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