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

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 from .common import InfoExtractor
   5 from ..utils import parse_iso8601, ExtractorError
   6
   7
   8 class CtsNewsIE(InfoExtractor):
   9     _VALID_URL = r'http://news\.cts\.com\.tw/[a-z]+/[a-z]+/\d+/(?P<id>\d+)\.html'
  10     _TESTS = [{
  11         'url': 'http://news.cts.com.tw/cts/international/201309/201309031304098.html',
  12         'md5': '3aee7e0df7cdff94e43581f54c22619e',
  13         'info_dict': {
  14             'id': '201309031304098',
  15             'ext': 'mp4',
  16             'title': '韓國31歲童顏男 貌如十多歲小孩',
  17             'description': 'md5:f183feeba3752b683827aab71adad584',
  18             'thumbnail': 're:^https?://.*\.jpg$',
  19             'timestamp': 1378205880,
  20             'upload_date': '20130903',
  21         }
  22     }]
  23
  24     def _real_extract(self, url):
  25         news_id = self._match_id(url)
  26         page = self._download_webpage(url, news_id)
  27
  28         if not self._search_regex(r'(CTSPlayer2)', page, 'CTSPlayer2 identifier', fatal=False):
  29             raise ExtractorError('The news includes no videos!')
  30
  31         feed_pattern = r'(http://news.cts.com.tw/action/mp4feed.php\?news_id=\d+)'
  32         feed_url = self._html_search_regex(feed_pattern, page, 'feed url')
  33         feed_page = self._download_webpage(feed_url, news_id)
  34
  35         description = self._html_search_meta('description', page)
  36         title = self._html_search_meta('title', page)
  37         thumbnail = self._html_search_meta('image', page)
  38
  39         datetime_pattern = r'(\d{4}/\d{2}/\d{2} \d{2}:\d{2})'
  40         datetime_str = self._html_search_regex(datetime_pattern, page, 'date and time')
  41         time = (datetime_str + ':00+08:00').replace('/', '-')
  42         timestamp = parse_iso8601(time, delimiter=' ')
  43
  44         return {
  45             'id': news_id,
  46             'title': title,
  47             'description': description,
  48             'url': feed_page,
  49             'thumbnail': thumbnail,
  50             'timestamp': timestamp
  51         }