youtube_dl/extractor/nextmedia.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 from .common import InfoExtractor
   5 from ..compat import compat_urlparse
   6 from ..utils import parse_iso8601
   7
   8
   9 class NextMediaIE(InfoExtractor):
  10     IE_DESC = '蘋果日報'
  11     _VALID_URL = r'https?://hk\.apple\.nextmedia\.com/[^/]+/[^/]+/(?P<date>\d+)/(?P<id>\d+)'
  12     _TESTS = [{
  13         'url': 'http://hk.apple.nextmedia.com/realtime/news/20141108/53109199',
  14         'md5': 'dff9fad7009311c421176d1ac90bfe4f',
  15         'info_dict': {
  16             'id': '53109199',
  17             'ext': 'mp4',
  18             'title': '【佔領金鐘】50外國領事議員撐場 讚學生勇敢香港有希望',
  19             'thumbnail': r're:^https?://.*\.jpg$',
  20             'description': 'md5:28222b9912b6665a21011b034c70fcc7',
  21             'timestamp': 1415456273,
  22             'upload_date': '20141108',
  23         }
  24     }]
  25
  26     _URL_PATTERN = r'\{ url: \'(.+)\' \}'
  27
  28     def _real_extract(self, url):
  29         news_id = self._match_id(url)
  30         page = self._download_webpage(url, news_id)
  31         return self._extract_from_nextmedia_page(news_id, url, page)
  32
  33     def _extract_from_nextmedia_page(self, news_id, url, page):
  34         redirection_url = self._search_regex(
  35             r'window\.location\.href\s*=\s*([\'"])(?P<url>(?!\1).+)\1',
  36             page, 'redirection URL', default=None, group='url')
  37         if redirection_url:
  38             return self.url_result(compat_urlparse.urljoin(url, redirection_url))
  39
  40         title = self._fetch_title(page)
  41         video_url = self._search_regex(self._URL_PATTERN, page, 'video url')
  42
  43         attrs = {
  44             'id': news_id,
  45             'title': title,
  46             'url': video_url,  # ext can be inferred from url
  47             'thumbnail': self._fetch_thumbnail(page),
  48             'description': self._fetch_description(page),
  49         }
  50
  51         timestamp = self._fetch_timestamp(page)
  52         if timestamp:
  53             attrs['timestamp'] = timestamp
  54         else:
  55             attrs['upload_date'] = self._fetch_upload_date(url)
  56
  57         return attrs
  58
  59     def _fetch_title(self, page):
  60         return self._og_search_title(page)
  61
  62     def _fetch_thumbnail(self, page):
  63         return self._og_search_thumbnail(page)
  64
  65     def _fetch_timestamp(self, page):
  66         dateCreated = self._search_regex('"dateCreated":"([^"]+)"', page, 'created time')
  67         return parse_iso8601(dateCreated)
  68
  69     def _fetch_upload_date(self, url):
  70         return self._search_regex(self._VALID_URL, url, 'upload date', group='date')
  71
  72     def _fetch_description(self, page):
  73         return self._og_search_property('description', page)
  74
  75
  76 class NextMediaActionNewsIE(NextMediaIE):
  77     IE_DESC = '蘋果日報 - 動新聞'
  78     _VALID_URL = r'https?://hk\.dv\.nextmedia\.com/actionnews/[^/]+/(?P<date>\d+)/(?P<id>\d+)/\d+'
  79     _TESTS = [{
  80         'url': 'http://hk.dv.nextmedia.com/actionnews/hit/20150121/19009428/20061460',
  81         'md5': '05fce8ffeed7a5e00665d4b7cf0f9201',
  82         'info_dict': {
  83             'id': '19009428',
  84             'ext': 'mp4',
  85             'title': '【壹週刊】細10年男友偷食　50歲邵美琪再失戀',
  86             'thumbnail': r're:^https?://.*\.jpg$',
  87             'description': 'md5:cd802fad1f40fd9ea178c1e2af02d659',
  88             'timestamp': 1421791200,
  89             'upload_date': '20150120',
  90         }
  91     }]
  92
  93     def _real_extract(self, url):
  94         news_id = self._match_id(url)
  95         actionnews_page = self._download_webpage(url, news_id)
  96         article_url = self._og_search_url(actionnews_page)
  97         article_page = self._download_webpage(article_url, news_id)
  98         return self._extract_from_nextmedia_page(news_id, url, article_page)
  99
 100
 101 class AppleDailyIE(NextMediaIE):
 102     IE_DESC = '臺灣蘋果日報'
 103     _VALID_URL = r'https?://(www|ent)\.appledaily\.com\.tw/[^/]+/[^/]+/[^/]+/(?P<date>\d+)/(?P<id>\d+)(/.*)?'
 104     _TESTS = [{
 105         'url': 'http://ent.appledaily.com.tw/enews/article/entertainment/20150128/36354694',
 106         'md5': 'a843ab23d150977cc55ef94f1e2c1e4d',
 107         'info_dict': {
 108             'id': '36354694',
 109             'ext': 'mp4',
 110             'title': '周亭羽走過摩鐵陰霾2男陪吃 九把刀孤寒看醫生',
 111             'thumbnail': r're:^https?://.*\.jpg$',
 112             'description': 'md5:2acd430e59956dc47cd7f67cb3c003f4',
 113             'upload_date': '20150128',
 114         }
 115     }, {
 116         'url': 'http://www.appledaily.com.tw/realtimenews/article/strange/20150128/550549/%E4%B8%8D%E6%BB%BF%E8%A2%AB%E8%B8%A9%E8%85%B3%E3%80%80%E5%B1%B1%E6%9D%B1%E5%85%A9%E5%A4%A7%E5%AA%BD%E4%B8%80%E8%B7%AF%E6%89%93%E4%B8%8B%E8%BB%8A',
 117         'md5': '86b4e9132d158279c7883822d94ccc49',
 118         'info_dict': {
 119             'id': '550549',
 120             'ext': 'mp4',
 121             'title': '不滿被踩腳　山東兩大媽一路打下車',
 122             'thumbnail': r're:^https?://.*\.jpg$',
 123             'description': 'md5:175b4260c1d7c085993474217e4ab1b4',
 124             'upload_date': '20150128',
 125         }
 126     }, {
 127         'url': 'http://www.appledaily.com.tw/animation/realtimenews/new/20150128/5003671',
 128         'md5': '03df296d95dedc2d5886debbb80cb43f',
 129         'info_dict': {
 130             'id': '5003671',
 131             'ext': 'mp4',
 132             'title': '20正妹熱舞　《刀龍傳說Online》火辣上市',
 133             'thumbnail': r're:^https?://.*\.jpg$',
 134             'description': 'md5:23c0aac567dc08c9c16a3161a2c2e3cd',
 135             'upload_date': '20150128',
 136         },
 137         'skip': 'redirect to http://www.appledaily.com.tw/animation/',
 138     }, {
 139         # No thumbnail
 140         'url': 'http://www.appledaily.com.tw/animation/realtimenews/new/20150128/5003673/',
 141         'md5': 'b06182cd386ea7bc6115ec7ff0f72aeb',
 142         'info_dict': {
 143             'id': '5003673',
 144             'ext': 'mp4',
 145             'title': '半夜尿尿　好像會看到___',
 146             'description': 'md5:61d2da7fe117fede148706cdb85ac066',
 147             'upload_date': '20150128',
 148         },
 149         'expected_warnings': [
 150             'video thumbnail',
 151         ],
 152         'skip': 'redirect to http://www.appledaily.com.tw/animation/',
 153     }, {
 154         'url': 'http://www.appledaily.com.tw/appledaily/article/supplement/20140417/35770334/',
 155         'md5': 'eaa20e6b9df418c912d7f5dec2ba734d',
 156         'info_dict': {
 157             'id': '35770334',
 158             'ext': 'mp4',
 159             'title': '咖啡占卜測 XU裝熟指數',
 160             'thumbnail': r're:^https?://.*\.jpg$',
 161             'description': 'md5:7b859991a6a4fedbdf3dd3b66545c748',
 162             'upload_date': '20140417',
 163         },
 164     }, {
 165         'url': 'http://www.appledaily.com.tw/actionnews/appledaily/7/20161003/960588/',
 166         'only_matching': True,
 167     }, {
 168         # Redirected from http://ent.appledaily.com.tw/enews/article/entertainment/20150128/36354694
 169         'url': 'http://ent.appledaily.com.tw/section/article/headline/20150128/36354694',
 170         'only_matching': True,
 171     }]
 172
 173     _URL_PATTERN = r'\{url: \'(.+)\'\}'
 174
 175     def _fetch_title(self, page):
 176         return (self._html_search_regex(r'<h1 id="h1">([^<>]+)</h1>', page, 'news title', default=None) or
 177                 self._html_search_meta('description', page, 'news title'))
 178
 179     def _fetch_thumbnail(self, page):
 180         return self._html_search_regex(r"setInitialImage\(\'([^']+)'\)", page, 'video thumbnail', fatal=False)
 181
 182     def _fetch_timestamp(self, page):
 183         return None
 184
 185     def _fetch_description(self, page):
 186         return self._html_search_meta('description', page, 'news description')