]> Raphaël G. Git Repositories - youtubedl/blobdiff - youtube_dl/extractor/yahoo.py
debian/copyright: Update my copyright years.
[youtubedl] / youtube_dl / extractor / yahoo.py
index 4951414e91ffbc34dc83629403c8b64ffb5e5682..e5ebdd1806ec30944d04ddb56707fea1228e2fae 100644 (file)
@@ -1,22 +1,28 @@
 # coding: utf-8
 from __future__ import unicode_literals
 
 # coding: utf-8
 from __future__ import unicode_literals
 
+import hashlib
 import itertools
 import json
 import re
 
 from .common import InfoExtractor, SearchInfoExtractor
 from ..compat import (
 import itertools
 import json
 import re
 
 from .common import InfoExtractor, SearchInfoExtractor
 from ..compat import (
+    compat_str,
     compat_urllib_parse,
     compat_urlparse,
 )
 from ..utils import (
     clean_html,
     compat_urllib_parse,
     compat_urlparse,
 )
 from ..utils import (
     clean_html,
-    unescapeHTML,
+    determine_ext,
     ExtractorError,
     ExtractorError,
+    extract_attributes,
     int_or_none,
     mimetype2ext,
     int_or_none,
     mimetype2ext,
-    determine_ext,
+    smuggle_url,
+    try_get,
+    unescapeHTML,
+    url_or_none,
 )
 
 from .brightcove import (
 )
 
 from .brightcove import (
@@ -28,7 +34,7 @@ from .nbc import NBCSportsVPlayerIE
 
 class YahooIE(InfoExtractor):
     IE_DESC = 'Yahoo screen and movies'
 
 class YahooIE(InfoExtractor):
     IE_DESC = 'Yahoo screen and movies'
-    _VALID_URL = r'(?P<url>(?P<host>https?://(?:[a-zA-Z]{2}\.)?[\da-zA-Z_-]+\.yahoo\.com)/(?:[^/]+/)*(?P<display_id>.+)?-(?P<id>[0-9]+)(?:-[a-z]+)?(?:\.html)?)'
+    _VALID_URL = r'(?P<host>https?://(?:(?P<country>[a-zA-Z]{2})\.)?[\da-zA-Z_-]+\.yahoo\.com)/(?:[^/]+/)*(?:(?P<display_id>.+)?-)?(?P<id>[0-9]+)(?:-[a-z]+)?(?:\.html)?'
     _TESTS = [
         {
             'url': 'http://screen.yahoo.com/julian-smith-travis-legg-watch-214727115.html',
     _TESTS = [
         {
             'url': 'http://screen.yahoo.com/julian-smith-travis-legg-watch-214727115.html',
@@ -50,6 +56,7 @@ class YahooIE(InfoExtractor):
                 'description': 'md5:66b627ab0a282b26352136ca96ce73c1',
                 'duration': 151,
             },
                 'description': 'md5:66b627ab0a282b26352136ca96ce73c1',
                 'duration': 151,
             },
+            'skip': 'HTTP Error 404',
         },
         {
             'url': 'https://screen.yahoo.com/community/community-sizzle-reel-203225340.html?format=embed',
         },
         {
             'url': 'https://screen.yahoo.com/community/community-sizzle-reel-203225340.html?format=embed',
@@ -142,7 +149,7 @@ class YahooIE(InfoExtractor):
             'skip': 'Domain name in.lifestyle.yahoo.com gone',
         }, {
             'url': 'https://www.yahoo.com/movies/v/true-story-trailer-173000497.html',
             'skip': 'Domain name in.lifestyle.yahoo.com gone',
         }, {
             'url': 'https://www.yahoo.com/movies/v/true-story-trailer-173000497.html',
-            'md5': '2a9752f74cb898af5d1083ea9f661b58',
+            'md5': '989396ae73d20c6f057746fb226aa215',
             'info_dict': {
                 'id': '071c4013-ce30-3a93-a5b2-e0413cd4a9d1',
                 'ext': 'mp4',
             'info_dict': {
                 'id': '071c4013-ce30-3a93-a5b2-e0413cd4a9d1',
                 'ext': 'mp4',
@@ -227,13 +234,33 @@ class YahooIE(InfoExtractor):
                 'skip_download': True,
             },
         },
                 'skip_download': True,
             },
         },
+        {
+            # custom brightcove
+            'url': 'https://au.tv.yahoo.com/plus7/sunrise/-/watch/37083565/clown-entertainers-say-it-is-hurting-their-business/',
+            'info_dict': {
+                'id': '5575377707001',
+                'ext': 'mp4',
+                'title': "Clown entertainers say 'It' is hurting their business",
+                'description': 'Stephen King s horror film has much to answer for. Jelby and Mr Loopy the Clowns join us.',
+                'timestamp': 1505341164,
+                'upload_date': '20170913',
+                'uploader_id': '2376984109001',
+            },
+            'params': {
+                'skip_download': True,
+            },
+        },
+        {
+            # custom brightcove, geo-restricted to Australia, bypassable
+            'url': 'https://au.tv.yahoo.com/plus7/sunrise/-/watch/37263964/sunrise-episode-wed-27-sep/',
+            'only_matching': True,
+        }
     ]
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
     ]
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
-        display_id = mobj.group('display_id') or self._match_id(url)
         page_id = mobj.group('id')
         page_id = mobj.group('id')
-        url = mobj.group('url')
+        display_id = mobj.group('display_id') or page_id
         host = mobj.group('host')
         webpage, urlh = self._download_webpage_handle(url, display_id)
         if 'err=404' in urlh.geturl():
         host = mobj.group('host')
         webpage, urlh = self._download_webpage_handle(url, display_id)
         if 'err=404' in urlh.geturl():
@@ -257,10 +284,31 @@ class YahooIE(InfoExtractor):
         if bc_url:
             return self.url_result(bc_url, BrightcoveLegacyIE.ie_key())
 
         if bc_url:
             return self.url_result(bc_url, BrightcoveLegacyIE.ie_key())
 
+        def brightcove_url_result(bc_url):
+            return self.url_result(
+                smuggle_url(bc_url, {'geo_countries': [mobj.group('country')]}),
+                BrightcoveNewIE.ie_key())
+
         # Look for Brightcove New Studio embeds
         # Look for Brightcove New Studio embeds
-        bc_url = BrightcoveNewIE._extract_url(webpage)
+        bc_url = BrightcoveNewIE._extract_url(self, webpage)
         if bc_url:
         if bc_url:
-            return self.url_result(bc_url, BrightcoveNewIE.ie_key())
+            return brightcove_url_result(bc_url)
+
+        brightcove_iframe = self._search_regex(
+            r'(<iframe[^>]+data-video-id=["\']\d+[^>]+>)', webpage,
+            'brightcove iframe', default=None)
+        if brightcove_iframe:
+            attr = extract_attributes(brightcove_iframe)
+            src = attr.get('src')
+            if src:
+                parsed_src = compat_urlparse.urlparse(src)
+                qs = compat_urlparse.parse_qs(parsed_src.query)
+                account_id = qs.get('accountId', ['2376984109001'])[0]
+                brightcove_id = attr.get('data-video-id') or qs.get('videoId', [None])[0]
+                if account_id and brightcove_id:
+                    return brightcove_url_result(
+                        'http://players.brightcove.net/%s/default_default/index.html?videoId=%s'
+                        % (account_id, brightcove_id))
 
         # Query result is often embedded in webpage as JSON. Sometimes explicit requests
         # to video API results in a failure with geo restriction reason therefore using
 
         # Query result is often embedded in webpage as JSON. Sometimes explicit requests
         # to video API results in a failure with geo restriction reason therefore using
@@ -433,3 +481,209 @@ class YahooSearchIE(SearchInfoExtractor):
             'id': query,
             'entries': entries,
         }
             'id': query,
             'entries': entries,
         }
+
+
+class YahooGyaOPlayerIE(InfoExtractor):
+    IE_NAME = 'yahoo:gyao:player'
+    _VALID_URL = r'https?://(?:gyao\.yahoo\.co\.jp/(?:player|episode/[^/]+)|streaming\.yahoo\.co\.jp/c/y)/(?P<id>\d+/v\d+/v\d+|[\da-f]{8}-[\da-f]{4}-[\da-f]{4}-[\da-f]{4}-[\da-f]{12})'
+    _TESTS = [{
+        'url': 'https://gyao.yahoo.co.jp/player/00998/v00818/v0000000000000008564/',
+        'info_dict': {
+            'id': '5993125228001',
+            'ext': 'mp4',
+            'title': 'フューリー 【字幕版】',
+            'description': 'md5:21e691c798a15330eda4db17a8fe45a5',
+            'uploader_id': '4235717419001',
+            'upload_date': '20190124',
+            'timestamp': 1548294365,
+        },
+        'params': {
+            # m3u8 download
+            'skip_download': True,
+        },
+    }, {
+        'url': 'https://streaming.yahoo.co.jp/c/y/01034/v00133/v0000000000000000706/',
+        'only_matching': True,
+    }, {
+        'url': 'https://gyao.yahoo.co.jp/episode/%E3%81%8D%E3%81%AE%E3%81%86%E4%BD%95%E9%A3%9F%E3%81%B9%E3%81%9F%EF%BC%9F%20%E7%AC%AC2%E8%A9%B1%202019%2F4%2F12%E6%94%BE%E9%80%81%E5%88%86/5cb02352-b725-409e-9f8d-88f947a9f682',
+        'only_matching': True,
+    }]
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url).replace('/', ':')
+        video = self._download_json(
+            'https://gyao.yahoo.co.jp/dam/v1/videos/' + video_id,
+            video_id, query={
+                'fields': 'longDescription,title,videoId',
+            }, headers={
+                'X-User-Agent': 'Unknown Pc GYAO!/2.0.0 Web',
+            })
+        return {
+            '_type': 'url_transparent',
+            'id': video_id,
+            'title': video['title'],
+            'url': smuggle_url(
+                'http://players.brightcove.net/4235717419001/default_default/index.html?videoId=' + video['videoId'],
+                {'geo_countries': ['JP']}),
+            'description': video.get('longDescription'),
+            'ie_key': BrightcoveNewIE.ie_key(),
+        }
+
+
+class YahooGyaOIE(InfoExtractor):
+    IE_NAME = 'yahoo:gyao'
+    _VALID_URL = r'https?://(?:gyao\.yahoo\.co\.jp/(?:p|title/[^/]+)|streaming\.yahoo\.co\.jp/p/y)/(?P<id>\d+/v\d+|[\da-f]{8}-[\da-f]{4}-[\da-f]{4}-[\da-f]{4}-[\da-f]{12})'
+    _TESTS = [{
+        'url': 'https://gyao.yahoo.co.jp/p/00449/v03102/',
+        'info_dict': {
+            'id': '00449:v03102',
+        },
+        'playlist_count': 2,
+    }, {
+        'url': 'https://streaming.yahoo.co.jp/p/y/01034/v00133/',
+        'only_matching': True,
+    }, {
+        'url': 'https://gyao.yahoo.co.jp/title/%E3%81%97%E3%82%83%E3%81%B9%E3%81%8F%E3%82%8A007/5b025a49-b2e5-4dc7-945c-09c6634afacf',
+        'only_matching': True,
+    }]
+
+    def _real_extract(self, url):
+        program_id = self._match_id(url).replace('/', ':')
+        videos = self._download_json(
+            'https://gyao.yahoo.co.jp/api/programs/%s/videos' % program_id, program_id)['videos']
+        entries = []
+        for video in videos:
+            video_id = video.get('id')
+            if not video_id:
+                continue
+            entries.append(self.url_result(
+                'https://gyao.yahoo.co.jp/player/%s/' % video_id.replace(':', '/'),
+                YahooGyaOPlayerIE.ie_key(), video_id))
+        return self.playlist_result(entries, program_id)
+
+
+class YahooJapanNewsIE(InfoExtractor):
+    IE_NAME = 'yahoo:japannews'
+    IE_DESC = 'Yahoo! Japan News'
+    _VALID_URL = r'https?://(?P<host>(?:news|headlines)\.yahoo\.co\.jp)[^\d]*(?P<id>\d[\d-]*\d)?'
+    _GEO_COUNTRIES = ['JP']
+    _TESTS = [{
+        'url': 'https://headlines.yahoo.co.jp/videonews/ann?a=20190716-00000071-ann-int',
+        'info_dict': {
+            'id': '1736242',
+            'ext': 'mp4',
+            'title': 'ムン大統領が対日批判を強化“現金化”効果は?(テレビ朝日系(ANN)) - Yahoo!ニュース',
+            'description': '韓国の元徴用工らを巡る裁判の原告が弁護士が差し押さえた三菱重工業の資産を売却して - Yahoo!ニュース(テレビ朝日系(ANN))',
+            'thumbnail': r're:^https?://.*\.[a-zA-Z\d]{3,4}$',
+        },
+        'params': {
+            'skip_download': True,
+        },
+    }, {
+        # geo restricted
+        'url': 'https://headlines.yahoo.co.jp/hl?a=20190721-00000001-oxv-l04',
+        'only_matching': True,
+    }, {
+        'url': 'https://headlines.yahoo.co.jp/videonews/',
+        'only_matching': True,
+    }, {
+        'url': 'https://news.yahoo.co.jp',
+        'only_matching': True,
+    }, {
+        'url': 'https://news.yahoo.co.jp/byline/hashimotojunji/20190628-00131977/',
+        'only_matching': True,
+    }, {
+        'url': 'https://news.yahoo.co.jp/feature/1356',
+        'only_matching': True
+    }]
+
+    def _extract_formats(self, json_data, content_id):
+        formats = []
+
+        video_data = try_get(
+            json_data,
+            lambda x: x['ResultSet']['Result'][0]['VideoUrlSet']['VideoUrl'],
+            list)
+        for vid in video_data or []:
+            delivery = vid.get('delivery')
+            url = url_or_none(vid.get('Url'))
+            if not delivery or not url:
+                continue
+            elif delivery == 'hls':
+                formats.extend(
+                    self._extract_m3u8_formats(
+                        url, content_id, 'mp4', 'm3u8_native',
+                        m3u8_id='hls', fatal=False))
+            else:
+                formats.append({
+                    'url': url,
+                    'format_id': 'http-%s' % compat_str(vid.get('bitrate', '')),
+                    'height': int_or_none(vid.get('height')),
+                    'width': int_or_none(vid.get('width')),
+                    'tbr': int_or_none(vid.get('bitrate')),
+                })
+        self._remove_duplicate_formats(formats)
+        self._sort_formats(formats)
+
+        return formats
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        host = mobj.group('host')
+        display_id = mobj.group('id') or host
+
+        webpage = self._download_webpage(url, display_id)
+
+        title = self._html_search_meta(
+            ['og:title', 'twitter:title'], webpage, 'title', default=None
+        ) or self._html_search_regex('<title>([^<]+)</title>', webpage, 'title')
+
+        if display_id == host:
+            # Headline page (w/ multiple BC playlists) ('news.yahoo.co.jp', 'headlines.yahoo.co.jp/videonews/', ...)
+            stream_plists = re.findall(r'plist=(\d+)', webpage) or re.findall(r'plist["\']:\s*["\']([^"\']+)', webpage)
+            entries = [
+                self.url_result(
+                    smuggle_url(
+                        'http://players.brightcove.net/5690807595001/HyZNerRl7_default/index.html?playlistId=%s' % plist_id,
+                        {'geo_countries': ['JP']}),
+                    ie='BrightcoveNew', video_id=plist_id)
+                for plist_id in stream_plists]
+            return self.playlist_result(entries, playlist_title=title)
+
+        # Article page
+        description = self._html_search_meta(
+            ['og:description', 'description', 'twitter:description'],
+            webpage, 'description', default=None)
+        thumbnail = self._og_search_thumbnail(
+            webpage, default=None) or self._html_search_meta(
+            'twitter:image', webpage, 'thumbnail', default=None)
+        space_id = self._search_regex([
+            r'<script[^>]+class=["\']yvpub-player["\'][^>]+spaceid=([^&"\']+)',
+            r'YAHOO\.JP\.srch\.\w+link\.onLoad[^;]+spaceID["\' ]*:["\' ]+([^"\']+)',
+            r'<!--\s+SpaceID=(\d+)'
+        ], webpage, 'spaceid')
+
+        content_id = self._search_regex(
+            r'<script[^>]+class=["\']yvpub-player["\'][^>]+contentid=(?P<contentid>[^&"\']+)',
+            webpage, 'contentid', group='contentid')
+
+        json_data = self._download_json(
+            'https://feapi-yvpub.yahooapis.jp/v1/content/%s' % content_id,
+            content_id,
+            query={
+                'appid': 'dj0zaiZpPVZMTVFJR0FwZWpiMyZzPWNvbnN1bWVyc2VjcmV0Jng9YjU-',
+                'output': 'json',
+                'space_id': space_id,
+                'domain': host,
+                'ak': hashlib.md5('_'.join((space_id, host)).encode()).hexdigest(),
+                'device_type': '1100',
+            })
+        formats = self._extract_formats(json_data, content_id)
+
+        return {
+            'id': content_id,
+            'title': title,
+            'description': description,
+            'thumbnail': thumbnail,
+            'formats': formats,
+        }