_ Git - youtube-dl/blob - youtube_dl/extractor/nextmedia.py

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