]>
Commit | Line | Data |
---|---|---|
1 | # coding: utf-8 | |
2 | from __future__ import unicode_literals | |
3 | ||
4 | import itertools | |
5 | import json | |
6 | import re | |
7 | ||
8 | from .common import InfoExtractor, SearchInfoExtractor | |
9 | from ..compat import ( | |
10 | compat_urllib_parse, | |
11 | compat_urlparse, | |
12 | ) | |
13 | from ..utils import ( | |
14 | clean_html, | |
15 | unescapeHTML, | |
16 | ExtractorError, | |
17 | int_or_none, | |
18 | ) | |
19 | ||
20 | ||
21 | class YahooIE(InfoExtractor): | |
22 | IE_DESC = 'Yahoo screen and movies' | |
23 | _VALID_URL = r'(?P<url>(?P<host>https?://(?:[a-zA-Z]{2}\.)?[\da-zA-Z_-]+\.yahoo\.com)/(?:[^/]+/)*(?P<display_id>.+?)-(?P<id>[0-9]+)(?:-[a-z]+)?\.html)' | |
24 | _TESTS = [ | |
25 | { | |
26 | 'url': 'http://screen.yahoo.com/julian-smith-travis-legg-watch-214727115.html', | |
27 | 'info_dict': { | |
28 | 'id': '2d25e626-2378-391f-ada0-ddaf1417e588', | |
29 | 'ext': 'mp4', | |
30 | 'title': 'Julian Smith & Travis Legg Watch Julian Smith', | |
31 | 'description': 'Julian and Travis watch Julian Smith', | |
32 | 'duration': 6863, | |
33 | }, | |
34 | }, | |
35 | { | |
36 | 'url': 'http://screen.yahoo.com/wired/codefellas-s1-ep12-cougar-lies-103000935.html', | |
37 | 'md5': 'd6e6fc6e1313c608f316ddad7b82b306', | |
38 | 'info_dict': { | |
39 | 'id': 'd1dedf8c-d58c-38c3-8963-e899929ae0a9', | |
40 | 'ext': 'mp4', | |
41 | 'title': 'Codefellas - The Cougar Lies with Spanish Moss', | |
42 | 'description': 'md5:66b627ab0a282b26352136ca96ce73c1', | |
43 | 'duration': 151, | |
44 | }, | |
45 | }, | |
46 | { | |
47 | 'url': 'https://screen.yahoo.com/community/community-sizzle-reel-203225340.html?format=embed', | |
48 | 'md5': '60e8ac193d8fb71997caa8fce54c6460', | |
49 | 'info_dict': { | |
50 | 'id': '4fe78544-8d48-39d8-97cd-13f205d9fcdb', | |
51 | 'ext': 'mp4', | |
52 | 'title': "Yahoo Saves 'Community'", | |
53 | 'description': 'md5:4d4145af2fd3de00cbb6c1d664105053', | |
54 | 'duration': 170, | |
55 | } | |
56 | }, | |
57 | { | |
58 | 'url': 'https://tw.screen.yahoo.com/election-2014-askmayor/敢問市長-黃秀霜批賴清德-非常高傲-033009720.html', | |
59 | 'md5': '3a09cf59349cfaddae1797acc3c087fc', | |
60 | 'info_dict': { | |
61 | 'id': 'cac903b3-fcf4-3c14-b632-643ab541712f', | |
62 | 'ext': 'mp4', | |
63 | 'title': '敢問市長/黃秀霜批賴清德「非常高傲」', | |
64 | 'description': '直言台南沒捷運 交通居五都之末', | |
65 | 'duration': 396, | |
66 | } | |
67 | }, | |
68 | { | |
69 | 'url': 'https://uk.screen.yahoo.com/editor-picks/cute-raccoon-freed-drain-using-091756545.html', | |
70 | 'md5': '0b51660361f0e27c9789e7037ef76f4b', | |
71 | 'info_dict': { | |
72 | 'id': 'b3affa53-2e14-3590-852b-0e0db6cd1a58', | |
73 | 'ext': 'mp4', | |
74 | 'title': 'Cute Raccoon Freed From Drain\u00a0Using Angle Grinder', | |
75 | 'description': 'md5:f66c890e1490f4910a9953c941dee944', | |
76 | 'duration': 97, | |
77 | } | |
78 | }, | |
79 | { | |
80 | 'url': 'https://ca.sports.yahoo.com/video/program-makes-hockey-more-affordable-013127711.html', | |
81 | 'md5': '57e06440778b1828a6079d2f744212c4', | |
82 | 'info_dict': { | |
83 | 'id': 'c9fa2a36-0d4d-3937-b8f6-cc0fb1881e73', | |
84 | 'ext': 'mp4', | |
85 | 'title': 'Program that makes hockey more affordable not offered in Manitoba', | |
86 | 'description': 'md5:c54a609f4c078d92b74ffb9bf1f496f4', | |
87 | 'duration': 121, | |
88 | } | |
89 | }, { | |
90 | 'url': 'https://ca.finance.yahoo.com/news/hackers-sony-more-trouble-well-154609075.html', | |
91 | 'md5': '226a895aae7e21b0129e2a2006fe9690', | |
92 | 'info_dict': { | |
93 | 'id': 'e624c4bc-3389-34de-9dfc-025f74943409', | |
94 | 'ext': 'mp4', | |
95 | 'title': '\'The Interview\' TV Spot: War', | |
96 | 'description': 'The Interview', | |
97 | 'duration': 30, | |
98 | } | |
99 | }, { | |
100 | 'url': 'http://news.yahoo.com/video/china-moses-crazy-blues-104538833.html', | |
101 | 'md5': '67010fdf3a08d290e060a4dd96baa07b', | |
102 | 'info_dict': { | |
103 | 'id': 'f885cf7f-43d4-3450-9fac-46ac30ece521', | |
104 | 'ext': 'mp4', | |
105 | 'title': 'China Moses Is Crazy About the Blues', | |
106 | 'description': 'md5:9900ab8cd5808175c7b3fe55b979bed0', | |
107 | 'duration': 128, | |
108 | } | |
109 | }, { | |
110 | 'url': 'https://in.lifestyle.yahoo.com/video/connect-dots-dark-side-virgo-090247395.html', | |
111 | 'md5': 'd9a083ccf1379127bf25699d67e4791b', | |
112 | 'info_dict': { | |
113 | 'id': '52aeeaa3-b3d1-30d8-9ef8-5d0cf05efb7c', | |
114 | 'ext': 'mp4', | |
115 | 'title': 'Connect the Dots: Dark Side of Virgo', | |
116 | 'description': 'md5:1428185051cfd1949807ad4ff6d3686a', | |
117 | 'duration': 201, | |
118 | } | |
119 | }, { | |
120 | 'url': 'https://www.yahoo.com/movies/v/true-story-trailer-173000497.html', | |
121 | 'md5': '989396ae73d20c6f057746fb226aa215', | |
122 | 'info_dict': { | |
123 | 'id': '071c4013-ce30-3a93-a5b2-e0413cd4a9d1', | |
124 | 'ext': 'mp4', | |
125 | 'title': '\'True Story\' Trailer', | |
126 | 'description': 'True Story', | |
127 | 'duration': 150, | |
128 | }, | |
129 | }, { | |
130 | 'url': 'https://gma.yahoo.com/pizza-delivery-man-surprised-huge-tip-college-kids-195200785.html', | |
131 | 'only_matching': True, | |
132 | } | |
133 | ] | |
134 | ||
135 | def _real_extract(self, url): | |
136 | mobj = re.match(self._VALID_URL, url) | |
137 | display_id = mobj.group('display_id') | |
138 | page_id = mobj.group('id') | |
139 | url = mobj.group('url') | |
140 | host = mobj.group('host') | |
141 | webpage = self._download_webpage(url, display_id) | |
142 | ||
143 | # Look for iframed media first | |
144 | iframe_m = re.search(r'<iframe[^>]+src="(/video/.+?-\d+\.html\?format=embed.*?)"', webpage) | |
145 | if iframe_m: | |
146 | iframepage = self._download_webpage( | |
147 | host + iframe_m.group(1), display_id, 'Downloading iframe webpage') | |
148 | items_json = self._search_regex( | |
149 | r'mediaItems: (\[.+?\])$', iframepage, 'items', flags=re.MULTILINE, default=None) | |
150 | if items_json: | |
151 | items = json.loads(items_json) | |
152 | video_id = items[0]['id'] | |
153 | return self._get_info(video_id, display_id, webpage) | |
154 | ||
155 | items_json = self._search_regex( | |
156 | r'mediaItems: ({.*?})$', webpage, 'items', flags=re.MULTILINE, | |
157 | default=None) | |
158 | if items_json is None: | |
159 | CONTENT_ID_REGEXES = [ | |
160 | r'YUI\.namespace\("Media"\)\.CONTENT_ID\s*=\s*"([^"]+)"', | |
161 | r'root\.App\.Cache\.context\.videoCache\.curVideo = \{"([^"]+)"', | |
162 | r'"first_videoid"\s*:\s*"([^"]+)"', | |
163 | r'%s[^}]*"ccm_id"\s*:\s*"([^"]+)"' % re.escape(page_id), | |
164 | ] | |
165 | video_id = self._search_regex(CONTENT_ID_REGEXES, webpage, 'content ID') | |
166 | else: | |
167 | items = json.loads(items_json) | |
168 | info = items['mediaItems']['query']['results']['mediaObj'][0] | |
169 | # The 'meta' field is not always in the video webpage, we request it | |
170 | # from another page | |
171 | video_id = info['id'] | |
172 | return self._get_info(video_id, display_id, webpage) | |
173 | ||
174 | def _get_info(self, video_id, display_id, webpage): | |
175 | region = self._search_regex( | |
176 | r'\\?"region\\?"\s*:\s*\\?"([^"]+?)\\?"', | |
177 | webpage, 'region', fatal=False, default='US') | |
178 | data = compat_urllib_parse.urlencode({ | |
179 | 'protocol': 'http', | |
180 | 'region': region, | |
181 | }) | |
182 | query_url = ( | |
183 | 'https://video.media.yql.yahoo.com/v1/video/sapi/streams/' | |
184 | '{id}?{data}'.format(id=video_id, data=data)) | |
185 | query_result = self._download_json( | |
186 | query_url, display_id, 'Downloading video info') | |
187 | ||
188 | info = query_result['query']['results']['mediaObj'][0] | |
189 | meta = info.get('meta') | |
190 | ||
191 | if not meta: | |
192 | msg = info['status'].get('msg') | |
193 | if msg: | |
194 | raise ExtractorError( | |
195 | '%s returned error: %s' % (self.IE_NAME, msg), expected=True) | |
196 | raise ExtractorError('Unable to extract media object meta') | |
197 | ||
198 | formats = [] | |
199 | for s in info['streams']: | |
200 | format_info = { | |
201 | 'width': int_or_none(s.get('width')), | |
202 | 'height': int_or_none(s.get('height')), | |
203 | 'tbr': int_or_none(s.get('bitrate')), | |
204 | } | |
205 | ||
206 | host = s['host'] | |
207 | path = s['path'] | |
208 | if host.startswith('rtmp'): | |
209 | format_info.update({ | |
210 | 'url': host, | |
211 | 'play_path': path, | |
212 | 'ext': 'flv', | |
213 | }) | |
214 | else: | |
215 | format_url = compat_urlparse.urljoin(host, path) | |
216 | format_info['url'] = format_url | |
217 | formats.append(format_info) | |
218 | ||
219 | self._sort_formats(formats) | |
220 | ||
221 | return { | |
222 | 'id': video_id, | |
223 | 'display_id': display_id, | |
224 | 'title': unescapeHTML(meta['title']), | |
225 | 'formats': formats, | |
226 | 'description': clean_html(meta['description']), | |
227 | 'thumbnail': meta['thumbnail'] if meta.get('thumbnail') else self._og_search_thumbnail(webpage), | |
228 | 'duration': int_or_none(meta.get('duration')), | |
229 | } | |
230 | ||
231 | ||
232 | class YahooSearchIE(SearchInfoExtractor): | |
233 | IE_DESC = 'Yahoo screen search' | |
234 | _MAX_RESULTS = 1000 | |
235 | IE_NAME = 'screen.yahoo:search' | |
236 | _SEARCH_KEY = 'yvsearch' | |
237 | ||
238 | def _get_n_results(self, query, n): | |
239 | """Get a specified number of results for a query""" | |
240 | entries = [] | |
241 | for pagenum in itertools.count(0): | |
242 | result_url = 'http://video.search.yahoo.com/search/?p=%s&fr=screen&o=js&gs=0&b=%d' % (compat_urllib_parse.quote_plus(query), pagenum * 30) | |
243 | info = self._download_json(result_url, query, | |
244 | note='Downloading results page ' + str(pagenum + 1)) | |
245 | m = info['m'] | |
246 | results = info['results'] | |
247 | ||
248 | for (i, r) in enumerate(results): | |
249 | if (pagenum * 30) + i >= n: | |
250 | break | |
251 | mobj = re.search(r'(?P<url>screen\.yahoo\.com/.*?-\d*?\.html)"', r) | |
252 | e = self.url_result('http://' + mobj.group('url'), 'Yahoo') | |
253 | entries.append(e) | |
254 | if (pagenum * 30 + i >= n) or (m['last'] >= (m['total'] - 1)): | |
255 | break | |
256 | ||
257 | return { | |
258 | '_type': 'playlist', | |
259 | 'id': query, | |
260 | 'entries': entries, | |
261 | } |