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

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 from .common import InfoExtractor
   5 from ..compat import compat_str
   6 from ..utils import (
   7     clean_html,
   8     ExtractorError,
   9     int_or_none,
  10     merge_dicts,
  11     str_or_none,
  12     str_to_int,
  13     try_get,
  14     unified_strdate,
  15     urlencode_postdata,
  16     urljoin,
  17 )
  18
  19
  20 class HKETVIE(InfoExtractor):
  21     IE_NAME = 'hketv'
  22     IE_DESC = '香港教育局教育電視 (HKETV) Educational Television, Hong Kong Educational Bureau'
  23     _GEO_BYPASS = False
  24     _GEO_COUNTRIES = ['HK']
  25     _VALID_URL = r'https?://(?:www\.)?hkedcity\.net/etv/resource/(?P<id>[0-9]+)'
  26     _TESTS = [{
  27         'url': 'https://www.hkedcity.net/etv/resource/2932360618',
  28         'md5': 'f193712f5f7abb208ddef3c5ea6ed0b7',
  29         'info_dict': {
  30             'id': '2932360618',
  31             'ext': 'mp4',
  32             'title': '喜閱一生(共享閱讀樂) (中、英文字幕可供選擇)',
  33             'description': '本節目輯錄了「閱讀滿Fun嘉年華」和「二○一八響應世界閱讀日――悅愛閱讀・愈讀愈愛」的活動花絮，並由學者、作家、演藝界人士等，分享培養子女閱讀興趣和習慣的方法，以及呼籲大家一同分享閱讀的樂趣。',
  34             'upload_date': '20181024',
  35             'duration': 900,
  36             'subtitles': {
  37                 'en': [{
  38                     'url': 'https://apps.hkedcity.net/media/mediaplayer/caption.php?f=74395&lang=en',
  39                     'ext': 'srt',
  40                 }],
  41                 'zh-Hant': [{
  42                     'url': 'https://apps.hkedcity.net/media/mediaplayer/caption.php?f=74395&lang=qmt',
  43                     'ext': 'srt',
  44                 }],
  45             }
  46         },
  47     }, {
  48         'url': 'https://www.hkedcity.net/etv/resource/972641418',
  49         'md5': '1ed494c1c6cf7866a8290edad9b07dc9',
  50         'info_dict': {
  51             'id': '972641418',
  52             'ext': 'mp4',
  53             'title': '衣冠楚楚 (天使系列之一)',
  54             'description': '天國仙境，有兩位可愛的天使小姐妹。她們對幾千年來天使衣著一成不變頗有不滿。她們下望人世間：只見人們穿著七彩繽紛、款式各異的服裝，漂亮極了。天使姐妹決定下凡考察衣著，以設計天使新裝。  下到人間，姐妹試穿各式各樣的衣著，引發連串奇特有趣的情節：她們穿著校服在街上閒逛時，被女警誤認為逃學而送回學校，到校後又被體育老師誤認為是新同學，匆匆忙忙換上運動服後在操場上大顯神通。她們穿著護士服在醫院散步時，又被誤認為當班護士，而投入追尋失蹤病童、治病救人的工作中去。姐妹倆還到過玩具店，與布娃娃們談論衣著。她們也去過服裝設計學校，被當成小模特兒而試穿各式服裝。最令姐妹倆興奮的是一場盛大的民族服裝表演會。身穿盛裝的十二個民族的少女在台上翩翩起舞，各種服飾七彩繽紛、美不勝收。姐妹們情不自禁地穿上民族服裝，小天使變成了少數民族姑娘……最後天使姐妹回到天上，對於天使究竟穿甚麼樣的衣服好，她們還是拿不定主意。  節目通過天使姐妹的奇特經歷，反復示範各式衣服鞋襪的正確讀音及談論衣著時的常用句式，並以盛大的民族服裝表演活動，帶出有關服裝的文化知識。內容豐富而饒有趣味。',
  55             'upload_date': '20070109',
  56             'duration': 907,
  57             'subtitles': {},
  58         },
  59     }]
  60
  61     _CC_LANGS = {
  62         '中文（繁體中文）': 'zh-Hant',
  63         '中文（简体中文）': 'zh-Hans',
  64         'English': 'en',
  65         'Bahasa Indonesia': 'id',
  66         '\u0939\u093f\u0928\u094d\u0926\u0940': 'hi',
  67         '\u0928\u0947\u092a\u093e\u0932\u0940': 'ne',
  68         'Tagalog': 'tl',
  69         '\u0e44\u0e17\u0e22': 'th',
  70         '\u0627\u0631\u062f\u0648': 'ur',
  71     }
  72
  73     def _real_extract(self, url):
  74         video_id = self._match_id(url)
  75         webpage = self._download_webpage(url, video_id)
  76         title = self._html_search_meta('ed_title', webpage, fatal=True)
  77
  78         file_id = self._html_search_regex(r'post_var\["file_id"\]\s*=\s*(.+?);', webpage, 'file ID')
  79         curr_url = self._html_search_regex(r'post_var\["curr_url"\]\s*=\s*"(.+?)";', webpage, 'curr URL')
  80         data = {
  81             'action': 'get_info',
  82             'curr_url': curr_url,
  83             'file_id': file_id,
  84             'video_url': file_id,
  85         }
  86         _APPS_BASE_URL = 'https://apps.hkedcity.net'
  87         handler_url = _APPS_BASE_URL + '/media/play/handler.php'
  88
  89         response = self._download_json(
  90             handler_url, video_id,
  91             data=urlencode_postdata(data),
  92             headers=merge_dicts({'Content-Type': 'application/x-www-form-urlencoded'},
  93                                 self.geo_verification_headers()))
  94
  95         result = response['result']
  96
  97         formats = []
  98         subtitles = {}
  99
 100         if response.get('success') and response.get('access'):
 101             width = int_or_none(result.get('width'))
 102             height = int_or_none(result.get('height'))
 103
 104             playlist0 = try_get(result, lambda x: x['playlist'][0], dict)
 105             fmts = playlist0.get('sources')
 106             for fmt in fmts:
 107                 file_path = fmt.get('file')
 108                 if file_path:
 109                     file_url = urljoin(_APPS_BASE_URL, file_path)
 110                     # If we ever wanted to provide the final resolved URL that
 111                     # does not require cookies, albeit with a shorter lifespan:
 112                     #     urlh = self._downloader.urlopen(file_url)
 113                     #     resolved_url = urlh.geturl()
 114
 115                     label = fmt.get('label')
 116                     w = None
 117                     h = None
 118                     if label == 'HD':
 119                         h = 720
 120                     elif label == 'SD':
 121                         h = 360
 122                     if h:
 123                         if width and height:
 124                             w = h * width // height
 125                         else:
 126                             w = h * 4 // 3
 127
 128                     formats.append({
 129                         'format_id': label,
 130                         'ext': fmt.get('type'),
 131                         'url': file_url,
 132                         'width': w,
 133                         'height': h,
 134                     })
 135
 136             tracks = playlist0.get('tracks', [])
 137             for track in tracks:
 138                 if not isinstance(track, dict):
 139                     continue
 140                 track_kind = str_or_none(track.get('kind'))
 141                 if not track_kind or not isinstance(track_kind, compat_str):
 142                     continue
 143                 if track_kind.lower() not in ('captions', 'subtitles'):
 144                     continue
 145                 track_url = urljoin(_APPS_BASE_URL, track.get('file'))
 146                 if not track_url:
 147                     continue
 148                 track_label = track.get('label')
 149                 subtitles.setdefault(self._CC_LANGS.get(track_label, track_label), []).append({
 150                     'url': self._proto_relative_url(track_url),
 151                     'ext': 'srt',
 152                 })
 153
 154         else:
 155             error = clean_html(response.get('access_err_msg'))
 156             if 'Video streaming is not available in your country' in error:
 157                 self.raise_geo_restricted(msg=error, countries=self._GEO_COUNTRIES)
 158             else:
 159                 raise ExtractorError(error)
 160
 161         # Likes
 162         emotion = self._download_json(
 163             'https://emocounter.hkedcity.net/handler.php',
 164             video_id,
 165             data=urlencode_postdata({
 166                 'action': 'get_emotion',
 167                 'data[bucket_id]': 'etv',
 168                 'data[identifier]': video_id,
 169             }),
 170             headers={'Content-Type': 'application/x-www-form-urlencoded'},
 171             fatal=False)
 172         like_count = int_or_none(try_get(emotion, lambda x: x['data']['emotion_data'][0]['count']))
 173
 174         return {
 175             'id': video_id,
 176             'title': title,
 177             'description': self._html_search_meta('description', webpage, fatal=False),
 178             'upload_date': unified_strdate(self._html_search_meta('ed_date', webpage, fatal=False), day_first=False),
 179             'duration': int_or_none(result.get('length')),
 180             'formats': formats,
 181             'subtitles': subtitles,
 182             'thumbnail': urljoin(_APPS_BASE_URL, result.get('image')),
 183             'view_count': str_to_int(result.get('view_count')),
 184             'like_count': like_count,
 185         }