1 from .common
import InfoExtractor
2 from ..compat
import compat_urlparse
13 class NextMediaIE(InfoExtractor
):
15 _VALID_URL
= r
'https?://hk\.apple\.nextmedia\.com/[^/]+/[^/]+/(?P<date>\d+)/(?P<id>\d+)'
17 'url': 'http://hk.apple.nextmedia.com/realtime/news/20141108/53109199',
18 'md5': 'dff9fad7009311c421176d1ac90bfe4f',
22 'title': '【佔領金鐘】50外國領事議員撐場 讚學生勇敢香港有希望',
23 'thumbnail': r
're:^https?://.*\.jpg$',
24 'description': 'md5:28222b9912b6665a21011b034c70fcc7',
25 'timestamp': 1415456273,
26 'upload_date': '20141108',
30 _URL_PATTERN
= r
'\{ url: \'(.+)\' \}'
32 def _real_extract(self
, url
):
33 news_id
= self
._match
_id
(url
)
34 page
= self
._download
_webpage
(url
, news_id
)
35 return self
._extract
_from
_nextmedia
_page
(news_id
, url
, page
)
37 def _extract_from_nextmedia_page(self
, news_id
, url
, page
):
38 redirection_url
= self
._search
_regex
(
39 r
'window\.location\.href\s*=\s*([\'"])(?P<url>(?!\1).+)\1',
40 page, 'redirection URL', default=None, group='url')
42 return self.url_result(compat_urlparse.urljoin(url, redirection_url))
44 title = self._fetch_title(page)
45 video_url = self._search_regex(self._URL_PATTERN, page, 'video url')
50 'url': video_url, # ext can be inferred from url
51 'thumbnail': self._fetch_thumbnail(page),
52 'description': self._fetch_description(page),
55 timestamp = self._fetch_timestamp(page)
57 attrs['timestamp'] = timestamp
59 attrs['upload_date'] = self._fetch_upload_date(url)
63 def _fetch_title(self, page):
64 return self._og_search_title(page)
66 def _fetch_thumbnail(self, page):
67 return self._og_search_thumbnail(page)
69 def _fetch_timestamp(self, page):
70 dateCreated = self._search_regex('"dateCreated
":"([^
"]+)"', page, 'created time
')
71 return parse_iso8601(dateCreated)
73 def _fetch_upload_date(self, url):
74 return self._search_regex(self._VALID_URL, url, 'upload date
', group='date
')
76 def _fetch_description(self, page):
77 return self._og_search_property('description
', page)
80 class NextMediaActionNewsIE(NextMediaIE): # XXX: Do not subclass from concrete IE
81 IE_DESC = '蘋果日報
- 動新聞
'
82 _VALID_URL = r'https?
://hk\
.dv\
.nextmedia\
.com
/actionnews
/[^
/]+/(?P
<date
>\d
+)/(?P
<id>\d
+)/\d
+'
84 'url
': 'http
://hk
.dv
.nextmedia
.com
/actionnews
/hit
/20150121/19009428/20061460',
85 'md5
': '05fce8ffeed7a5e00665d4b7cf0f9201
',
89 'title
': '【壹週刊】細
10年男友偷食
50歲邵美琪再失戀
',
90 'thumbnail
': r're
:^https?
://.*\
.jpg$
',
91 'description
': 'md5
:cd802fad1f40fd9ea178c1e2af02d659
',
92 'timestamp
': 1421791200,
93 'upload_date
': '20150120',
97 def _real_extract(self, url):
98 news_id = self._match_id(url)
99 actionnews_page = self._download_webpage(url, news_id)
100 article_url = self._og_search_url(actionnews_page)
101 article_page = self._download_webpage(article_url, news_id)
102 return self._extract_from_nextmedia_page(news_id, url, article_page)
105 class AppleDailyIE(NextMediaIE): # XXX: Do not subclass from concrete IE
107 _VALID_URL = r'https?
://(www|ent
)\
.appledaily\
.com\
.tw
/[^
/]+/[^
/]+/[^
/]+/(?P
<date
>\d
+)/(?P
<id>\d
+)(/.*)?
'
109 'url
': 'http
://ent
.appledaily
.com
.tw
/enews
/article
/entertainment
/20150128/36354694',
110 'md5
': 'a843ab23d150977cc55ef94f1e2c1e4d
',
114 'title
': '周亭羽走過摩鐵陰霾
2男陪吃 九把刀孤寒看醫生
',
115 'thumbnail
': r're
:^https?
://.*\
.jpg$
',
116 'description
': 'md5
:2acd430e59956dc47cd7f67cb3c003f4
',
117 'upload_date
': '20150128',
120 '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
',
121 'md5
': '86b4e9132d158279c7883822d94ccc49
',
125 'title
': '不滿被踩腳 山東兩大媽一路打下車
',
126 'thumbnail
': r're
:^https?
://.*\
.jpg$
',
127 'description
': 'md5
:175b4260c1d7c085993474217e4ab1b4
',
128 'upload_date
': '20150128',
131 'url
': 'http
://www
.appledaily
.com
.tw
/animation
/realtimenews
/new
/20150128/5003671',
132 'md5
': '03df296d95dedc2d5886debbb80cb43f
',
136 'title
': '20正妹熱舞 《刀龍傳說Online》火辣上市
',
137 'thumbnail
': r're
:^https?
://.*\
.jpg$
',
138 'description
': 'md5
:23c0aac567dc08c9c16a3161a2c2e3cd
',
139 'upload_date
': '20150128',
141 'skip
': 'redirect to http
://www
.appledaily
.com
.tw
/animation
/',
144 'url
': 'http
://www
.appledaily
.com
.tw
/animation
/realtimenews
/new
/20150128/5003673/',
145 'md5
': 'b06182cd386ea7bc6115ec7ff0f72aeb
',
149 'title
': '半夜尿尿 好像會看到___
',
150 'description
': 'md5
:61d2da7fe117fede148706cdb85ac066
',
151 'upload_date
': '20150128',
153 'expected_warnings
': [
156 'skip
': 'redirect to http
://www
.appledaily
.com
.tw
/animation
/',
158 'url
': 'http
://www
.appledaily
.com
.tw
/appledaily
/article
/supplement
/20140417/35770334/',
159 'md5
': 'eaa20e6b9df418c912d7f5dec2ba734d
',
163 'title
': '咖啡占卜測 XU裝熟指數
',
164 'thumbnail
': r're
:^https?
://.*\
.jpg$
',
165 'description
': 'md5
:7b859991a6a4fedbdf3dd3b66545c748
',
166 'upload_date
': '20140417',
169 'url
': 'http
://www
.appledaily
.com
.tw
/actionnews
/appledaily
/7/20161003/960588/',
170 'only_matching
': True,
172 # Redirected from http://ent.appledaily.com.tw/enews/article/entertainment/20150128/36354694
173 'url
': 'http
://ent
.appledaily
.com
.tw
/section
/article
/headline
/20150128/36354694',
174 'only_matching
': True,
177 _URL_PATTERN = r'\{url: \'(.+)\'\}
'
179 def _fetch_title(self, page):
180 return (self._html_search_regex(r'<h1
id="h1">([^
<>]+)</h1
>', page, 'news title
', default=None)
181 or self._html_search_meta('description
', page, 'news title
'))
183 def _fetch_thumbnail(self, page):
184 return self._html_search_regex(r"setInitialImage\(\'([^']+)'\)", page, 'video thumbnail
', fatal=False)
186 def _fetch_timestamp(self, page):
189 def _fetch_description(self, page):
190 return self._html_search_meta('description
', page, 'news description
')
193 class NextTVIE(InfoExtractor):
195 _VALID_URL = r'https?
://(?
:www\
.)?nexttv\
.com\
.tw
/(?
:[^
/]+/)+(?P
<id>\d
+)'
198 'url
': 'http
://www
.nexttv
.com
.tw
/news
/realtime
/politics
/11779671',
202 'title
': '「超收稅」近
4千億! 藍議員籲發消費券
',
203 'thumbnail
': r're
:^https?
://.*\
.jpg$
',
204 'timestamp
': 1484825400,
205 'upload_date
': '20170119',
210 def _real_extract(self, url):
211 video_id = self._match_id(url)
213 webpage = self._download_webpage(url, video_id)
215 title = self._html_search_regex(
216 r'<h1
[^
>]*>([^
<]+)</h1
>', webpage, 'title
')
218 data = self._hidden_inputs(webpage)
220 video_url = data['ntt
-vod
-src
-detailview
']
222 date_str = get_element_by_class('date
', webpage)
223 timestamp = unified_timestamp(date_str + '+0800') if date_str else None
225 view_count = int_or_none(remove_start(
226 clean_html(get_element_by_class('click
', webpage)), '點閱:
'))
232 'thumbnail
': data.get('ntt
-vod
-img
-src
'),
233 'timestamp
': timestamp,
234 'view_count
': view_count,