]> Raphaƫl G. Git Repositories - youtubedl/blob - youtube_dl/extractor/tiktok.py
Fix extraction from youtube.
[youtubedl] / youtube_dl / extractor / tiktok.py
1 # coding: utf-8
2 from __future__ import unicode_literals
3
4 from .common import InfoExtractor
5 from ..utils import (
6 compat_str,
7 ExtractorError,
8 int_or_none,
9 str_or_none,
10 try_get,
11 url_or_none,
12 )
13
14
15 class TikTokBaseIE(InfoExtractor):
16 def _extract_aweme(self, data):
17 video = data['video']
18 description = str_or_none(try_get(data, lambda x: x['desc']))
19 width = int_or_none(try_get(data, lambda x: video['width']))
20 height = int_or_none(try_get(data, lambda x: video['height']))
21
22 format_urls = set()
23 formats = []
24 for format_id in (
25 'play_addr_lowbr', 'play_addr', 'play_addr_h264',
26 'download_addr'):
27 for format in try_get(
28 video, lambda x: x[format_id]['url_list'], list) or []:
29 format_url = url_or_none(format)
30 if not format_url:
31 continue
32 if format_url in format_urls:
33 continue
34 format_urls.add(format_url)
35 formats.append({
36 'url': format_url,
37 'ext': 'mp4',
38 'height': height,
39 'width': width,
40 })
41 self._sort_formats(formats)
42
43 thumbnail = url_or_none(try_get(
44 video, lambda x: x['cover']['url_list'][0], compat_str))
45 uploader = try_get(data, lambda x: x['author']['nickname'], compat_str)
46 timestamp = int_or_none(data.get('create_time'))
47 comment_count = int_or_none(data.get('comment_count')) or int_or_none(
48 try_get(data, lambda x: x['statistics']['comment_count']))
49 repost_count = int_or_none(try_get(
50 data, lambda x: x['statistics']['share_count']))
51
52 aweme_id = data['aweme_id']
53
54 return {
55 'id': aweme_id,
56 'title': uploader or aweme_id,
57 'description': description,
58 'thumbnail': thumbnail,
59 'uploader': uploader,
60 'timestamp': timestamp,
61 'comment_count': comment_count,
62 'repost_count': repost_count,
63 'formats': formats,
64 }
65
66
67 class TikTokIE(TikTokBaseIE):
68 _VALID_URL = r'https?://(?:m\.)?tiktok\.com/v/(?P<id>\d+)'
69 _TEST = {
70 'url': 'https://m.tiktok.com/v/6606727368545406213.html',
71 'md5': 'd584b572e92fcd48888051f238022420',
72 'info_dict': {
73 'id': '6606727368545406213',
74 'ext': 'mp4',
75 'title': 'Zureeal',
76 'description': '#bowsette#mario#cosplay#uk#lgbt#gaming#asian#bowsettecosplay',
77 'thumbnail': r're:^https?://.*~noop.image',
78 'uploader': 'Zureeal',
79 'timestamp': 1538248586,
80 'upload_date': '20180929',
81 'comment_count': int,
82 'repost_count': int,
83 }
84 }
85
86 def _real_extract(self, url):
87 video_id = self._match_id(url)
88 webpage = self._download_webpage(url, video_id)
89 data = self._parse_json(self._search_regex(
90 r'\bdata\s*=\s*({.+?})\s*;', webpage, 'data'), video_id)
91 return self._extract_aweme(data)
92
93
94 class TikTokUserIE(TikTokBaseIE):
95 _VALID_URL = r'https?://(?:m\.)?tiktok\.com/h5/share/usr/(?P<id>\d+)'
96 _TEST = {
97 'url': 'https://m.tiktok.com/h5/share/usr/188294915489964032.html',
98 'info_dict': {
99 'id': '188294915489964032',
100 },
101 'playlist_mincount': 24,
102 }
103
104 def _real_extract(self, url):
105 user_id = self._match_id(url)
106 data = self._download_json(
107 'https://m.tiktok.com/h5/share/usr/list/%s/' % user_id, user_id,
108 query={'_signature': '_'})
109 entries = []
110 for aweme in data['aweme_list']:
111 try:
112 entry = self._extract_aweme(aweme)
113 except ExtractorError:
114 continue
115 entry['extractor_key'] = TikTokIE.ie_key()
116 entries.append(entry)
117 return self.playlist_result(entries, user_id)