[youtube] Fix extraction.
[youtube-dl] / youtube_dl / extractor / camdemy.py
1 # coding: utf-8
2 from __future__ import unicode_literals
3
4 import re
5
6 from .common import InfoExtractor
7 from ..compat import (
8     compat_urllib_parse_urlencode,
9     compat_urlparse,
10 )
11 from ..utils import (
12     clean_html,
13     parse_duration,
14     str_to_int,
15     unified_strdate,
16 )
17
18
19 class CamdemyIE(InfoExtractor):
20     _VALID_URL = r'https?://(?:www\.)?camdemy\.com/media/(?P<id>\d+)'
21     _TESTS = [{
22         # single file
23         'url': 'http://www.camdemy.com/media/5181/',
24         'md5': '5a5562b6a98b37873119102e052e311b',
25         'info_dict': {
26             'id': '5181',
27             'ext': 'mp4',
28             'title': 'Ch1-1 Introduction, Signals (02-23-2012)',
29             'thumbnail': r're:^https?://.*\.jpg$',
30             'creator': 'ss11spring',
31             'duration': 1591,
32             'upload_date': '20130114',
33             'view_count': int,
34         }
35     }, {
36         # With non-empty description
37         # webpage returns "No permission or not login"
38         'url': 'http://www.camdemy.com/media/13885',
39         'md5': '4576a3bb2581f86c61044822adbd1249',
40         'info_dict': {
41             'id': '13885',
42             'ext': 'mp4',
43             'title': 'EverCam + Camdemy QuickStart',
44             'thumbnail': r're:^https?://.*\.jpg$',
45             'description': 'md5:2a9f989c2b153a2342acee579c6e7db6',
46             'creator': 'evercam',
47             'duration': 318,
48         }
49     }, {
50         # External source (YouTube)
51         'url': 'http://www.camdemy.com/media/14842',
52         'info_dict': {
53             'id': '2vsYQzNIsJo',
54             'ext': 'mp4',
55             'title': 'Excel 2013 Tutorial - How to add Password Protection',
56             'description': 'Excel 2013 Tutorial for Beginners - How to add Password Protection',
57             'upload_date': '20130211',
58             'uploader': 'Hun Kim',
59             'uploader_id': 'hunkimtutorials',
60         },
61         'params': {
62             'skip_download': True,
63         },
64     }]
65
66     def _real_extract(self, url):
67         video_id = self._match_id(url)
68
69         webpage = self._download_webpage(url, video_id)
70
71         src_from = self._html_search_regex(
72             r"class=['\"]srcFrom['\"][^>]*>Sources?(?:\s+from)?\s*:\s*<a[^>]+(?:href|title)=(['\"])(?P<url>(?:(?!\1).)+)\1",
73             webpage, 'external source', default=None, group='url')
74         if src_from:
75             return self.url_result(src_from)
76
77         oembed_obj = self._download_json(
78             'http://www.camdemy.com/oembed/?format=json&url=' + url, video_id)
79
80         title = oembed_obj['title']
81         thumb_url = oembed_obj['thumbnail_url']
82         video_folder = compat_urlparse.urljoin(thumb_url, 'video/')
83         file_list_doc = self._download_xml(
84             compat_urlparse.urljoin(video_folder, 'fileList.xml'),
85             video_id, 'Downloading filelist XML')
86         file_name = file_list_doc.find('./video/item/fileName').text
87         video_url = compat_urlparse.urljoin(video_folder, file_name)
88
89         # Some URLs return "No permission or not login" in a webpage despite being
90         # freely available via oembed JSON URL (e.g. http://www.camdemy.com/media/13885)
91         upload_date = unified_strdate(self._search_regex(
92             r'>published on ([^<]+)<', webpage,
93             'upload date', default=None))
94         view_count = str_to_int(self._search_regex(
95             r'role=["\']viewCnt["\'][^>]*>([\d,.]+) views',
96             webpage, 'view count', default=None))
97         description = self._html_search_meta(
98             'description', webpage, default=None) or clean_html(
99             oembed_obj.get('description'))
100
101         return {
102             'id': video_id,
103             'url': video_url,
104             'title': title,
105             'thumbnail': thumb_url,
106             'description': description,
107             'creator': oembed_obj.get('author_name'),
108             'duration': parse_duration(oembed_obj.get('duration')),
109             'upload_date': upload_date,
110             'view_count': view_count,
111         }
112
113
114 class CamdemyFolderIE(InfoExtractor):
115     _VALID_URL = r'https?://(?:www\.)?camdemy\.com/folder/(?P<id>\d+)'
116     _TESTS = [{
117         # links with trailing slash
118         'url': 'http://www.camdemy.com/folder/450',
119         'info_dict': {
120             'id': '450',
121             'title': '信號與系統 2012 & 2011 (Signals and Systems)',
122         },
123         'playlist_mincount': 145
124     }, {
125         # links without trailing slash
126         # and multi-page
127         'url': 'http://www.camdemy.com/folder/853',
128         'info_dict': {
129             'id': '853',
130             'title': '科學計算 - 使用 Matlab'
131         },
132         'playlist_mincount': 20
133     }, {
134         # with displayMode parameter. For testing the codes to add parameters
135         'url': 'http://www.camdemy.com/folder/853/?displayMode=defaultOrderByOrg',
136         'info_dict': {
137             'id': '853',
138             'title': '科學計算 - 使用 Matlab'
139         },
140         'playlist_mincount': 20
141     }]
142
143     def _real_extract(self, url):
144         folder_id = self._match_id(url)
145
146         # Add displayMode=list so that all links are displayed in a single page
147         parsed_url = list(compat_urlparse.urlparse(url))
148         query = dict(compat_urlparse.parse_qsl(parsed_url[4]))
149         query.update({'displayMode': 'list'})
150         parsed_url[4] = compat_urllib_parse_urlencode(query)
151         final_url = compat_urlparse.urlunparse(parsed_url)
152
153         page = self._download_webpage(final_url, folder_id)
154         matches = re.findall(r"href='(/media/\d+/?)'", page)
155
156         entries = [self.url_result('http://www.camdemy.com' + media_path)
157                    for media_path in matches]
158
159         folder_title = self._html_search_meta('keywords', page)
160
161         return self.playlist_result(entries, folder_id, folder_title)