yt-dlp/yt_dlp/extractor/tiktok.py

150 lines
6 KiB
Python
Raw Normal View History

2018-11-09 21:49:20 +00:00
# coding: utf-8
from __future__ import unicode_literals
2019-10-25 08:05:54 +00:00
from datetime import datetime
2018-11-09 21:49:20 +00:00
from .common import InfoExtractor
from ..utils import (
ExtractorError,
2018-11-09 21:49:20 +00:00
int_or_none,
str_or_none,
2019-10-25 08:05:54 +00:00
try_get
2018-11-09 21:49:20 +00:00
)
class TikTokBaseIE(InfoExtractor):
2020-10-26 01:15:46 +00:00
def _extract_aweme(self, props_data, webpage, url):
video_data = try_get(props_data, lambda x: x['pageProps'], expected_type=dict)
2020-02-15 06:39:53 +00:00
video_info = try_get(
video_data, lambda x: x['itemInfo']['itemStruct'], dict)
2020-02-15 06:39:53 +00:00
author_info = try_get(
video_data, lambda x: x['itemInfo']['itemStruct']['author'], dict) or {}
share_info = try_get(video_data, lambda x: x['itemInfo']['shareMeta'], dict) or {}
2019-10-25 08:05:54 +00:00
2020-02-15 06:39:53 +00:00
unique_id = str_or_none(author_info.get('uniqueId'))
timestamp = try_get(video_info, lambda x: int(x['createTime']), int)
date = datetime.fromtimestamp(timestamp).strftime('%Y%m%d')
2019-10-25 08:05:54 +00:00
height = try_get(video_info, lambda x: x['video']['height'], int)
width = try_get(video_info, lambda x: x['video']['width'], int)
2019-10-25 08:05:54 +00:00
thumbnails = []
thumbnails.append({
'url': video_info.get('thumbnail') or self._og_search_thumbnail(webpage),
2020-02-15 06:39:53 +00:00
'width': width,
'height': height
2019-10-25 08:05:54 +00:00
})
2020-10-26 01:15:46 +00:00
url = ''
if not url:
url = try_get(video_info, lambda x: x['video']['playAddr'])
if not url:
url = try_get(video_info, lambda x: x['video']['downloadAddr'])
2018-11-09 21:49:20 +00:00
formats = []
2019-10-25 08:05:54 +00:00
formats.append({
2020-10-26 01:15:46 +00:00
'url': url,
2019-10-25 08:05:54 +00:00
'ext': 'mp4',
2020-02-15 06:39:53 +00:00
'height': height,
'width': width
2019-10-25 08:05:54 +00:00
})
2018-11-09 21:49:20 +00:00
2020-10-26 01:15:46 +00:00
tracker = try_get(props_data, lambda x: x['initialProps']['$wid'])
2018-11-09 21:49:20 +00:00
return {
2020-02-15 06:39:53 +00:00
'comment_count': int_or_none(video_info.get('commentCount')),
'duration': try_get(video_info, lambda x: x['video']['videoMeta']['duration'], int),
'height': height,
'id': str_or_none(video_info.get('id')),
'like_count': int_or_none(video_info.get('diggCount')),
'repost_count': int_or_none(video_info.get('shareCount')),
'thumbnail': try_get(video_info, lambda x: x['covers'][0]),
2020-02-15 06:39:53 +00:00
'timestamp': timestamp,
'width': width,
'title': str_or_none(share_info.get('title')) or self._og_search_title(webpage),
2020-02-15 06:39:53 +00:00
'creator': str_or_none(author_info.get('nickName')),
'uploader': unique_id,
'uploader_id': str_or_none(author_info.get('userId')),
'uploader_url': 'https://www.tiktok.com/@' + unique_id,
2019-10-25 08:05:54 +00:00
'thumbnails': thumbnails,
'upload_date': date,
'webpage_url': self._og_search_url(webpage),
2020-02-15 06:39:53 +00:00
'description': str_or_none(video_info.get('text')) or str_or_none(share_info.get('desc')),
'ext': 'mp4',
'formats': formats,
'http_headers': {
'Referer': url,
2020-10-26 01:15:46 +00:00
'Cookie': 'tt_webid=%s; tt_webid_v2=%s' % (tracker, tracker),
}
2018-11-09 21:49:20 +00:00
}
class TikTokIE(TikTokBaseIE):
2020-02-15 07:00:12 +00:00
_VALID_URL = r'https?://www\.tiktok\.com/@[\w\._]+/video/(?P<id>\d+)'
2019-10-25 08:05:54 +00:00
_TESTS = [{
2019-10-25 08:05:54 +00:00
'url': 'https://www.tiktok.com/@leenabhushan/video/6748451240264420610',
'md5': '34a7543afd5a151b0840ba6736fb633b',
'info_dict': {
'comment_count': int,
2019-10-25 08:05:54 +00:00
'creator': 'facestoriesbyleenabh',
2020-02-15 06:39:53 +00:00
'description': 'md5:a9f6c0c44a1ff2249cae610372d0ae95',
2019-10-25 08:05:54 +00:00
'duration': 13,
2020-02-15 06:39:53 +00:00
'ext': 'mp4',
2019-10-25 08:05:54 +00:00
'formats': list,
'height': 1280,
2020-02-15 06:39:53 +00:00
'id': '6748451240264420610',
'like_count': int,
'repost_count': int,
'thumbnail': r're:^https?://[\w\/\.\-]+(~[\w\-]+\.image)?',
2019-10-25 08:05:54 +00:00
'thumbnails': list,
2020-02-15 06:39:53 +00:00
'timestamp': 1571246252,
'title': 'facestoriesbyleenabh on TikTok',
'upload_date': '20191016',
'uploader': 'leenabhushan',
2019-10-25 08:05:54 +00:00
'uploader_id': '6691488002098119685',
'uploader_url': r're:https://www.tiktok.com/@leenabhushan',
'webpage_url': r're:https://www.tiktok.com/@leenabhushan/(video/)?6748451240264420610',
'width': 720,
}
}, {
2019-10-25 08:05:54 +00:00
'url': 'https://www.tiktok.com/@patroxofficial/video/6742501081818877190?langCountry=en',
'md5': '06b9800d47d5fe51a19e322dd86e61c9',
'info_dict': {
'comment_count': int,
'creator': 'patroX',
'description': 'md5:5e2a23877420bb85ce6521dbee39ba94',
'duration': 27,
'ext': 'mp4',
'formats': list,
'height': 960,
'id': '6742501081818877190',
'like_count': int,
'repost_count': int,
'thumbnail': r're:^https?://[\w\/\.\-]+(~[\w\-]+\.image)?',
'thumbnails': list,
'timestamp': 1569860870,
'title': 'patroX on TikTok',
'upload_date': '20190930',
2020-02-15 06:39:53 +00:00
'uploader': 'patroxofficial',
2019-10-25 08:05:54 +00:00
'uploader_id': '18702747',
'uploader_url': r're:https://www.tiktok.com/@patroxofficial',
'webpage_url': r're:https://www.tiktok.com/@patroxofficial/(video/)?6742501081818877190',
'width': 540,
}
}]
def _real_extract(self, url):
video_id = self._match_id(url)
2020-10-26 12:38:25 +00:00
# If we only call once, we get a 403 when downlaoding the video.
self._download_webpage(url, video_id)
2020-02-15 06:39:53 +00:00
webpage = self._download_webpage(url, video_id, note='Downloading video webpage')
2019-10-25 08:05:54 +00:00
json_string = self._search_regex(
2020-02-15 07:00:12 +00:00
r'id=\"__NEXT_DATA__\"\s+type=\"application\/json\"\s*[^>]+>\s*(?P<json_string_ld>[^<]+)',
webpage, 'json_string', group='json_string_ld')
2019-10-25 08:05:54 +00:00
json_data = self._parse_json(json_string, video_id)
2020-10-26 01:15:46 +00:00
props_data = try_get(json_data, lambda x: x['props'], expected_type=dict)
2019-10-25 08:05:54 +00:00
# Chech statusCode for success
2020-10-26 01:15:46 +00:00
if props_data.get('pageProps').get('statusCode') == 0:
return self._extract_aweme(props_data, webpage, url)
2019-10-25 08:05:54 +00:00
2020-02-15 06:39:53 +00:00
raise ExtractorError('Video not available', video_id=video_id)