etc/yt-dlp
0
0
Fork 0
mirror of https://github.com/yt-dlp/yt-dlp.git synced 2024-11-18 01:35:12 +00:00
Code Issues Projects Releases Wiki Activity

[tiktok] fix regex

Browse source
This commit is contained in:
Aakash Gajjar 2020-02-15 12:30:12 +05:30
parent 6fb11ca851
commit 6255e567d9
1 changed files with 3 additions and 12 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

15
youtube_dl/extractor/tiktok.py
View file

@ -65,16 +65,7 @@ def _extract_aweme(self, video_data, webpage):
class TikTokIE(TikTokBaseIE): class TikTokIE(TikTokBaseIE):
_VALID_URL = r'''(?x) _VALID_URL = r'https?://www\.tiktok\.com/@[\w\._]+/video/(?P<id>\d+)'
https?://
(?:
(?:www|vm)\.
(?:tiktok.com)\/
(@(?P<username>[\w\.]+))\/
(?:video)\/
)
(?P<id>[\d]{6,})
'''
_TESTS = [{ _TESTS = [{
'url': 'https://www.tiktok.com/@leenabhushan/video/6748451240264420610', 'url': 'https://www.tiktok.com/@leenabhushan/video/6748451240264420610',
@ -133,8 +124,8 @@ def _real_extract(self, url):
webpage = self._download_webpage(url, video_id, note='Downloading video webpage') webpage = self._download_webpage(url, video_id, note='Downloading video webpage')
json_string = self._search_regex( json_string = self._search_regex(
r'id=\"__NEXT_DATA__\"\s+type=\"application\/json\"\s*[^>]+>\s*(?P<json_string>[^<]+)', r'id=\"__NEXT_DATA__\"\s+type=\"application\/json\"\s*[^>]+>\s*(?P<json_string_ld>[^<]+)',
webpage, 'json_string', group='json_string') webpage, 'json_string', group='json_string_ld')
json_data = self._parse_json(json_string, video_id) json_data = self._parse_json(json_string, video_id)
video_data = try_get(json_data, lambda x: x['props']['pageProps'], expected_type=dict) video_data = try_get(json_data, lambda x: x['props']['pageProps'], expected_type=dict)