debian/control: Upgrade debhelper-compat to 13.

[youtubedl] / youtube_dl / extractor / nova.py

diff --git a/youtube_dl/extractor/nova.py b/youtube_dl/extractor/nova.py
index 06cb8cb3f5a867104a8bd67abd9ce60bbc1df25f..47b9748f0202d76ddef8ee5f96257bebe88e4169 100644 (file)
--- a/youtube_dl/extractor/nova.py
+++ b/youtube_dl/extractor/nova.py
@@ -6,30 +6,134 @@ import re
 from .common import InfoExtractor
 from ..utils import (
     clean_html,
 from .common import InfoExtractor
 from ..utils import (
     clean_html,

+    determine_ext,
+    int_or_none,
+    js_to_json,
+    qualities,

     unified_strdate,
     unified_strdate,

+    url_or_none,

 )
 
 
 )
 
 

+class NovaEmbedIE(InfoExtractor):
+    _VALID_URL = r'https?://media\.cms\.nova\.cz/embed/(?P<id>[^/?#&]+)'
+    _TEST = {
+        'url': 'https://media.cms.nova.cz/embed/8o0n0r?autoplay=1',
+        'md5': 'ee009bafcc794541570edd44b71cbea3',
+        'info_dict': {
+            'id': '8o0n0r',
+            'ext': 'mp4',
+            'title': '2180. díl',
+            'thumbnail': r're:^https?://.*\.jpg',
+            'duration': 2578,
+        },
+    }
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+
+        webpage = self._download_webpage(url, video_id)
+
+        duration = None
+        formats = []
+
+        player = self._parse_json(
+            self._search_regex(
+                r'Player\.init\s*\([^,]+,\s*({.+?})\s*,\s*{.+?}\s*\)\s*;',
+                webpage, 'player', default='{}'), video_id, fatal=False)
+        if player:
+            for format_id, format_list in player['tracks'].items():
+                if not isinstance(format_list, list):
+                    format_list = [format_list]
+                for format_dict in format_list:
+                    if not isinstance(format_dict, dict):
+                        continue
+                    format_url = url_or_none(format_dict.get('src'))
+                    format_type = format_dict.get('type')
+                    ext = determine_ext(format_url)
+                    if (format_type == 'application/x-mpegURL'
+                            or format_id == 'HLS' or ext == 'm3u8'):
+                        formats.extend(self._extract_m3u8_formats(
+                            format_url, video_id, 'mp4',
+                            entry_protocol='m3u8_native', m3u8_id='hls',
+                            fatal=False))
+                    elif (format_type == 'application/dash+xml'
+                          or format_id == 'DASH' or ext == 'mpd'):
+                        formats.extend(self._extract_mpd_formats(
+                            format_url, video_id, mpd_id='dash', fatal=False))
+                    else:
+                        formats.append({
+                            'url': format_url,
+                        })
+            duration = int_or_none(player.get('duration'))
+        else:
+            # Old path, not actual as of 08.04.2020
+            bitrates = self._parse_json(
+                self._search_regex(
+                    r'(?s)(?:src|bitrates)\s*=\s*({.+?})\s*;', webpage, 'formats'),
+                video_id, transform_source=js_to_json)
+
+            QUALITIES = ('lq', 'mq', 'hq', 'hd')
+            quality_key = qualities(QUALITIES)
+
+            for format_id, format_list in bitrates.items():
+                if not isinstance(format_list, list):
+                    format_list = [format_list]
+                for format_url in format_list:
+                    format_url = url_or_none(format_url)
+                    if not format_url:
+                        continue
+                    if format_id == 'hls':
+                        formats.extend(self._extract_m3u8_formats(
+                            format_url, video_id, ext='mp4',
+                            entry_protocol='m3u8_native', m3u8_id='hls',
+                            fatal=False))
+                        continue
+                    f = {
+                        'url': format_url,
+                    }
+                    f_id = format_id
+                    for quality in QUALITIES:
+                        if '%s.mp4' % quality in format_url:
+                            f_id += '-%s' % quality
+                            f.update({
+                                'quality': quality_key(quality),
+                                'format_note': quality.upper(),
+                            })
+                            break
+                    f['format_id'] = f_id
+                    formats.append(f)
+
+        self._sort_formats(formats)
+
+        title = self._og_search_title(
+            webpage, default=None) or self._search_regex(
+            (r'<value>(?P<title>[^<]+)',
+             r'videoTitle\s*:\s*(["\'])(?P<value>(?:(?!\1).)+)\1'), webpage,
+            'title', group='value')
+        thumbnail = self._og_search_thumbnail(
+            webpage, default=None) or self._search_regex(
+            r'poster\s*:\s*(["\'])(?P<value>(?:(?!\1).)+)\1', webpage,
+            'thumbnail', fatal=False, group='value')
+        duration = int_or_none(self._search_regex(
+            r'videoDuration\s*:\s*(\d+)', webpage, 'duration',
+            default=duration))
+
+        return {
+            'id': video_id,
+            'title': title,
+            'thumbnail': thumbnail,
+            'duration': duration,
+            'formats': formats,
+        }
+
+

 class NovaIE(InfoExtractor):
     IE_DESC = 'TN.cz, Prásk.tv, Nova.cz, Novaplus.cz, FANDA.tv, Krásná.cz and Doma.cz'
     _VALID_URL = r'https?://(?:[^.]+\.)?(?P<site>tv(?:noviny)?|tn|novaplus|vymena|fanda|krasna|doma|prask)\.nova\.cz/(?:[^/]+/)+(?P<id>[^/]+?)(?:\.html|/|$)'
     _TESTS = [{
 class NovaIE(InfoExtractor):
     IE_DESC = 'TN.cz, Prásk.tv, Nova.cz, Novaplus.cz, FANDA.tv, Krásná.cz and Doma.cz'
     _VALID_URL = r'https?://(?:[^.]+\.)?(?P<site>tv(?:noviny)?|tn|novaplus|vymena|fanda|krasna|doma|prask)\.nova\.cz/(?:[^/]+/)+(?P<id>[^/]+?)(?:\.html|/|$)'
     _TESTS = [{

-        'url': 'http://tvnoviny.nova.cz/clanek/novinky/co-na-sebe-sportaci-praskli-vime-jestli-pujde-hrdlicka-na-materskou.html?utm_source=tvnoviny&utm_medium=cpfooter&utm_campaign=novaplus',
-        'info_dict': {
-            'id': '1608920',
-            'display_id': 'co-na-sebe-sportaci-praskli-vime-jestli-pujde-hrdlicka-na-materskou',
-            'ext': 'flv',
-            'title': 'Duel: Michal Hrdlička a Petr Suchoň',
-            'description': 'md5:d0cc509858eee1b1374111c588c6f5d5',
-            'thumbnail': r're:^https?://.*\.(?:jpg)',
-        },
-        'params': {
-            # rtmp download
-            'skip_download': True,
-        }
-    }, {

         'url': 'http://tn.nova.cz/clanek/tajemstvi-ukryte-v-podzemi-specialni-nemocnice-v-prazske-krci.html#player_13260',
         'url': 'http://tn.nova.cz/clanek/tajemstvi-ukryte-v-podzemi-specialni-nemocnice-v-prazske-krci.html#player_13260',

-        'md5': '1dd7b9d5ea27bc361f110cd855a19bd3',
+        'md5': '249baab7d0104e186e78b0899c7d5f28',

         'info_dict': {
             'id': '1757139',
             'display_id': 'tajemstvi-ukryte-v-podzemi-specialni-nemocnice-v-prazske-krci',
         'info_dict': {
             'id': '1757139',
             'display_id': 'tajemstvi-ukryte-v-podzemi-specialni-nemocnice-v-prazske-krci',


@@ -38,33 +142,6 @@ class NovaIE(InfoExtractor):
             'description': 'md5:f0a42dd239c26f61c28f19e62d20ef53',
             'thumbnail': r're:^https?://.*\.(?:jpg)',
         }
             'description': 'md5:f0a42dd239c26f61c28f19e62d20ef53',
             'thumbnail': r're:^https?://.*\.(?:jpg)',
         }

-    }, {
-        'url': 'http://novaplus.nova.cz/porad/policie-modrava/video/5591-policie-modrava-15-dil-blondynka-na-hrbitove',
-        'info_dict': {
-            'id': '1756825',
-            'display_id': '5591-policie-modrava-15-dil-blondynka-na-hrbitove',
-            'ext': 'flv',
-            'title': 'Policie Modrava - 15. díl - Blondýnka na hřbitově',
-            'description': 'md5:dc24e50be5908df83348e50d1431295e',  # Make sure this description is clean of html tags
-            'thumbnail': r're:^https?://.*\.(?:jpg)',
-        },
-        'params': {
-            # rtmp download
-            'skip_download': True,
-        }
-    }, {
-        'url': 'http://novaplus.nova.cz/porad/televizni-noviny/video/5585-televizni-noviny-30-5-2015/',
-        'info_dict': {
-            'id': '1756858',
-            'ext': 'flv',
-            'title': 'Televizní noviny - 30. 5. 2015',
-            'thumbnail': r're:^https?://.*\.(?:jpg)',
-            'upload_date': '20150530',
-        },
-        'params': {
-            # rtmp download
-            'skip_download': True,
-        }

     }, {
         'url': 'http://fanda.nova.cz/clanek/fun-and-games/krvavy-epos-zaklinac-3-divoky-hon-vychazi-vyhrajte-ho-pro-sebe.html',
         'info_dict': {
     }, {
         'url': 'http://fanda.nova.cz/clanek/fun-and-games/krvavy-epos-zaklinac-3-divoky-hon-vychazi-vyhrajte-ho-pro-sebe.html',
         'info_dict': {


@@ -78,7 +155,23 @@ class NovaIE(InfoExtractor):
         'params': {
             # rtmp download
             'skip_download': True,
         'params': {
             # rtmp download
             'skip_download': True,

-        }
+        },
+        'skip': 'gone',
+    }, {
+        # media.cms.nova.cz embed
+        'url': 'https://novaplus.nova.cz/porad/ulice/epizoda/18760-2180-dil',
+        'info_dict': {
+            'id': '8o0n0r',
+            'ext': 'mp4',
+            'title': '2180. díl',
+            'thumbnail': r're:^https?://.*\.jpg',
+            'duration': 2578,
+        },
+        'params': {
+            'skip_download': True,
+        },
+        'add_ie': [NovaEmbedIE.ie_key()],
+        'skip': 'CHYBA 404: STRÁNKA NENALEZENA',

     }, {
         'url': 'http://sport.tn.nova.cz/clanek/sport/hokej/nhl/zivot-jde-dal-hodnotil-po-vyrazeni-z-playoff-jiri-sekac.html',
         'only_matching': True,
     }, {
         'url': 'http://sport.tn.nova.cz/clanek/sport/hokej/nhl/zivot-jde-dal-hodnotil-po-vyrazeni-z-playoff-jiri-sekac.html',
         'only_matching': True,


@@ -103,6 +196,30 @@ class NovaIE(InfoExtractor):
 
         webpage = self._download_webpage(url, display_id)
 
 
         webpage = self._download_webpage(url, display_id)
 

+        description = clean_html(self._og_search_description(webpage, default=None))
+        if site == 'novaplus':
+            upload_date = unified_strdate(self._search_regex(
+                r'(\d{1,2}-\d{1,2}-\d{4})$', display_id, 'upload date', default=None))
+        elif site == 'fanda':
+            upload_date = unified_strdate(self._search_regex(
+                r'<span class="date_time">(\d{1,2}\.\d{1,2}\.\d{4})', webpage, 'upload date', default=None))
+        else:
+            upload_date = None
+
+        # novaplus
+        embed_id = self._search_regex(
+            r'<iframe[^>]+\bsrc=["\'](?:https?:)?//media\.cms\.nova\.cz/embed/([^/?#&]+)',
+            webpage, 'embed url', default=None)
+        if embed_id:
+            return {
+                '_type': 'url_transparent',
+                'url': 'https://media.cms.nova.cz/embed/%s' % embed_id,
+                'ie_key': NovaEmbedIE.ie_key(),
+                'id': embed_id,
+                'description': description,
+                'upload_date': upload_date
+            }
+

         video_id = self._search_regex(
             [r"(?:media|video_id)\s*:\s*'(\d+)'",
              r'media=(\d+)',
         video_id = self._search_regex(
             [r"(?:media|video_id)\s*:\s*'(\d+)'",
              r'media=(\d+)',


@@ -111,8 +228,21 @@ class NovaIE(InfoExtractor):
             webpage, 'video id')
 
         config_url = self._search_regex(
             webpage, 'video id')
 
         config_url = self._search_regex(

-            r'src="(http://tn\.nova\.cz/bin/player/videojs/config\.php\?[^"]+)"',
+            r'src="(https?://(?:tn|api)\.nova\.cz/bin/player/videojs/config\.php\?[^"]+)"',

             webpage, 'config url', default=None)
             webpage, 'config url', default=None)

+        config_params = {}
+
+        if not config_url:
+            player = self._parse_json(
+                self._search_regex(
+                    r'(?s)Player\s*\(.+?\s*,\s*({.+?\bmedia\b["\']?\s*:\s*["\']?\d+.+?})\s*\)', webpage,
+                    'player', default='{}'),
+                video_id, transform_source=js_to_json, fatal=False)
+            if player:
+                config_url = url_or_none(player.get('configUrl'))
+                params = player.get('configParams')
+                if isinstance(params, dict):
+                    config_params = params

 
         if not config_url:
             DEFAULT_SITE_ID = '23000'
 
         if not config_url:
             DEFAULT_SITE_ID = '23000'


@@ -127,14 +257,20 @@ class NovaIE(InfoExtractor):
             }
 
             site_id = self._search_regex(
             }
 
             site_id = self._search_regex(

-                r'site=(\d+)', webpage, 'site id', default=None) or SITES.get(site, DEFAULT_SITE_ID)
+                r'site=(\d+)', webpage, 'site id', default=None) or SITES.get(
+                site, DEFAULT_SITE_ID)

 
 

-            config_url = ('http://tn.nova.cz/bin/player/videojs/config.php?site=%s&media=%s&jsVar=vjsconfig'
-                          % (site_id, video_id))
+            config_url = 'https://api.nova.cz/bin/player/videojs/config.php'
+            config_params = {
+                'site': site_id,
+                'media': video_id,
+                'quality': 3,
+                'version': 1,
+            }

 
         config = self._download_json(
             config_url, display_id,
 
         config = self._download_json(
             config_url, display_id,

-            'Downloading config JSON',
+            'Downloading config JSON', query=config_params,

             transform_source=lambda s: s[s.index('{'):s.rindex('}') + 1])
 
         mediafile = config['mediafile']
             transform_source=lambda s: s[s.index('{'):s.rindex('}') + 1])
 
         mediafile = config['mediafile']


@@ -156,18 +292,8 @@ class NovaIE(InfoExtractor):
         self._sort_formats(formats)
 
         title = mediafile.get('meta', {}).get('title') or self._og_search_title(webpage)
         self._sort_formats(formats)
 
         title = mediafile.get('meta', {}).get('title') or self._og_search_title(webpage)

-        description = clean_html(self._og_search_description(webpage, default=None))

         thumbnail = config.get('poster')
 
         thumbnail = config.get('poster')
 

-        if site == 'novaplus':
-            upload_date = unified_strdate(self._search_regex(
-                r'(\d{1,2}-\d{1,2}-\d{4})$', display_id, 'upload date', default=None))
-        elif site == 'fanda':
-            upload_date = unified_strdate(self._search_regex(
-                r'<span class="date_time">(\d{1,2}\.\d{1,2}\.\d{4})', webpage, 'upload date', default=None))
-        else:
-            upload_date = None
-

         return {
             'id': video_id,
             'display_id': display_id,
         return {
             'id': video_id,
             'display_id': display_id,