1 # coding: utf-8
2 from __future__ import unicode_literals
3
4 import re
5
6 from .common import InfoExtractor
7 from ..utils import (
8 clean_html,
9 float_or_none,
10 get_element_by_class,
11 get_element_by_id,
12 parse_duration,
13 str_to_int,
14 unified_timestamp,
15 urlencode_postdata,
16 )
17
18
19 class TwitCastingIE(InfoExtractor):
20 _VALID_URL = r'https?://(?:[^/]+\.)?twitcasting\.tv/(?P<uploader_id>[^/]+)/movie/(?P<id>\d+)'
21 _TESTS = [{
22 'url': 'https://twitcasting.tv/ivetesangalo/movie/2357609',
23 'md5': '745243cad58c4681dc752490f7540d7f',
24 'info_dict': {
25 'id': '2357609',
26 'ext': 'mp4',
27 'title': 'Live #2357609',
28 'uploader_id': 'ivetesangalo',
29 'description': 'Twitter Oficial da cantora brasileira Ivete Sangalo.',
30 'thumbnail': r're:^https?://.*\.jpg$',
31 'upload_date': '20110822',
32 'timestamp': 1314010824,
33 'duration': 32,
34 'view_count': int,
35 },
36 'params': {
37 'skip_download': True,
38 },
39 }, {
40 'url': 'https://twitcasting.tv/mttbernardini/movie/3689740',
41 'info_dict': {
42 'id': '3689740',
43 'ext': 'mp4',
44 'title': 'Live playing something #3689740',
45 'uploader_id': 'mttbernardini',
46 'description': 'Salve, io sono Matto (ma con la e). Questa è la mia presentazione, in quanto sono letteralmente matto (nel senso di strano), con qualcosa in più.',
47 'thumbnail': r're:^https?://.*\.jpg$',
48 'upload_date': '20120212',
49 'timestamp': 1329028024,
50 'duration': 681,
51 'view_count': int,
52 },
53 'params': {
54 'skip_download': True,
55 'videopassword': 'abc',
56 },
57 }]
58
59 def _real_extract(self, url):
60 uploader_id, video_id = re.match(self._VALID_URL, url).groups()
61
62 video_password = self._downloader.params.get('videopassword')
63 request_data = None
64 if video_password:
65 request_data = urlencode_postdata({
66 'password': video_password,
67 })
68 webpage = self._download_webpage(url, video_id, data=request_data)
69
70 title = clean_html(get_element_by_id(
71 'movietitle', webpage)) or self._html_search_meta(
72 ['og:title', 'twitter:title'], webpage, fatal=True)
73
74 video_js_data = {}
75 m3u8_url = self._search_regex(
76 r'data-movie-url=(["\'])(?P<url>(?:(?!\1).)+)\1',
77 webpage, 'm3u8 url', group='url', default=None)
78 if not m3u8_url:
79 video_js_data = self._parse_json(self._search_regex(
80 r"data-movie-playlist='(\[[^']+\])'",
81 webpage, 'movie playlist'), video_id)[0]
82 m3u8_url = video_js_data['source']['url']
83
84 # use `m3u8` entry_protocol until EXT-X-MAP is properly supported by `m3u8_native` entry_protocol
85 formats = self._extract_m3u8_formats(
86 m3u8_url, video_id, 'mp4', m3u8_id='hls')
87
88 thumbnail = video_js_data.get('thumbnailUrl') or self._og_search_thumbnail(webpage)
89 description = clean_html(get_element_by_id(
90 'authorcomment', webpage)) or self._html_search_meta(
91 ['description', 'og:description', 'twitter:description'], webpage)
92 duration = float_or_none(video_js_data.get(
93 'duration'), 1000) or parse_duration(clean_html(
94 get_element_by_class('tw-player-duration-time', webpage)))
95 view_count = str_to_int(self._search_regex(
96 r'Total\s*:\s*([\d,]+)\s*Views', webpage, 'views', None))
97 timestamp = unified_timestamp(self._search_regex(
98 r'data-toggle="true"[^>]+datetime="([^"]+)"',
99 webpage, 'datetime', None))
100
101 return {
102 'id': video_id,
103 'title': title,
104 'description': description,
105 'thumbnail': thumbnail,
106 'timestamp': timestamp,
107 'uploader_id': uploader_id,
108 'duration': duration,
109 'view_count': view_count,
110 'formats': formats,
111 }
|
