debian/copyright: Use HTTPS for upstream's source.

[youtubedl] / youtube_dl / extractor / ceskatelevize.py

diff --git a/youtube_dl/extractor/ceskatelevize.py b/youtube_dl/extractor/ceskatelevize.py
index 97feb6704075831fb8b5ef95a547428ecc57ec3f..e250de18ceb555e4750df54fcb7de1f1b92d6d49 100644 (file)
--- a/youtube_dl/extractor/ceskatelevize.py
+++ b/youtube_dl/extractor/ceskatelevize.py
@@ -1,126 +1,279 @@
-# -*- coding: utf-8 -*-
+# coding: utf-8

 from __future__ import unicode_literals
 
 import re
 
 from .common import InfoExtractor
 from __future__ import unicode_literals
 
 import re
 
 from .common import InfoExtractor

-from ..utils import (
-    compat_urllib_request,
-    compat_urllib_parse,
+from ..compat import (
+    compat_urllib_parse_unquote,

     compat_urllib_parse_urlparse,
     compat_urllib_parse_urlparse,

+)
+from ..utils import (

     ExtractorError,
     ExtractorError,

+    float_or_none,
+    sanitized_Request,
+    unescapeHTML,
+    urlencode_postdata,
+    USER_AGENTS,

 )
 
 
 class CeskaTelevizeIE(InfoExtractor):
 )
 
 
 class CeskaTelevizeIE(InfoExtractor):

-    _VALID_URL = r'https?://www\.ceskatelevize\.cz/(porady|ivysilani)/(.+/)?(?P<id>[^?#]+)'
-
-    _TESTS = [
-        {
-            'url': 'http://www.ceskatelevize.cz/ivysilani/10532695142-prvni-republika/213512120230004-spanelska-chripka',
-            'info_dict': {
-                'id': '213512120230004',
-                'ext': 'flv',
-                'title': 'První republika: Španělská chřipka',
-                'duration': 3107.4,
-            },
-            'params': {
-                'skip_download': True,  # requires rtmpdump
-            },
-            'skip': 'Works only from Czech Republic.',
+    _VALID_URL = r'https?://(?:www\.)?ceskatelevize\.cz/ivysilani/(?:[^/?#&]+/)*(?P<id>[^/#?]+)'
+    _TESTS = [{
+        'url': 'http://www.ceskatelevize.cz/ivysilani/ivysilani/10441294653-hyde-park-civilizace/214411058091220',
+        'info_dict': {
+            'id': '61924494877246241',
+            'ext': 'mp4',
+            'title': 'Hyde Park Civilizace: Život v Grónsku',
+            'description': 'md5:3fec8f6bb497be5cdb0c9e8781076626',
+            'thumbnail': r're:^https?://.*\.jpg',
+            'duration': 3350,

         },
         },

-        {
-            'url': 'http://www.ceskatelevize.cz/ivysilani/1030584952-tsatsiki-maminka-a-policajt',
-            'info_dict': {
-                'id': '20138143440',
-                'ext': 'flv',
-                'title': 'Tsatsiki, maminka a policajt',
-                'duration': 6754.1,
-            },
-            'params': {
-                'skip_download': True,  # requires rtmpdump
-            },
-            'skip': 'Works only from Czech Republic.',
+        'params': {
+            # m3u8 download
+            'skip_download': True,

         },
         },

-        {
-            'url': 'http://www.ceskatelevize.cz/ivysilani/10532695142-prvni-republika/bonus/14716-zpevacka-z-duparny-bobina',
-            'info_dict': {
-                'id': '14716',
-                'ext': 'flv',
-                'title': 'První republika: Zpěvačka z Dupárny Bobina',
-                'duration': 90,
-            },
-            'params': {
-                'skip_download': True,  # requires rtmpdump
-            },
+    }, {
+        'url': 'http://www.ceskatelevize.cz/ivysilani/10441294653-hyde-park-civilizace/215411058090502/bonus/20641-bonus-01-en',
+        'info_dict': {
+            'id': '61924494877028507',
+            'ext': 'mp4',
+            'title': 'Hyde Park Civilizace: Bonus 01 - En',
+            'description': 'English Subtittles',
+            'thumbnail': r're:^https?://.*\.jpg',
+            'duration': 81.3,

         },
         },

-    ]
+        'params': {
+            # m3u8 download
+            'skip_download': True,
+        },
+    }, {
+        # live stream
+        'url': 'http://www.ceskatelevize.cz/ivysilani/zive/ct4/',
+        'info_dict': {
+            'id': 402,
+            'ext': 'mp4',
+            'title': r're:^ČT Sport \d{4}-\d{2}-\d{2} \d{2}:\d{2}$',
+            'is_live': True,
+        },
+        'params': {
+            # m3u8 download
+            'skip_download': True,
+        },
+        'skip': 'Georestricted to Czech Republic',
+    }, {
+        'url': 'http://www.ceskatelevize.cz/ivysilani/embed/iFramePlayer.php?hash=d6a3e1370d2e4fa76296b90bad4dfc19673b641e&IDEC=217 562 22150/0004&channelID=1&width=100%25',
+        'only_matching': True,
+    }]

 
     def _real_extract(self, url):
 
     def _real_extract(self, url):

-        url = url.replace('/porady/', '/ivysilani/').replace('/video/', '')
+        playlist_id = self._match_id(url)

 
 

-        mobj = re.match(self._VALID_URL, url)
-        video_id = mobj.group('id')
-
-        webpage = self._download_webpage(url, video_id)
+        webpage = self._download_webpage(url, playlist_id)

 
         NOT_AVAILABLE_STRING = 'This content is not available at your territory due to limited copyright.'
         if '%s</p>' % NOT_AVAILABLE_STRING in webpage:
             raise ExtractorError(NOT_AVAILABLE_STRING, expected=True)
 
 
         NOT_AVAILABLE_STRING = 'This content is not available at your territory due to limited copyright.'
         if '%s</p>' % NOT_AVAILABLE_STRING in webpage:
             raise ExtractorError(NOT_AVAILABLE_STRING, expected=True)
 

-        typ = self._html_search_regex(r'getPlaylistUrl\(\[\{"type":"(.+?)","id":".+?"\}\],', webpage, 'type')
-        episode_id = self._html_search_regex(r'getPlaylistUrl\(\[\{"type":".+?","id":"(.+?)"\}\],', webpage, 'episode_id')
+        type_ = None
+        episode_id = None
+
+        playlist = self._parse_json(
+            self._search_regex(
+                r'getPlaylistUrl\(\[({.+?})\]', webpage, 'playlist',
+                default='{}'), playlist_id)
+        if playlist:
+            type_ = playlist.get('type')
+            episode_id = playlist.get('id')
+
+        if not type_:
+            type_ = self._html_search_regex(
+                r'getPlaylistUrl\(\[\{"type":"(.+?)","id":".+?"\}\],',
+                webpage, 'type')
+        if not episode_id:
+            episode_id = self._html_search_regex(
+                r'getPlaylistUrl\(\[\{"type":".+?","id":"(.+?)"\}\],',
+                webpage, 'episode_id')

 
         data = {
 
         data = {

-            'playlist[0][type]': typ,
+            'playlist[0][type]': type_,

             'playlist[0][id]': episode_id,
             'requestUrl': compat_urllib_parse_urlparse(url).path,
             'requestSource': 'iVysilani',
         }
 
             'playlist[0][id]': episode_id,
             'requestUrl': compat_urllib_parse_urlparse(url).path,
             'requestSource': 'iVysilani',
         }
 

-        req = compat_urllib_request.Request('http://www.ceskatelevize.cz/ivysilani/ajax/get-playlist-url',
-                                            data=compat_urllib_parse.urlencode(data))
+        entries = []
+
+        for user_agent in (None, USER_AGENTS['Safari']):
+            req = sanitized_Request(
+                'http://www.ceskatelevize.cz/ivysilani/ajax/get-client-playlist',
+                data=urlencode_postdata(data))
+
+            req.add_header('Content-type', 'application/x-www-form-urlencoded')
+            req.add_header('x-addr', '127.0.0.1')
+            req.add_header('X-Requested-With', 'XMLHttpRequest')
+            if user_agent:
+                req.add_header('User-Agent', user_agent)
+            req.add_header('Referer', url)
+
+            playlistpage = self._download_json(req, playlist_id, fatal=False)
+
+            if not playlistpage:
+                continue
+
+            playlist_url = playlistpage['url']
+            if playlist_url == 'error_region':
+                raise ExtractorError(NOT_AVAILABLE_STRING, expected=True)

 
 

-        req.add_header('Content-type', 'application/x-www-form-urlencoded')
-        req.add_header('x-addr', '127.0.0.1')
-        req.add_header('X-Requested-With', 'XMLHttpRequest')
-        req.add_header('Referer', url)
+            req = sanitized_Request(compat_urllib_parse_unquote(playlist_url))
+            req.add_header('Referer', url)

 
 

-        playlistpage = self._download_json(req, video_id)
+            playlist_title = self._og_search_title(webpage, default=None)
+            playlist_description = self._og_search_description(webpage, default=None)

 
 

-        req = compat_urllib_request.Request(compat_urllib_parse.unquote(playlistpage['url']))
-        req.add_header('Referer', url)
+            playlist = self._download_json(req, playlist_id, fatal=False)
+            if not playlist:
+                continue

 
 

-        playlist = self._download_xml(req, video_id)
+            playlist = playlist.get('playlist')
+            if not isinstance(playlist, list):
+                continue

 
 

-        formats = []
-        for i in playlist.find('smilRoot/body'):
-            if 'AD' not in i.attrib['id']:
-                base_url = i.attrib['base']
-                parsedurl = compat_urllib_parse_urlparse(base_url)
-                duration = i.attrib['duration']
+            playlist_len = len(playlist)

 
 

-                for video in i.findall('video'):
-                    if video.attrib['label'] != 'AD':
-                        format_id = video.attrib['label']
-                        play_path = video.attrib['src']
-                        vbr = int(video.attrib['system-bitrate'])
+            for num, item in enumerate(playlist):
+                is_live = item.get('type') == 'LIVE'
+                formats = []
+                for format_id, stream_url in item.get('streamUrls', {}).items():
+                    if 'playerType=flash' in stream_url:
+                        stream_formats = self._extract_m3u8_formats(
+                            stream_url, playlist_id, 'mp4', 'm3u8_native',
+                            m3u8_id='hls-%s' % format_id, fatal=False)
+                    else:
+                        stream_formats = self._extract_mpd_formats(
+                            stream_url, playlist_id,
+                            mpd_id='dash-%s' % format_id, fatal=False)
+                    # See https://github.com/rg3/youtube-dl/issues/12119#issuecomment-280037031
+                    if format_id == 'audioDescription':
+                        for f in stream_formats:
+                            f['source_preference'] = -10
+                    formats.extend(stream_formats)

 
 

-                        formats.append({
-                            'format_id': format_id,
-                            'url': base_url,
-                            'vbr': vbr,
-                            'play_path': play_path,
-                            'app': parsedurl.path[1:] + '?' + parsedurl.query,
-                            'rtmp_live': True,
-                            'ext': 'flv',
-                        })
+                if user_agent and len(entries) == playlist_len:
+                    entries[num]['formats'].extend(formats)
+                    continue

 
 

-        self._sort_formats(formats)
+                item_id = item.get('id') or item['assetId']
+                title = item['title']

 
 

+                duration = float_or_none(item.get('duration'))
+                thumbnail = item.get('previewImageUrl')
+
+                subtitles = {}
+                if item.get('type') == 'VOD':
+                    subs = item.get('subtitles')
+                    if subs:
+                        subtitles = self.extract_subtitles(episode_id, subs)
+
+                if playlist_len == 1:
+                    final_title = playlist_title or title
+                    if is_live:
+                        final_title = self._live_title(final_title)
+                else:
+                    final_title = '%s (%s)' % (playlist_title, title)
+
+                entries.append({
+                    'id': item_id,
+                    'title': final_title,
+                    'description': playlist_description if playlist_len == 1 else None,
+                    'thumbnail': thumbnail,
+                    'duration': duration,
+                    'formats': formats,
+                    'subtitles': subtitles,
+                    'is_live': is_live,
+                })
+
+        for e in entries:
+            self._sort_formats(e['formats'])
+
+        return self.playlist_result(entries, playlist_id, playlist_title, playlist_description)
+
+    def _get_subtitles(self, episode_id, subs):
+        original_subtitles = self._download_webpage(
+            subs[0]['url'], episode_id, 'Downloading subtitles')
+        srt_subs = self._fix_subtitles(original_subtitles)

         return {
         return {

-            'id': episode_id,
-            'title': self._html_search_regex(r'<title>(.+?) — iVysílání — Česká televize</title>', webpage, 'title'),
-            'duration': float(duration),
-            'formats': formats,
+            'cs': [{
+                'ext': 'srt',
+                'data': srt_subs,
+            }]

         }
         }

+
+    @staticmethod
+    def _fix_subtitles(subtitles):
+        """ Convert millisecond-based subtitles to SRT """
+
+        def _msectotimecode(msec):
+            """ Helper utility to convert milliseconds to timecode """
+            components = []
+            for divider in [1000, 60, 60, 100]:
+                components.append(msec % divider)
+                msec //= divider
+            return '{3:02}:{2:02}:{1:02},{0:03}'.format(*components)
+
+        def _fix_subtitle(subtitle):
+            for line in subtitle.splitlines():
+                m = re.match(r'^\s*([0-9]+);\s*([0-9]+)\s+([0-9]+)\s*$', line)
+                if m:
+                    yield m.group(1)
+                    start, stop = (_msectotimecode(int(t)) for t in m.groups()[1:])
+                    yield '{0} --> {1}'.format(start, stop)
+                else:
+                    yield line
+
+        return '\r\n'.join(_fix_subtitle(subtitles))
+
+
+class CeskaTelevizePoradyIE(InfoExtractor):
+    _VALID_URL = r'https?://(?:www\.)?ceskatelevize\.cz/porady/(?:[^/?#&]+/)*(?P<id>[^/#?]+)'
+    _TESTS = [{
+        # video with 18+ caution trailer
+        'url': 'http://www.ceskatelevize.cz/porady/10520528904-queer/215562210900007-bogotart/',
+        'info_dict': {
+            'id': '215562210900007-bogotart',
+            'title': 'Queer: Bogotart',
+            'description': 'Alternativní průvodce současným queer světem',
+        },
+        'playlist': [{
+            'info_dict': {
+                'id': '61924494876844842',
+                'ext': 'mp4',
+                'title': 'Queer: Bogotart (Varování 18+)',
+                'duration': 10.2,
+            },
+        }, {
+            'info_dict': {
+                'id': '61924494877068022',
+                'ext': 'mp4',
+                'title': 'Queer: Bogotart (Queer)',
+                'thumbnail': r're:^https?://.*\.jpg',
+                'duration': 1558.3,
+            },
+        }],
+        'params': {
+            # m3u8 download
+            'skip_download': True,
+        },
+    }]
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+
+        webpage = self._download_webpage(url, video_id)
+
+        data_url = unescapeHTML(self._search_regex(
+            r'<span[^>]*\bdata-url=(["\'])(?P<url>(?:(?!\1).)+)\1',
+            webpage, 'iframe player url', group='url'))
+
+        return self.url_result(data_url, ie=CeskaTelevizeIE.ie_key())