]> Raphaël G. Git Repositories - youtubedl/blobdiff - youtube_dl/extractor/tvplay.py
Merge tag 'upstream/2016.12.01'
[youtubedl] / youtube_dl / extractor / tvplay.py
index df70a6b230a4217261f3f69a3e1213a88f07afbf..3eda0a399cf602d8e717c75e392a18e9dac82ec9 100644 (file)
@@ -4,47 +4,58 @@ from __future__ import unicode_literals
 import re
 
 from .common import InfoExtractor
-from ..compat import compat_str
+from ..compat import (
+    compat_HTTPError,
+    compat_str,
+    compat_urlparse,
+)
 from ..utils import (
+    determine_ext,
+    ExtractorError,
+    int_or_none,
     parse_iso8601,
     qualities,
+    try_get,
+    update_url_query,
 )
 
 
 class TVPlayIE(InfoExtractor):
-    IE_DESC = 'TV3Play and related services'
-    _VALID_URL = r'''(?x)https?://(?:www\.)?
-        (?:tvplay\.lv/parraides|
-           tv3play\.lt/programos|
-           play\.tv3\.lt/programos|
-           tv3play\.ee/sisu|
-           tv3play\.se/program|
-           tv6play\.se/program|
-           tv8play\.se/program|
-           tv10play\.se/program|
-           tv3play\.no/programmer|
-           viasat4play\.no/programmer|
-           tv6play\.no/programmer|
-           tv3play\.dk/programmer|
-           play\.novatv\.bg/programi
-        )/[^/]+/(?P<id>\d+)
-        '''
+    IE_NAME = 'mtg'
+    IE_DESC = 'MTG services'
+    _VALID_URL = r'''(?x)
+                    (?:
+                        mtg:|
+                        https?://
+                            (?:www\.)?
+                            (?:
+                                tvplay(?:\.skaties)?\.lv/parraides|
+                                (?:tv3play|play\.tv3)\.lt/programos|
+                                tv3play(?:\.tv3)?\.ee/sisu|
+                                (?:tv(?:3|6|8|10)play|viafree)\.se/program|
+                                (?:(?:tv3play|viasat4play|tv6play|viafree)\.no|(?:tv3play|viafree)\.dk)/programmer|
+                                play\.novatv\.bg/programi
+                            )
+                            /(?:[^/]+/)+
+                        )
+                        (?P<id>\d+)
+                    '''
     _TESTS = [
         {
             'url': 'http://www.tvplay.lv/parraides/vinas-melo-labak/418113?autostart=true',
+            'md5': 'a1612fe0849455423ad8718fe049be21',
             'info_dict': {
                 'id': '418113',
-                'ext': 'flv',
+                'ext': 'mp4',
                 'title': 'Kādi ir īri? - Viņas melo labāk',
                 'description': 'Baiba apsmej īrus, kādi tie ir un ko viņi dara.',
+                'series': 'Viņas melo labāk',
+                'season': '2.sezona',
+                'season_number': 2,
                 'duration': 25,
                 'timestamp': 1406097056,
                 'upload_date': '20140723',
             },
-            'params': {
-                # rtmp download
-                'skip_download': True,
-            },
         },
         {
             'url': 'http://play.tv3.lt/programos/moterys-meluoja-geriau/409229?autostart=true',
@@ -53,6 +64,10 @@ class TVPlayIE(InfoExtractor):
                 'ext': 'flv',
                 'title': 'Moterys meluoja geriau',
                 'description': 'md5:9aec0fc68e2cbc992d2a140bd41fa89e',
+                'series': 'Moterys meluoja geriau',
+                'episode_number': 47,
+                'season': '1 sezonas',
+                'season_number': 1,
                 'duration': 1330,
                 'timestamp': 1403769181,
                 'upload_date': '20140626',
@@ -82,7 +97,7 @@ class TVPlayIE(InfoExtractor):
             'url': 'http://www.tv3play.se/program/husraddarna/395385?autostart=true',
             'info_dict': {
                 'id': '395385',
-                'ext': 'flv',
+                'ext': 'mp4',
                 'title': 'Husräddarna S02E07',
                 'description': 'md5:f210c6c89f42d4fc39faa551be813777',
                 'duration': 2574,
@@ -90,7 +105,6 @@ class TVPlayIE(InfoExtractor):
                 'upload_date': '20140520',
             },
             'params': {
-                # rtmp download
                 'skip_download': True,
             },
         },
@@ -98,7 +112,7 @@ class TVPlayIE(InfoExtractor):
             'url': 'http://www.tv6play.se/program/den-sista-dokusapan/266636?autostart=true',
             'info_dict': {
                 'id': '266636',
-                'ext': 'flv',
+                'ext': 'mp4',
                 'title': 'Den sista dokusåpan S01E08',
                 'description': 'md5:295be39c872520221b933830f660b110',
                 'duration': 1492,
@@ -107,7 +121,6 @@ class TVPlayIE(InfoExtractor):
                 'age_limit': 18,
             },
             'params': {
-                # rtmp download
                 'skip_download': True,
             },
         },
@@ -115,7 +128,7 @@ class TVPlayIE(InfoExtractor):
             'url': 'http://www.tv8play.se/program/antikjakten/282756?autostart=true',
             'info_dict': {
                 'id': '282756',
-                'ext': 'flv',
+                'ext': 'mp4',
                 'title': 'Antikjakten S01E10',
                 'description': 'md5:1b201169beabd97e20c5ad0ad67b13b8',
                 'duration': 2646,
@@ -123,7 +136,6 @@ class TVPlayIE(InfoExtractor):
                 'upload_date': '20120925',
             },
             'params': {
-                # rtmp download
                 'skip_download': True,
             },
         },
@@ -131,7 +143,7 @@ class TVPlayIE(InfoExtractor):
             'url': 'http://www.tv3play.no/programmer/anna-anka-soker-assistent/230898?autostart=true',
             'info_dict': {
                 'id': '230898',
-                'ext': 'flv',
+                'ext': 'mp4',
                 'title': 'Anna Anka søker assistent - Ep. 8',
                 'description': 'md5:f80916bf5bbe1c5f760d127f8dd71474',
                 'duration': 2656,
@@ -139,7 +151,6 @@ class TVPlayIE(InfoExtractor):
                 'upload_date': '20100628',
             },
             'params': {
-                # rtmp download
                 'skip_download': True,
             },
         },
@@ -147,7 +158,7 @@ class TVPlayIE(InfoExtractor):
             'url': 'http://www.viasat4play.no/programmer/budbringerne/21873?autostart=true',
             'info_dict': {
                 'id': '21873',
-                'ext': 'flv',
+                'ext': 'mp4',
                 'title': 'Budbringerne program 10',
                 'description': 'md5:4db78dc4ec8a85bb04fd322a3ee5092d',
                 'duration': 1297,
@@ -155,7 +166,6 @@ class TVPlayIE(InfoExtractor):
                 'upload_date': '20090929',
             },
             'params': {
-                # rtmp download
                 'skip_download': True,
             },
         },
@@ -163,7 +173,7 @@ class TVPlayIE(InfoExtractor):
             'url': 'http://www.tv6play.no/programmer/hotelinspektor-alex-polizzi/361883?autostart=true',
             'info_dict': {
                 'id': '361883',
-                'ext': 'flv',
+                'ext': 'mp4',
                 'title': 'Hotelinspektør Alex Polizzi - Ep. 10',
                 'description': 'md5:3ecf808db9ec96c862c8ecb3a7fdaf81',
                 'duration': 2594,
@@ -171,7 +181,6 @@ class TVPlayIE(InfoExtractor):
                 'upload_date': '20140224',
             },
             'params': {
-                # rtmp download
                 'skip_download': True,
             },
         },
@@ -191,59 +200,226 @@ class TVPlayIE(InfoExtractor):
                 'skip_download': True,
             },
         },
+        {
+            'url': 'http://tvplay.skaties.lv/parraides/vinas-melo-labak/418113?autostart=true',
+            'only_matching': True,
+        },
+        {
+            # views is null
+            'url': 'http://tvplay.skaties.lv/parraides/tv3-zinas/760183',
+            'only_matching': True,
+        },
+        {
+            'url': 'http://tv3play.tv3.ee/sisu/kodu-keset-linna/238551?autostart=true',
+            'only_matching': True,
+        },
+        {
+            'url': 'http://www.viafree.se/program/underhallning/i-like-radio-live/sasong-1/676869',
+            'only_matching': True,
+        },
+        {
+            'url': 'mtg:418113',
+            'only_matching': True,
+        }
     ]
 
     def _real_extract(self, url):
         video_id = self._match_id(url)
 
         video = self._download_json(
-            'http://playapi.mtgx.tv/v1/videos/%s' % video_id, video_id, 'Downloading video JSON')
+            'http://playapi.mtgx.tv/v3/videos/%s' % video_id, video_id, 'Downloading video JSON')
 
-        if video['is_geo_blocked']:
-            self.report_warning(
-                'This content might not be available in your country due to copyright reasons')
+        title = video['title']
 
-        streams = self._download_json(
-            'http://playapi.mtgx.tv/v1/videos/stream/%s' % video_id, video_id, 'Downloading streams JSON')
+        try:
+            streams = self._download_json(
+                'http://playapi.mtgx.tv/v3/videos/stream/%s' % video_id,
+                video_id, 'Downloading streams JSON')
+        except ExtractorError as e:
+            if isinstance(e.cause, compat_HTTPError) and e.cause.code == 403:
+                msg = self._parse_json(e.cause.read().decode('utf-8'), video_id)
+                raise ExtractorError(msg['msg'], expected=True)
+            raise
 
         quality = qualities(['hls', 'medium', 'high'])
         formats = []
-        for format_id, video_url in streams['streams'].items():
+        for format_id, video_url in streams.get('streams', {}).items():
             if not video_url or not isinstance(video_url, compat_str):
                 continue
-            fmt = {
-                'format_id': format_id,
-                'preference': quality(format_id),
-            }
-            if video_url.startswith('rtmp'):
-                m = re.search(r'^(?P<url>rtmp://[^/]+/(?P<app>[^/]+))/(?P<playpath>.+)$', video_url)
-                if not m:
-                    continue
-                fmt.update({
-                    'ext': 'flv',
-                    'url': m.group('url'),
-                    'app': m.group('app'),
-                    'play_path': m.group('playpath'),
-                })
-            elif video_url.endswith('.f4m'):
+            ext = determine_ext(video_url)
+            if ext == 'f4m':
                 formats.extend(self._extract_f4m_formats(
-                    video_url + '?hdcore=3.5.0&plugin=aasp-3.5.0.151.81', video_id))
-                continue
+                    update_url_query(video_url, {
+                        'hdcore': '3.5.0',
+                        'plugin': 'aasp-3.5.0.151.81'
+                    }), video_id, f4m_id='hds', fatal=False))
+            elif ext == 'm3u8':
+                formats.extend(self._extract_m3u8_formats(
+                    video_url, video_id, 'mp4', 'm3u8_native',
+                    m3u8_id='hls', fatal=False))
             else:
-                fmt.update({
-                    'url': video_url,
-                })
-            formats.append(fmt)
+                fmt = {
+                    'format_id': format_id,
+                    'quality': quality(format_id),
+                    'ext': ext,
+                }
+                if video_url.startswith('rtmp'):
+                    m = re.search(
+                        r'^(?P<url>rtmp://[^/]+/(?P<app>[^/]+))/(?P<playpath>.+)$', video_url)
+                    if not m:
+                        continue
+                    fmt.update({
+                        'ext': 'flv',
+                        'url': m.group('url'),
+                        'app': m.group('app'),
+                        'play_path': m.group('playpath'),
+                    })
+                else:
+                    fmt.update({
+                        'url': video_url,
+                    })
+                formats.append(fmt)
+
+        if not formats and video.get('is_geo_blocked'):
+            self.raise_geo_restricted(
+                'This content might not be available in your country due to copyright reasons')
 
         self._sort_formats(formats)
 
+        # TODO: webvtt in m3u8
+        subtitles = {}
+        sami_path = video.get('sami_path')
+        if sami_path:
+            lang = self._search_regex(
+                r'_([a-z]{2})\.xml', sami_path, 'lang',
+                default=compat_urlparse.urlparse(url).netloc.rsplit('.', 1)[-1])
+            subtitles[lang] = [{
+                'url': sami_path,
+            }]
+
+        series = video.get('format_title')
+        episode_number = int_or_none(video.get('format_position', {}).get('episode'))
+        season = video.get('_embedded', {}).get('season', {}).get('title')
+        season_number = int_or_none(video.get('format_position', {}).get('season'))
+
         return {
             'id': video_id,
-            'title': video['title'],
-            'description': video['description'],
-            'duration': video['duration'],
-            'timestamp': parse_iso8601(video['created_at']),
-            'view_count': video['views']['total'],
-            'age_limit': video.get('age_limit', 0),
+            'title': title,
+            'description': video.get('description'),
+            'series': series,
+            'episode_number': episode_number,
+            'season': season,
+            'season_number': season_number,
+            'duration': int_or_none(video.get('duration')),
+            'timestamp': parse_iso8601(video.get('created_at')),
+            'view_count': try_get(video, lambda x: x['views']['total'], int),
+            'age_limit': int_or_none(video.get('age_limit', 0)),
             'formats': formats,
+            'subtitles': subtitles,
         }
+
+
+class ViafreeIE(InfoExtractor):
+    _VALID_URL = r'''(?x)
+                    https?://
+                        (?:www\.)?
+                        viafree\.
+                        (?:
+                            (?:dk|no)/programmer|
+                            se/program
+                        )
+                        /(?:[^/]+/)+(?P<id>[^/?#&]+)
+                    '''
+    _TESTS = [{
+        'url': 'http://www.viafree.se/program/livsstil/husraddarna/sasong-2/avsnitt-2',
+        'info_dict': {
+            'id': '395375',
+            'ext': 'mp4',
+            'title': 'Husräddarna S02E02',
+            'description': 'md5:4db5c933e37db629b5a2f75dfb34829e',
+            'series': 'Husräddarna',
+            'season': 'Säsong 2',
+            'season_number': 2,
+            'duration': 2576,
+            'timestamp': 1400596321,
+            'upload_date': '20140520',
+        },
+        'params': {
+            'skip_download': True,
+        },
+        'add_ie': [TVPlayIE.ie_key()],
+    }, {
+        # with relatedClips
+        'url': 'http://www.viafree.se/program/reality/sommaren-med-youtube-stjarnorna/sasong-1/avsnitt-1',
+        'info_dict': {
+            'id': '758770',
+            'ext': 'mp4',
+            'title': 'Sommaren med YouTube-stjärnorna S01E01',
+            'description': 'md5:2bc69dce2c4bb48391e858539bbb0e3f',
+            'series': 'Sommaren med YouTube-stjärnorna',
+            'season': 'Säsong 1',
+            'season_number': 1,
+            'duration': 1326,
+            'timestamp': 1470905572,
+            'upload_date': '20160811',
+        },
+        'params': {
+            'skip_download': True,
+        },
+        'add_ie': [TVPlayIE.ie_key()],
+    }, {
+        # Different og:image URL schema
+        'url': 'http://www.viafree.se/program/reality/sommaren-med-youtube-stjarnorna/sasong-1/avsnitt-2',
+        'only_matching': True,
+    }, {
+        'url': 'http://www.viafree.no/programmer/underholdning/det-beste-vorspielet/sesong-2/episode-1',
+        'only_matching': True,
+    }, {
+        'url': 'http://www.viafree.dk/programmer/reality/paradise-hotel/saeson-7/episode-5',
+        'only_matching': True,
+    }]
+
+    @classmethod
+    def suitable(cls, url):
+        return False if TVPlayIE.suitable(url) else super(ViafreeIE, cls).suitable(url)
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+
+        webpage = self._download_webpage(url, video_id)
+
+        data = self._parse_json(
+            self._search_regex(
+                r'(?s)window\.App\s*=\s*({.+?})\s*;\s*</script',
+                webpage, 'data', default='{}'),
+            video_id, transform_source=lambda x: re.sub(
+                r'(?s)function\s+[a-zA-Z_][\da-zA-Z_]*\s*\([^)]*\)\s*{[^}]*}\s*',
+                'null', x), fatal=False)
+
+        video_id = None
+
+        if data:
+            video_id = try_get(
+                data, lambda x: x['context']['dispatcher']['stores'][
+                    'ContentPageProgramStore']['currentVideo']['id'],
+                compat_str)
+
+        # Fallback #1 (extract from og:image URL schema)
+        if not video_id:
+            thumbnail = self._og_search_thumbnail(webpage, default=None)
+            if thumbnail:
+                video_id = self._search_regex(
+                    # Patterns seen:
+                    #  http://cdn.playapi.mtgx.tv/imagecache/600x315/cloud/content-images/inbox/765166/a2e95e5f1d735bab9f309fa345cc3f25.jpg
+                    #  http://cdn.playapi.mtgx.tv/imagecache/600x315/cloud/content-images/seasons/15204/758770/4a5ba509ca8bc043e1ebd1a76131cdf2.jpg
+                    r'https?://[^/]+/imagecache/(?:[^/]+/)+(\d{6,})/',
+                    thumbnail, 'video id', default=None)
+
+        # Fallback #2. Extract from raw JSON string.
+        # May extract wrong video id if relatedClips is present.
+        if not video_id:
+            video_id = self._search_regex(
+                r'currentVideo["\']\s*:\s*.+?["\']id["\']\s*:\s*["\'](\d{6,})',
+                webpage, 'video id')
+
+        return self.url_result('mtg:%s' % video_id, TVPlayIE.ie_key())