aboutsummaryrefslogtreecommitdiffstats
path: root/hypervideo_dl/extractor/tver.py
diff options
context:
space:
mode:
Diffstat (limited to 'hypervideo_dl/extractor/tver.py')
-rw-r--r--hypervideo_dl/extractor/tver.py130
1 files changed, 79 insertions, 51 deletions
diff --git a/hypervideo_dl/extractor/tver.py b/hypervideo_dl/extractor/tver.py
index 9ff3136..cebd027 100644
--- a/hypervideo_dl/extractor/tver.py
+++ b/hypervideo_dl/extractor/tver.py
@@ -1,77 +1,105 @@
-# coding: utf-8
-from __future__ import unicode_literals
-
-
from .common import InfoExtractor
-from ..compat import compat_str
from ..utils import (
ExtractorError,
- int_or_none,
- remove_start,
+ join_nonempty,
smuggle_url,
+ str_or_none,
+ strip_or_none,
traverse_obj,
)
class TVerIE(InfoExtractor):
- _VALID_URL = r'https?://(?:www\.)?tver\.jp/(?P<path>corner|episode|feature|lp|tokyo2020/video)/(?P<id>[fc]?\d+)'
- # videos are only available for 7 days
+ _VALID_URL = r'https?://(?:www\.)?tver\.jp/(?:(?P<type>lp|corner|series|episodes?|feature|tokyo2020/video)/)+(?P<id>[a-zA-Z0-9]+)'
_TESTS = [{
- 'url': 'https://tver.jp/corner/f0062178',
- 'only_matching': True,
- }, {
- 'url': 'https://tver.jp/feature/f0062413',
- 'only_matching': True,
- }, {
- 'url': 'https://tver.jp/episode/79622438',
- 'only_matching': True,
+ 'skip': 'videos are only available for 7 days',
+ 'url': 'https://tver.jp/episodes/ep83nf3w4p',
+ 'info_dict': {
+ 'title': '家事ヤロウ!!! 売り場席巻のチーズSP&財前直見×森泉親子の脱東京暮らし密着!',
+ 'description': 'md5:dc2c06b6acc23f1e7c730c513737719b',
+ 'series': '家事ヤロウ!!!',
+ 'episode': '売り場席巻のチーズSP&財前直見×森泉親子の脱東京暮らし密着!',
+ 'alt_title': '売り場席巻のチーズSP&財前直見×森泉親子の脱東京暮らし密着!',
+ 'channel': 'テレビ朝日',
+ 'onair_label': '5月3日(火)放送分',
+ 'ext_title': '家事ヤロウ!!! 売り場席巻のチーズSP&財前直見×森泉親子の脱東京暮らし密着! テレビ朝日 5月3日(火)放送分',
+ },
+ 'add_ie': ['BrightcoveNew'],
}, {
- # subtitle = ' '
- 'url': 'https://tver.jp/corner/f0068870',
+ 'url': 'https://tver.jp/corner/f0103888',
'only_matching': True,
}, {
- 'url': 'https://tver.jp/lp/f0009694',
- 'only_matching': True,
- }, {
- 'url': 'https://tver.jp/lp/c0000239',
- 'only_matching': True,
- }, {
- 'url': 'https://tver.jp/tokyo2020/video/6264525510001',
+ 'url': 'https://tver.jp/lp/f0033031',
'only_matching': True,
}]
- _TOKEN = None
BRIGHTCOVE_URL_TEMPLATE = 'http://players.brightcove.net/%s/default_default/index.html?videoId=%s'
+ _PLATFORM_UID = None
+ _PLATFORM_TOKEN = None
def _real_initialize(self):
- self._TOKEN = self._download_json(
- 'https://tver.jp/api/access_token.php', None)['token']
+ create_response = self._download_json(
+ 'https://platform-api.tver.jp/v2/api/platform_users/browser/create', None,
+ note='Creating session', data=b'device_type=pc', headers={
+ 'Origin': 'https://s.tver.jp',
+ 'Referer': 'https://s.tver.jp/',
+ 'Content-Type': 'application/x-www-form-urlencoded',
+ })
+ self._PLATFORM_UID = traverse_obj(create_response, ('result', 'platform_uid'))
+ self._PLATFORM_TOKEN = traverse_obj(create_response, ('result', 'platform_token'))
def _real_extract(self, url):
- path, video_id = self._match_valid_url(url).groups()
- if path == 'lp':
- webpage = self._download_webpage(url, video_id)
- redirect_path = self._search_regex(r'to_href="([^"]+)', webpage, 'redirect path')
- path, video_id = self._match_valid_url(f'https://tver.jp{redirect_path}').groups()
- api_response = self._download_json(f'https://api.tver.jp/v4/{path}/{video_id}', video_id, query={'token': self._TOKEN})
- p_id = traverse_obj(api_response, ('main', 'publisher_id'))
- if not p_id:
- error_msg, expected = traverse_obj(api_response, ('episode', 0, 'textbar', 0, ('text', 'longer')), get_all=False), True
- if not error_msg:
- error_msg, expected = 'Failed to extract publisher ID', False
- raise ExtractorError(error_msg, expected=expected)
- service = remove_start(traverse_obj(api_response, ('main', 'service')), 'ts_')
+ video_id, video_type = self._match_valid_url(url).group('id', 'type')
+ if video_type not in {'series', 'episodes'}:
+ webpage = self._download_webpage(url, video_id, note='Resolving to new URL')
+ video_id = self._match_id(self._search_regex(
+ (r'canonical"\s*href="(https?://tver\.jp/[^"]+)"', r'&link=(https?://tver\.jp/[^?&]+)[?&]'),
+ webpage, 'url regex'))
+
+ episode_info = self._download_json(
+ f'https://platform-api.tver.jp/service/api/v1/callEpisode/{video_id}?require_data=mylist,later[epefy106ur],good[epefy106ur],resume[epefy106ur]',
+ video_id, fatal=False,
+ query={
+ 'platform_uid': self._PLATFORM_UID,
+ 'platform_token': self._PLATFORM_TOKEN,
+ }, headers={
+ 'x-tver-platform-type': 'web'
+ })
+ episode_content = traverse_obj(
+ episode_info, ('result', 'episode', 'content')) or {}
+
+ video_info = self._download_json(
+ f'https://statics.tver.jp/content/episode/{video_id}.json', video_id,
+ query={
+ 'v': str_or_none(episode_content.get('version')) or '5',
+ }, headers={
+ 'Origin': 'https://tver.jp',
+ 'Referer': 'https://tver.jp/',
+ })
+ p_id = video_info['video']['accountID']
+ r_id = traverse_obj(video_info, ('video', ('videoRefID', 'videoID')), get_all=False)
+ if not r_id:
+ raise ExtractorError('Failed to extract reference ID for Brightcove')
+ if not r_id.isdigit():
+ r_id = f'ref:{r_id}'
- r_id = traverse_obj(api_response, ('main', 'reference_id'))
- if service not in ('tx', 'russia2018', 'sebare2018live', 'gorin'):
- r_id = 'ref:' + r_id
- bc_url = smuggle_url(
- self.BRIGHTCOVE_URL_TEMPLATE % (p_id, r_id),
- {'geo_countries': ['JP']})
+ episode = strip_or_none(episode_content.get('title'))
+ series = str_or_none(episode_content.get('seriesTitle'))
+ title = (
+ join_nonempty(series, episode, delim=' ')
+ or str_or_none(video_info.get('title')))
+ provider = str_or_none(episode_content.get('productionProviderName'))
+ onair_label = str_or_none(episode_content.get('broadcastDateLabel'))
return {
'_type': 'url_transparent',
- 'description': traverse_obj(api_response, ('main', 'note', 0, 'text'), expected_type=compat_str),
- 'episode_number': int_or_none(traverse_obj(api_response, ('main', 'ext', 'episode_number'), expected_type=compat_str)),
- 'url': bc_url,
+ 'title': title,
+ 'series': series,
+ 'episode': episode,
+ # an another title which is considered "full title" for some viewers
+ 'alt_title': join_nonempty(title, provider, onair_label, delim=' '),
+ 'channel': provider,
+ 'description': str_or_none(video_info.get('description')),
+ 'url': smuggle_url(
+ self.BRIGHTCOVE_URL_TEMPLATE % (p_id, r_id), {'geo_countries': ['JP']}),
'ie_key': 'BrightcoveNew',
}