[kakao] improve extraction

- support embed URLs
- support Kakao Legacy vid based embed URLs
- only extract fields used for extraction
- strip description and extract tags
This commit is contained in:
Remita Amine 2019-11-01 11:37:41 +01:00
parent e993f1a095
commit 274bf5e4c5

View file

@ -6,14 +6,15 @@ from .common import InfoExtractor
from ..compat import compat_str from ..compat import compat_str
from ..utils import ( from ..utils import (
int_or_none, int_or_none,
strip_or_none,
unified_timestamp, unified_timestamp,
update_url_query, update_url_query,
) )
class KakaoIE(InfoExtractor): class KakaoIE(InfoExtractor):
_VALID_URL = r'https?://tv\.kakao\.com/channel/(?P<channel>\d+)/cliplink/(?P<id>\d+)' _VALID_URL = r'https?://(?:play-)?tv\.kakao\.com/(?:channel/\d+|embed/player)/cliplink/(?P<id>\d+|[^?#&]+@my)'
_API_BASE = 'http://tv.kakao.com/api/v1/ft/cliplinks' _API_BASE_TMPL = 'http://tv.kakao.com/api/v1/ft/cliplinks/%s/'
_TESTS = [{ _TESTS = [{
'url': 'http://tv.kakao.com/channel/2671005/cliplink/301965083', 'url': 'http://tv.kakao.com/channel/2671005/cliplink/301965083',
@ -36,7 +37,7 @@ class KakaoIE(InfoExtractor):
'description': '러블리즈 - Destiny (나의 지구) (Lovelyz - Destiny)\r\n\r\n[쇼! 음악중심] 20160611, 507회', 'description': '러블리즈 - Destiny (나의 지구) (Lovelyz - Destiny)\r\n\r\n[쇼! 음악중심] 20160611, 507회',
'title': '러블리즈 - Destiny (나의 지구) (Lovelyz - Destiny)', 'title': '러블리즈 - Destiny (나의 지구) (Lovelyz - Destiny)',
'uploader_id': 2653210, 'uploader_id': 2653210,
'uploader': ' 음악중심', 'uploader': '! 음악중심',
'timestamp': 1485684628, 'timestamp': 1485684628,
'upload_date': '20170129', 'upload_date': '20170129',
} }
@ -44,6 +45,8 @@ class KakaoIE(InfoExtractor):
def _real_extract(self, url): def _real_extract(self, url):
video_id = self._match_id(url) video_id = self._match_id(url)
display_id = video_id.rstrip('@my')
api_base = self._API_BASE_TMPL % video_id
player_header = { player_header = {
'Referer': update_url_query( 'Referer': update_url_query(
@ -55,20 +58,22 @@ class KakaoIE(InfoExtractor):
}) })
} }
QUERY_COMMON = { query = {
'player': 'monet_html5', 'player': 'monet_html5',
'referer': url, 'referer': url,
'uuid': '', 'uuid': '',
'service': 'kakao_tv', 'service': 'kakao_tv',
'section': '', 'section': '',
'dteType': 'PC', 'dteType': 'PC',
'fields': ','.join([
'-*', 'tid', 'clipLink', 'displayTitle', 'clip', 'title',
'description', 'channelId', 'createTime', 'duration', 'playCount',
'likeCount', 'commentCount', 'tagList', 'channel', 'name',
'clipChapterThumbnailList', 'thumbnailUrl', 'timeInSec', 'isDefault'])
} }
query = QUERY_COMMON.copy()
query['fields'] = 'clipLink,clip,channel,hasPlusFriend,-service,-tagList'
impress = self._download_json( impress = self._download_json(
'%s/%s/impress' % (self._API_BASE, video_id), api_base + 'impress', display_id, 'Downloading video info',
video_id, 'Downloading video info',
query=query, headers=player_header) query=query, headers=player_header)
clip_link = impress['clipLink'] clip_link = impress['clipLink']
@ -78,30 +83,27 @@ class KakaoIE(InfoExtractor):
tid = impress.get('tid', '') tid = impress.get('tid', '')
query = QUERY_COMMON.copy()
query.update({ query.update({
'fields': '-*,outputList,profile,width,height,label,filesize',
'tid': tid, 'tid': tid,
'profile': 'HIGH', 'profile': 'HIGH',
}) })
raw = self._download_json( raw = self._download_json(
'%s/%s/raw' % (self._API_BASE, video_id), api_base + 'raw', display_id, 'Downloading video formats info',
video_id, 'Downloading video formats info',
query=query, headers=player_header) query=query, headers=player_header)
formats = [] formats = []
for fmt in raw.get('outputList', []): for fmt in raw.get('outputList', []):
try: try:
profile_name = fmt['profile'] profile_name = fmt['profile']
query.update({
'profile': profile_name,
'fields': '-*,url',
})
fmt_url_json = self._download_json( fmt_url_json = self._download_json(
'%s/%s/raw/videolocation' % (self._API_BASE, video_id), api_base + 'raw/videolocation', display_id,
video_id,
'Downloading video URL for profile %s' % profile_name, 'Downloading video URL for profile %s' % profile_name,
query={ query=query, headers=player_header, fatal=False)
'service': 'kakao_tv',
'section': '',
'tid': tid,
'profile': profile_name
}, headers=player_header, fatal=False)
if fmt_url_json is None: if fmt_url_json is None:
continue continue
@ -134,9 +136,9 @@ class KakaoIE(InfoExtractor):
}) })
return { return {
'id': video_id, 'id': display_id,
'title': title, 'title': title,
'description': clip.get('description'), 'description': strip_or_none(clip.get('description')),
'uploader': clip_link.get('channel', {}).get('name'), 'uploader': clip_link.get('channel', {}).get('name'),
'uploader_id': clip_link.get('channelId'), 'uploader_id': clip_link.get('channelId'),
'thumbnails': thumbs, 'thumbnails': thumbs,
@ -146,4 +148,5 @@ class KakaoIE(InfoExtractor):
'like_count': int_or_none(clip.get('likeCount')), 'like_count': int_or_none(clip.get('likeCount')),
'comment_count': int_or_none(clip.get('commentCount')), 'comment_count': int_or_none(clip.get('commentCount')),
'formats': formats, 'formats': formats,
'tags': clip.get('tagList'),
} }