[yt-dlp.git] / yt_dlp / extractor / nextmedia.py

import urllib.parse

from .common import InfoExtractor
from ..utils import (
    clean_html,
    get_element_by_class,
    int_or_none,
    parse_iso8601,
    remove_start,
    unified_timestamp,
)


class NextMediaIE(InfoExtractor):
    IE_DESC = '蘋果日報'
    _VALID_URL = r'https?://hk\.apple\.nextmedia\.com/[^/]+/[^/]+/(?P<date>\d+)/(?P<id>\d+)'
    _TESTS = [{
        'url': 'http://hk.apple.nextmedia.com/realtime/news/20141108/53109199',
        'md5': 'dff9fad7009311c421176d1ac90bfe4f',
        'info_dict': {
            'id': '53109199',
            'ext': 'mp4',
            'title': '【佔領金鐘】50外國領事議員撐場 讚學生勇敢香港有希望',
            'thumbnail': r're:^https?://.*\.jpg$',
            'description': 'md5:28222b9912b6665a21011b034c70fcc7',
            'timestamp': 1415456273,
            'upload_date': '20141108',
        },
    }]

    _URL_PATTERN = r'\{ url: \'(.+)\' \}'

    def _real_extract(self, url):
        news_id = self._match_id(url)
        page = self._download_webpage(url, news_id)
        return self._extract_from_nextmedia_page(news_id, url, page)

    def _extract_from_nextmedia_page(self, news_id, url, page):
        redirection_url = self._search_regex(
            r'window\.location\.href\s*=\s*([\'"])(?P<url>(?!\1).+)\1',
            page, 'redirection URL', default=None, group='url')
        if redirection_url:
            return self.url_result(urllib.parse.urljoin(url, redirection_url))

        title = self._fetch_title(page)
        video_url = self._search_regex(self._URL_PATTERN, page, 'video url')

        attrs = {
            'id': news_id,
            'title': title,
            'url': video_url,  # ext can be inferred from url
            'thumbnail': self._fetch_thumbnail(page),
            'description': self._fetch_description(page),
        }

        timestamp = self._fetch_timestamp(page)
        if timestamp:
            attrs['timestamp'] = timestamp
        else:
            attrs['upload_date'] = self._fetch_upload_date(url)

        return attrs

    def _fetch_title(self, page):
        return self._og_search_title(page)

    def _fetch_thumbnail(self, page):
        return self._og_search_thumbnail(page)

    def _fetch_timestamp(self, page):
        date_created = self._search_regex('"dateCreated":"([^"]+)"', page, 'created time')
        return parse_iso8601(date_created)

    def _fetch_upload_date(self, url):
        return self._search_regex(self._VALID_URL, url, 'upload date', group='date')

    def _fetch_description(self, page):
        return self._og_search_property('description', page)


class NextMediaActionNewsIE(NextMediaIE):  # XXX: Do not subclass from concrete IE
    IE_DESC = '蘋果日報 - 動新聞'
    _VALID_URL = r'https?://hk\.dv\.nextmedia\.com/actionnews/[^/]+/(?P<date>\d+)/(?P<id>\d+)/\d+'
    _TESTS = [{
        'url': 'http://hk.dv.nextmedia.com/actionnews/hit/20150121/19009428/20061460',
        'md5': '05fce8ffeed7a5e00665d4b7cf0f9201',
        'info_dict': {
            'id': '19009428',
            'ext': 'mp4',
            'title': '【壹週刊】細10年男友偷食　50歲邵美琪再失戀',
            'thumbnail': r're:^https?://.*\.jpg$',
            'description': 'md5:cd802fad1f40fd9ea178c1e2af02d659',
            'timestamp': 1421791200,
            'upload_date': '20150120',
        },
    }]

    def _real_extract(self, url):
        news_id = self._match_id(url)
        actionnews_page = self._download_webpage(url, news_id)
        article_url = self._og_search_url(actionnews_page)
        article_page = self._download_webpage(article_url, news_id)
        return self._extract_from_nextmedia_page(news_id, url, article_page)


class AppleDailyIE(NextMediaIE):  # XXX: Do not subclass from concrete IE
    IE_DESC = '臺灣蘋果日報'
    _VALID_URL = r'https?://(www|ent)\.appledaily\.com\.tw/[^/]+/[^/]+/[^/]+/(?P<date>\d+)/(?P<id>\d+)(/.*)?'
    _TESTS = [{
        'url': 'http://ent.appledaily.com.tw/enews/article/entertainment/20150128/36354694',
        'md5': 'a843ab23d150977cc55ef94f1e2c1e4d',
        'info_dict': {
            'id': '36354694',
            'ext': 'mp4',
            'title': '周亭羽走過摩鐵陰霾2男陪吃 九把刀孤寒看醫生',
            'thumbnail': r're:^https?://.*\.jpg$',
            'description': 'md5:2acd430e59956dc47cd7f67cb3c003f4',
            'upload_date': '20150128',
        },
    }, {
        'url': 'http://www.appledaily.com.tw/realtimenews/article/strange/20150128/550549/%E4%B8%8D%E6%BB%BF%E8%A2%AB%E8%B8%A9%E8%85%B3%E3%80%80%E5%B1%B1%E6%9D%B1%E5%85%A9%E5%A4%A7%E5%AA%BD%E4%B8%80%E8%B7%AF%E6%89%93%E4%B8%8B%E8%BB%8A',
        'md5': '86b4e9132d158279c7883822d94ccc49',
        'info_dict': {
            'id': '550549',
            'ext': 'mp4',
            'title': '不滿被踩腳　山東兩大媽一路打下車',
            'thumbnail': r're:^https?://.*\.jpg$',
            'description': 'md5:175b4260c1d7c085993474217e4ab1b4',
            'upload_date': '20150128',
        },
    }, {
        'url': 'http://www.appledaily.com.tw/animation/realtimenews/new/20150128/5003671',
        'md5': '03df296d95dedc2d5886debbb80cb43f',
        'info_dict': {
            'id': '5003671',
            'ext': 'mp4',
            'title': '20正妹熱舞　《刀龍傳說Online》火辣上市',
            'thumbnail': r're:^https?://.*\.jpg$',
            'description': 'md5:23c0aac567dc08c9c16a3161a2c2e3cd',
            'upload_date': '20150128',
        },
        'skip': 'redirect to http://www.appledaily.com.tw/animation/',
    }, {
        # No thumbnail
        'url': 'http://www.appledaily.com.tw/animation/realtimenews/new/20150128/5003673/',
        'md5': 'b06182cd386ea7bc6115ec7ff0f72aeb',
        'info_dict': {
            'id': '5003673',
            'ext': 'mp4',
            'title': '半夜尿尿　好像會看到___',
            'description': 'md5:61d2da7fe117fede148706cdb85ac066',
            'upload_date': '20150128',
        },
        'expected_warnings': [
            'video thumbnail',
        ],
        'skip': 'redirect to http://www.appledaily.com.tw/animation/',
    }, {
        'url': 'http://www.appledaily.com.tw/appledaily/article/supplement/20140417/35770334/',
        'md5': 'eaa20e6b9df418c912d7f5dec2ba734d',
        'info_dict': {
            'id': '35770334',
            'ext': 'mp4',
            'title': '咖啡占卜測 XU裝熟指數',
            'thumbnail': r're:^https?://.*\.jpg$',
            'description': 'md5:7b859991a6a4fedbdf3dd3b66545c748',
            'upload_date': '20140417',
        },
    }, {
        'url': 'http://www.appledaily.com.tw/actionnews/appledaily/7/20161003/960588/',
        'only_matching': True,
    }, {
        # Redirected from http://ent.appledaily.com.tw/enews/article/entertainment/20150128/36354694
        'url': 'http://ent.appledaily.com.tw/section/article/headline/20150128/36354694',
        'only_matching': True,
    }]

    _URL_PATTERN = r'\{url: \'(.+)\'\}'

    def _fetch_title(self, page):
        return (self._html_search_regex(r'<h1 id="h1">([^<>]+)</h1>', page, 'news title', default=None)
                or self._html_search_meta('description', page, 'news title'))

    def _fetch_thumbnail(self, page):
        return self._html_search_regex(r"setInitialImage\(\'([^']+)'\)", page, 'video thumbnail', fatal=False)

    def _fetch_timestamp(self, page):
        return None

    def _fetch_description(self, page):
        return self._html_search_meta('description', page, 'news description')


class NextTVIE(InfoExtractor):
    _WORKING = False
    _ENABLED = None  # XXX: pass through to GenericIE
    IE_DESC = '壹電視'
    _VALID_URL = r'https?://(?:www\.)?nexttv\.com\.tw/(?:[^/]+/)+(?P<id>\d+)'

    _TEST = {
        'url': 'http://www.nexttv.com.tw/news/realtime/politics/11779671',
        'info_dict': {
            'id': '11779671',
            'ext': 'mp4',
            'title': '「超收稅」近4千億！　藍議員籲發消費券',
            'thumbnail': r're:^https?://.*\.jpg$',
            'timestamp': 1484825400,
            'upload_date': '20170119',
            'view_count': int,
        },
    }

    def _real_extract(self, url):
        video_id = self._match_id(url)

        webpage = self._download_webpage(url, video_id)

        title = self._html_search_regex(
            r'<h1[^>]*>([^<]+)</h1>', webpage, 'title')

        data = self._hidden_inputs(webpage)

        video_url = data['ntt-vod-src-detailview']

        date_str = get_element_by_class('date', webpage)
        timestamp = unified_timestamp(date_str + '+0800') if date_str else None

        view_count = int_or_none(remove_start(
            clean_html(get_element_by_class('click', webpage)), '點閱：'))

        return {
            'id': video_id,
            'title': title,
            'url': video_url,
            'thumbnail': data.get('ntt-vod-img-src'),
            'timestamp': timestamp,
            'view_count': view_count,
        }
Commit	Line	Data
add96eb9	1	import urllib.parse
add96eb9	2
206dba27	3	from .common import InfoExtractor
bc35ed3f YCH	4	from ..utils import (
	5	clean_html,
	6	get_element_by_class,
	7	int_or_none,
	8	parse_iso8601,
	9	remove_start,
	10	unified_timestamp,
	11	)
206dba27 YCH	12
	13
	14	class NextMediaIE(InfoExtractor):
9d16788a	15	IE_DESC = '蘋果日報'
92519402	16	_VALID_URL = r'https?://hk\.apple\.nextmedia\.com/[^/]+/[^/]+/(?P<date>\d+)/(?P<id>\d+)'
206dba27 YCH	17	_TESTS = [{
	18	'url': 'http://hk.apple.nextmedia.com/realtime/news/20141108/53109199',
	19	'md5': 'dff9fad7009311c421176d1ac90bfe4f',
	20	'info_dict': {
	21	'id': '53109199',
	22	'ext': 'mp4',
	23	'title': '【佔領金鐘】50外國領事議員撐場讚學生勇敢香港有希望',
ec85ded8	24	'thumbnail': r're:^https?://.*\.jpg$',
206dba27 YCH	25	'description': 'md5:28222b9912b6665a21011b034c70fcc7',
	26	'timestamp': 1415456273,
	27	'upload_date': '20141108',
add96eb9	28	},
206dba27 YCH	29	}]
	30
	31	_URL_PATTERN = r'\{ url: \'(.+)\' \}'
	32
	33	def _real_extract(self, url):
	34	news_id = self._match_id(url)
	35	page = self._download_webpage(url, news_id)
	36	return self._extract_from_nextmedia_page(news_id, url, page)
	37
	38	def _extract_from_nextmedia_page(self, news_id, url, page):
7c20b748 YCH	39	redirection_url = self._search_regex(
	40	r'window\.location\.href\s=\s([\'"])(?P<url>(?!\1).+)\1',
	41	page, 'redirection URL', default=None, group='url')
	42	if redirection_url:
add96eb9	43	return self.url_result(urllib.parse.urljoin(url, redirection_url))
7c20b748	44
206dba27 YCH	45	title = self._fetch_title(page)
	46	video_url = self._search_regex(self._URL_PATTERN, page, 'video url')
	47
	48	attrs = {
	49	'id': news_id,
	50	'title': title,
	51	'url': video_url, # ext can be inferred from url
	52	'thumbnail': self._fetch_thumbnail(page),
	53	'description': self._fetch_description(page),
	54	}
	55
	56	timestamp = self._fetch_timestamp(page)
	57	if timestamp:
	58	attrs['timestamp'] = timestamp
	59	else:
	60	attrs['upload_date'] = self._fetch_upload_date(url)
	61
	62	return attrs
	63
	64	def _fetch_title(self, page):
	65	return self._og_search_title(page)
	66
	67	def _fetch_thumbnail(self, page):
	68	return self._og_search_thumbnail(page)
	69
	70	def _fetch_timestamp(self, page):
add96eb9	71	date_created = self._search_regex('"dateCreated":"([^"]+)"', page, 'created time')
add96eb9	72	return parse_iso8601(date_created)
206dba27 YCH	73
	74	def _fetch_upload_date(self, url):
	75	return self._search_regex(self._VALID_URL, url, 'upload date', group='date')
	76
	77	def _fetch_description(self, page):
	78	return self._og_search_property('description', page)
	79
	80
6368e2e6	81	class NextMediaActionNewsIE(NextMediaIE): # XXX: Do not subclass from concrete IE
9d16788a	82	IE_DESC = '蘋果日報 - 動新聞'
92519402	83	_VALID_URL = r'https?://hk\.dv\.nextmedia\.com/actionnews/[^/]+/(?P<date>\d+)/(?P<id>\d+)/\d+'
206dba27 YCH	84	_TESTS = [{
	85	'url': 'http://hk.dv.nextmedia.com/actionnews/hit/20150121/19009428/20061460',
	86	'md5': '05fce8ffeed7a5e00665d4b7cf0f9201',
	87	'info_dict': {
	88	'id': '19009428',
	89	'ext': 'mp4',
	90	'title': '【壹週刊】細10年男友偷食　50歲邵美琪再失戀',
ec85ded8	91	'thumbnail': r're:^https?://.*\.jpg$',
206dba27 YCH	92	'description': 'md5:cd802fad1f40fd9ea178c1e2af02d659',
	93	'timestamp': 1421791200,
	94	'upload_date': '20150120',
add96eb9	95	},
206dba27 YCH	96	}]
	97
	98	def _real_extract(self, url):
	99	news_id = self._match_id(url)
	100	actionnews_page = self._download_webpage(url, news_id)
	101	article_url = self._og_search_url(actionnews_page)
	102	article_page = self._download_webpage(article_url, news_id)
	103	return self._extract_from_nextmedia_page(news_id, url, article_page)
	104
	105
6368e2e6	106	class AppleDailyIE(NextMediaIE): # XXX: Do not subclass from concrete IE
9d16788a	107	IE_DESC = '臺灣蘋果日報'
7c20b748	108	_VALID_URL = r'https?://(www\|ent)\.appledaily\.com\.tw/[^/]+/[^/]+/[^/]+/(?P<date>\d+)/(?P<id>\d+)(/.*)?'
206dba27 YCH	109	_TESTS = [{
	110	'url': 'http://ent.appledaily.com.tw/enews/article/entertainment/20150128/36354694',
	111	'md5': 'a843ab23d150977cc55ef94f1e2c1e4d',
	112	'info_dict': {
	113	'id': '36354694',
	114	'ext': 'mp4',
	115	'title': '周亭羽走過摩鐵陰霾2男陪吃九把刀孤寒看醫生',
ec85ded8	116	'thumbnail': r're:^https?://.*\.jpg$',
9bf87ae3	117	'description': 'md5:2acd430e59956dc47cd7f67cb3c003f4',
206dba27	118	'upload_date': '20150128',
add96eb9	119	},
206dba27 YCH	120	}, {
	121	'url': 'http://www.appledaily.com.tw/realtimenews/article/strange/20150128/550549/%E4%B8%8D%E6%BB%BF%E8%A2%AB%E8%B8%A9%E8%85%B3%E3%80%80%E5%B1%B1%E6%9D%B1%E5%85%A9%E5%A4%A7%E5%AA%BD%E4%B8%80%E8%B7%AF%E6%89%93%E4%B8%8B%E8%BB%8A',
	122	'md5': '86b4e9132d158279c7883822d94ccc49',
	123	'info_dict': {
	124	'id': '550549',
	125	'ext': 'mp4',
	126	'title': '不滿被踩腳　山東兩大媽一路打下車',
ec85ded8	127	'thumbnail': r're:^https?://.*\.jpg$',
9bf87ae3	128	'description': 'md5:175b4260c1d7c085993474217e4ab1b4',
206dba27	129	'upload_date': '20150128',
add96eb9	130	},
9bf87ae3	131	}, {
206dba27 YCH	132	'url': 'http://www.appledaily.com.tw/animation/realtimenews/new/20150128/5003671',
	133	'md5': '03df296d95dedc2d5886debbb80cb43f',
	134	'info_dict': {
	135	'id': '5003671',
	136	'ext': 'mp4',
	137	'title': '20正妹熱舞　《刀龍傳說Online》火辣上市',
ec85ded8	138	'thumbnail': r're:^https?://.*\.jpg$',
206dba27 YCH	139	'description': 'md5:23c0aac567dc08c9c16a3161a2c2e3cd',
206dba27 YCH	140	'upload_date': '20150128',
1f9fb20f	141	},
1f9fb20f	142	'skip': 'redirect to http://www.appledaily.com.tw/animation/',
206dba27 YCH	143	}, {
	144	# No thumbnail
	145	'url': 'http://www.appledaily.com.tw/animation/realtimenews/new/20150128/5003673/',
	146	'md5': 'b06182cd386ea7bc6115ec7ff0f72aeb',
	147	'info_dict': {
	148	'id': '5003673',
	149	'ext': 'mp4',
	150	'title': '半夜尿尿　好像會看到___',
	151	'description': 'md5:61d2da7fe117fede148706cdb85ac066',
	152	'upload_date': '20150128',
	153	},
	154	'expected_warnings': [
	155	'video thumbnail',
1f9fb20f	156	],
1f9fb20f	157	'skip': 'redirect to http://www.appledaily.com.tw/animation/',
30455ce2 YCH	158	}, {
30455ce2 YCH	159	'url': 'http://www.appledaily.com.tw/appledaily/article/supplement/20140417/35770334/',
1f9fb20f	160	'md5': 'eaa20e6b9df418c912d7f5dec2ba734d',
	161	'info_dict': {
	162	'id': '35770334',
	163	'ext': 'mp4',
	164	'title': '咖啡占卜測 XU裝熟指數',
ec85ded8	165	'thumbnail': r're:^https?://.*\.jpg$',
1f9fb20f	166	'description': 'md5:7b859991a6a4fedbdf3dd3b66545c748',
	167	'upload_date': '20140417',
	168	},
b0082629 YCH	169	}, {
	170	'url': 'http://www.appledaily.com.tw/actionnews/appledaily/7/20161003/960588/',
	171	'only_matching': True,
7c20b748 YCH	172	}, {
	173	# Redirected from http://ent.appledaily.com.tw/enews/article/entertainment/20150128/36354694
	174	'url': 'http://ent.appledaily.com.tw/section/article/headline/20150128/36354694',
	175	'only_matching': True,
206dba27 YCH	176	}]
206dba27 YCH	177
9bf87ae3 YCH	178	_URL_PATTERN = r'\{url: \'(.+)\'\}'
9bf87ae3 YCH	179
206dba27	180	def _fetch_title(self, page):
3089bc74 S	181	return (self._html_search_regex(r'<h1 id="h1">([^<>]+)</h1>', page, 'news title', default=None)
3089bc74 S	182	or self._html_search_meta('description', page, 'news title'))
9bf87ae3 YCH	183
	184	def _fetch_thumbnail(self, page):
	185	return self._html_search_regex(r"setInitialImage\(\'([^']+)'\)", page, 'video thumbnail', fatal=False)
	186
	187	def _fetch_timestamp(self, page):
	188	return None
206dba27 YCH	189
	190	def _fetch_description(self, page):
	191	return self._html_search_meta('description', page, 'news description')
bc35ed3f YCH	192
	193
	194	class NextTVIE(InfoExtractor):
df773c3d	195	_WORKING = False
df773c3d	196	_ENABLED = None # XXX: pass through to GenericIE
bc35ed3f YCH	197	IE_DESC = '壹電視'
	198	_VALID_URL = r'https?://(?:www\.)?nexttv\.com\.tw/(?:[^/]+/)+(?P<id>\d+)'
	199
	200	_TEST = {
	201	'url': 'http://www.nexttv.com.tw/news/realtime/politics/11779671',
	202	'info_dict': {
	203	'id': '11779671',
	204	'ext': 'mp4',
	205	'title': '「超收稅」近4千億！　藍議員籲發消費券',
	206	'thumbnail': r're:^https?://.*\.jpg$',
	207	'timestamp': 1484825400,
	208	'upload_date': '20170119',
	209	'view_count': int,
	210	},
	211	}
	212
	213	def _real_extract(self, url):
	214	video_id = self._match_id(url)
	215
	216	webpage = self._download_webpage(url, video_id)
	217
	218	title = self._html_search_regex(
	219	r'<h1[^>]*>([^<]+)</h1>', webpage, 'title')
	220
	221	data = self._hidden_inputs(webpage)
	222
	223	video_url = data['ntt-vod-src-detailview']
	224
	225	date_str = get_element_by_class('date', webpage)
	226	timestamp = unified_timestamp(date_str + '+0800') if date_str else None
	227
	228	view_count = int_or_none(remove_start(
	229	clean_html(get_element_by_class('click', webpage)), '點閱：'))
	230
	231	return {
	232	'id': video_id,
	233	'title': title,
	234	'url': video_url,
	235	'thumbnail': data.get('ntt-vod-img-src'),
	236	'timestamp': timestamp,
	237	'view_count': view_count,
	238	}