aboutsummaryrefslogtreecommitdiff
path: root/youtube_dl/extractor/xuite.py
blob: 0276c0dbb26f584c0fc49cf35614f399cc203508 (plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
# coding: utf-8
from __future__ import unicode_literals

from .common import InfoExtractor
from ..utils import (
    ExtractorError,
    float_or_none,
    get_element_by_attribute,
    parse_iso8601,
    remove_end,
)


class XuiteIE(InfoExtractor):
    IE_DESC = '隨意窩Xuite影音'
    _REGEX_BASE64 = r'(?:[A-Za-z0-9+/]{4})*(?:[A-Za-z0-9+/]{2}==|[A-Za-z0-9+/]{3}=)?'
    _VALID_URL = r'https?://vlog\.xuite\.net/(?:play|embed)/(?P<id>%s)' % _REGEX_BASE64
    _TESTS = [{
        # Audio
        'url': 'http://vlog.xuite.net/play/RGkzc1ZULTM4NjA5MTQuZmx2',
        'md5': 'e79284c87b371424885448d11f6398c8',
        'info_dict': {
            'id': '3860914',
            'ext': 'mp3',
            'title': '孤單南半球-歐德陽',
            'description': '孤單南半球-歐德陽',
            'thumbnail': r're:^https?://.*\.jpg$',
            'duration': 247.246,
            'timestamp': 1314932940,
            'upload_date': '20110902',
            'uploader': '阿能',
            'uploader_id': '15973816',
            'categories': ['個人短片'],
        },
    }, {
        # Video with only one format
        'url': 'http://vlog.xuite.net/play/WUxxR2xCLTI1OTI1MDk5LmZsdg==',
        'md5': '21f7b39c009b5a4615b4463df6eb7a46',
        'info_dict': {
            'id': '25925099',
            'ext': 'mp4',
            'title': 'BigBuckBunny_320x180',
            'thumbnail': r're:^https?://.*\.jpg$',
            'duration': 596.458,
            'timestamp': 1454242500,
            'upload_date': '20160131',
            'uploader': '屁姥',
            'uploader_id': '12158353',
            'categories': ['個人短片'],
            'description': 'http://download.blender.org/peach/bigbuckbunny_movies/BigBuckBunny_320x180.mp4',
        },
    }, {
        # Video with two formats
        'url': 'http://vlog.xuite.net/play/bWo1N1pLLTIxMzAxMTcwLmZsdg==',
        'md5': '1166e0f461efe55b62e26a2d2a68e6de',
        'info_dict': {
            'id': '21301170',
            'ext': 'mp4',
            'title': '暗殺教室 02',
            'description': '字幕:【極影字幕社】',
            'thumbnail': r're:^https?://.*\.jpg$',
            'duration': 1384.907,
            'timestamp': 1421481240,
            'upload_date': '20150117',
            'uploader': '我只是想認真點',
            'uploader_id': '242127761',
            'categories': ['電玩動漫'],
        },
        'skip': 'Video removed',
    }, {
        # Video with encoded media id
        # from http://forgetfulbc.blogspot.com/2016/06/date.html
        'url': 'http://vlog.xuite.net/embed/cE1xbENoLTI3NDQ3MzM2LmZsdg==?ar=0&as=0',
        'info_dict': {
            'id': '27447336',
            'ext': 'mp4',
            'title': '男女平權只是口號?專家解釋約會時男生是否該幫女生付錢 (中字)',
            'description': 'md5:1223810fa123b179083a3aed53574706',
            'timestamp': 1466160960,
            'upload_date': '20160617',
            'uploader': 'B.C. & Lowy',
            'uploader_id': '232279340',
        },
    }, {
        'url': 'http://vlog.xuite.net/play/S1dDUjdyLTMyOTc3NjcuZmx2/%E5%AD%AB%E7%87%95%E5%A7%BF-%E7%9C%BC%E6%B7%9A%E6%88%90%E8%A9%A9',
        'only_matching': True,
    }]

    def _real_extract(self, url):
        # /play/ URLs provide embedded video URL and more metadata
        url = url.replace('/embed/', '/play/')
        video_id = self._match_id(url)

        webpage = self._download_webpage(url, video_id)

        error_msg = self._search_regex(
            r'<div id="error-message-content">([^<]+)',
            webpage, 'error message', default=None)
        if error_msg:
            raise ExtractorError(
                '%s returned error: %s' % (self.IE_NAME, error_msg),
                expected=True)

        media_info = self._parse_json(self._search_regex(
            r'var\s+mediaInfo\s*=\s*({.*});', webpage, 'media info'), video_id)

        video_id = media_info['MEDIA_ID']

        formats = []
        for key in ('html5Url', 'html5HQUrl'):
            video_url = media_info.get(key)
            if not video_url:
                continue
            format_id = self._search_regex(
                r'\bq=(.+?)\b', video_url, 'format id', default=None)
            formats.append({
                'url': video_url,
                'ext': 'mp4' if format_id.isnumeric() else format_id,
                'format_id': format_id,
                'height': int(format_id) if format_id.isnumeric() else None,
            })
        self._sort_formats(formats)

        timestamp = media_info.get('PUBLISH_DATETIME')
        if timestamp:
            timestamp = parse_iso8601(timestamp + ' +0800', ' ')

        category = media_info.get('catName')
        categories = [category] if category else []

        uploader = media_info.get('NICKNAME')
        uploader_url = None

        author_div = get_element_by_attribute('itemprop', 'author', webpage)
        if author_div:
            uploader = uploader or self._html_search_meta('name', author_div)
            uploader_url = self._html_search_regex(
                r'<link[^>]+itemprop="url"[^>]+href="([^"]+)"', author_div,
                'uploader URL', fatal=False)

        return {
            'id': video_id,
            'title': media_info['TITLE'],
            'description': remove_end(media_info.get('metaDesc'), ' (Xuite 影音)'),
            'thumbnail': media_info.get('ogImageUrl'),
            'timestamp': timestamp,
            'uploader': uploader,
            'uploader_id': media_info.get('MEMBER_ID'),
            'uploader_url': uploader_url,
            'duration': float_or_none(media_info.get('MEDIA_DURATION'), 1000000),
            'categories': categories,
            'formats': formats,
        }