type)
         return formats

youtube_dl/extractor/egghead.py

@@ -12,7 +12,14 @@ from ..utils import (
 )
 
 
-class EggheadCourseIE(InfoExtractor):
+class EggheadBaseIE(InfoExtractor):
+    def _call_api(self, path, video_id, resource, fatal=True):
+        return self._download_json(
+            'https://app.egghead.io/api/v1/' + path,
+            video_id, 'Downloading %s JSON' % resource, fatal=fatal)
+
+
+class EggheadCourseIE(EggheadBaseIE):
     IE_DESC = 'egghead.io course'
     IE_NAME = 'egghead:course'
     _VALID_URL = r'https://egghead\.io/courses/(?P<id>[^/?#&]+)'
@@ -28,10 +35,9 @@ class EggheadCourseIE(InfoExtractor):
 
     def _real_extract(self, url):
         playlist_id = self._match_id(url)
-
-        lessons = self._download_json(
-            'https://egghead.io/api/v1/series/%s/lessons' % playlist_id,
-            playlist_id, 'Downloading course lessons JSON')
+        series_path = 'series/' + playlist_id
+        lessons = self._call_api(
+            series_path + '/lessons', playlist_id, 'course lessons')
 
         entries = []
         for lesson in lessons:
@@ -44,9 +50,8 @@ class EggheadCourseIE(InfoExtractor):
             entries.append(self.url_result(
                 lesson_url, ie=EggheadLessonIE.ie_key(), video_id=lesson_id))
 
-        course = self._download_json(
-            'https://egghead.io/api/v1/series/%s' % playlist_id,
-            playlist_id, 'Downloading course JSON', fatal=False) or {}
+        course = self._call_api(
+            series_path, playlist_id, 'course', False) or {}
 
         playlist_id = course.get('id')
         if playlist_id:
@@ -57,7 +62,7 @@ class EggheadCourseIE(InfoExtractor):
             course.get('description'))
 
 
-class EggheadLessonIE(InfoExtractor):
+class EggheadLessonIE(EggheadBaseIE):
     IE_DESC = 'egghead.io lesson'
     IE_NAME = 'egghead:lesson'
     _VALID_URL = r'https://egghead\.io/(?:api/v1/)?lessons/(?P<id>[^/?#&]+)'
@@ -74,7 +79,7 @@ class EggheadLessonIE(InfoExtractor):
             'upload_date': '20161209',
             'duration': 304,
             'view_count': 0,
-            'tags': ['javascript', 'free'],
+            'tags': 'count:2',
         },
         'params': {
             'skip_download': True,
@@ -88,8 +93,8 @@ class EggheadLessonIE(InfoExtractor):
     def _real_extract(self, url):
         display_id = self._match_id(url)
 
-        lesson = self._download_json(
-            'https://egghead.io/api/v1/lessons/%s' % display_id, display_id)
+        lesson = self._call_api(
+            'lessons/' + display_id, display_id, 'lesson')
 
         lesson_id = compat_str(lesson['id'])
         title = lesson['title']

youtube_dl/extractor/extractors.py

@@ -1400,7 +1400,6 @@ from .vidme import (
     VidmeUserIE,
     VidmeUserLikesIE,
 )
-from .vidzi import VidziIE
 from .vier import VierIE, VierVideosIE
 from .viewlift import (
     ViewLiftIE,

youtube_dl/extractor/vidio.py

@@ -4,7 +4,13 @@ from __future__ import unicode_literals
 import re
 
 from .common import InfoExtractor
-from ..utils import int_or_none
+from ..utils import (
+    int_or_none,
+    parse_iso8601,
+    str_or_none,
+    strip_or_none,
+    try_get,
+)
 
 
 class VidioIE(InfoExtractor):
@@ -21,57 +27,63 @@ class VidioIE(InfoExtractor):
             'thumbnail': r're:^https?://.*\.jpg$',
             'duration': 149,
             'like_count': int,
+            'uploader': 'TWELVE Pic',
+            'timestamp': 1444902800,
+            'upload_date': '20151015',
+            'uploader_id': 'twelvepictures',
+            'channel': 'Cover Music Video',
+            'channel_id': '280236',
+            'view_count': int,
+            'dislike_count': int,
+            'comment_count': int,
+            'tags': 'count:4',
         },
     }, {
         'url': 'https://www.vidio.com/watch/77949-south-korea-test-fires-missile-that-can-strike-all-of-the-north',
         'only_matching': True,
     }]
 
+    def _real_initialize(self):
+        self._api_key = self._download_json(
+            'https://www.vidio.com/auth', None, data=b'')['api_key']
+
     def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
-        video_id, display_id = mobj.group('id', 'display_id')
+        video_id, display_id = re.match(self._VALID_URL, url).groups()
+        data = self._download_json(
+            'https://api.vidio.com/videos/' + video_id, display_id, headers={
+                'Content-Type': 'application/vnd.api+json',
+                'X-API-KEY': self._api_key,
+            })
+        video = data['videos'][0]
+        title = video['title'].strip()
 
-        webpage = self._download_webpage(url, display_id)
-
-        title = self._og_search_title(webpage)
-
-        m3u8_url, duration, thumbnail = [None] * 3
-
-        clips = self._parse_json(
-            self._html_search_regex(
-                r'data-json-clips\s*=\s*(["\'])(?P<data>\[.+?\])\1',
-                webpage, 'video data', default='[]', group='data'),
-            display_id, fatal=False)
-        if clips:
-            clip = clips[0]
-            m3u8_url = clip.get('sources', [{}])[0].get('file')
-            duration = clip.get('clip_duration')
-            thumbnail = clip.get('image')
-
-        m3u8_url = m3u8_url or self._search_regex(
-            r'data(?:-vjs)?-clip-hls-url=(["\'])(?P<url>(?:(?!\1).)+)\1',
-            webpage, 'hls url', group='url')
         formats = self._extract_m3u8_formats(
-            m3u8_url, display_id, 'mp4', entry_protocol='m3u8_native')
+            data['clips'][0]['hls_url'], display_id, 'mp4', 'm3u8_native')
         self._sort_formats(formats)
 
-        duration = int_or_none(duration or self._search_regex(
-            r'data-video-duration=(["\'])(?P<duration>\d+)\1', webpage,
-            'duration', fatal=False, group='duration'))
-        thumbnail = thumbnail or self._og_search_thumbnail(webpage)
-
-        like_count = int_or_none(self._search_regex(
-            (r'<span[^>]+data-comment-vote-count=["\'](\d+)',
-             r'<span[^>]+class=["\'].*?\blike(?:__|-)count\b.*?["\'][^>]*>\s*(\d+)'),
-            webpage, 'like count', fatal=False))
+        get_first = lambda x: try_get(data, lambda y: y[x + 's'][0], dict) or {}
+        channel = get_first('channel')
+        user = get_first('user')
+        username = user.get('username')
+        get_count = lambda x: int_or_none(video.get('total_' + x))
 
         return {
             'id': video_id,
             'display_id': display_id,
             'title': title,
-            'description': self._og_search_description(webpage),
-            'thumbnail': thumbnail,
-            'duration': duration,
-            'like_count': like_count,
+            'description': strip_or_none(video.get('description')),
+            'thumbnail': video.get('image_url_medium'),
+            'duration': int_or_none(video.get('duration')),
+            'like_count': get_count('likes'),
             'formats': formats,
+            'uploader': user.get('name'),
+            'timestamp': parse_iso8601(video.get('created_at')),
+            'uploader_id': username,
+            'uploader_url': 'https://www.vidio.com/@' + username if username else None,
+            'channel': channel.get('name'),
+            'channel_id': str_or_none(channel.get('id')),
+            'view_count': get_count('view_count'),
+            'dislike_count': get_count('dislikes'),
+            'comment_count': get_count('comments'),
+            'tags': video.get('tag_list'),
         }

youtube_dl/extractor/vidzi.py

@@ -1,68 +0,0 @@
-# coding: utf-8
-from __future__ import unicode_literals
-
-import re
-
-from .common import InfoExtractor
-from ..utils import (
-    decode_packed_codes,
-    js_to_json,
-    NO_DEFAULT,
-    PACKED_CODES_RE,
-)
-
-
-class VidziIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?vidzi\.(?:tv|cc|si|nu)/(?:embed-)?(?P<id>[0-9a-zA-Z]+)'
-    _TESTS = [{
-        'url': 'http://vidzi.tv/cghql9yq6emu.html',
-        'md5': '4f16c71ca0c8c8635ab6932b5f3f1660',
-        'info_dict': {
-            'id': 'cghql9yq6emu',
-            'ext': 'mp4',
-            'title': 'youtube-dl test video  1\\\\2\'3/4<5\\\\6ä7↭',
-        },
-        'params': {
-            # m3u8 download
-            'skip_download': True,
-        },
-    }, {
-        'url': 'http://vidzi.tv/embed-4z2yb0rzphe9-600x338.html',
-        'only_matching': True,
-    }, {
-        'url': 'http://vidzi.cc/cghql9yq6emu.html',
-        'only_matching': True,
-    }, {
-        'url': 'https://vidzi.si/rph9gztxj1et.html',
-        'only_matching': True,
-    }, {
-        'url': 'http://vidzi.nu/cghql9yq6emu.html',
-        'only_matching': True,
-    }]
-
-    def _real_extract(self, url):
-        video_id = self._match_id(url)
-
-        webpage = self._download_webpage(
-            'http://vidzi.tv/%s' % video_id, video_id)
-        title = self._html_search_regex(
-            r'(?s)<h2 class="video-title">(.*?)</h2>', webpage, 'title')
-
-        codes = [webpage]
-        codes.extend([
-            decode_packed_codes(mobj.group(0)).replace('\\\'', '\'')
-            for mobj in re.finditer(PACKED_CODES_RE, webpage)])
-        for num, code in enumerate(codes, 1):
-            jwplayer_data = self._parse_json(
-                self._search_regex(
-                    r'setup\(([^)]+)\)', code, 'jwplayer data',
-                    default=NO_DEFAULT if num == len(codes) else '{}'),
-                video_id, transform_source=lambda s: js_to_json(
-                    re.sub(r'\s*\+\s*window\[.+?\]', '', s)))
-            if jwplayer_data:
-                break
-
-        info_dict = self._parse_jwplayer_data(jwplayer_data, video_id, require_title=False)
-        info_dict['title'] = title
-
-        return info_dict