]>
Commit | Line | Data |
---|---|---|
28746fbd PH |
1 | # coding: utf-8 |
2 | from __future__ import unicode_literals | |
3 | ||
04b32c8f | 4 | import hashlib |
6efb0711 | 5 | import itertools |
06167fbb | 6 | import json |
c34f505b | 7 | import functools |
520e7533 | 8 | import re |
c34f505b | 9 | import math |
28746fbd | 10 | |
06167fbb | 11 | from .common import InfoExtractor, SearchInfoExtractor |
bd8f48c7 | 12 | from ..compat import ( |
adc74b3c | 13 | compat_str, |
bd8f48c7 YCH |
14 | compat_parse_qs, |
15 | compat_urlparse, | |
c34f505b | 16 | compat_urllib_parse_urlparse |
bd8f48c7 | 17 | ) |
28746fbd | 18 | from ..utils import ( |
bd8f48c7 | 19 | ExtractorError, |
6461f2b7 YCH |
20 | int_or_none, |
21 | float_or_none, | |
bd8f48c7 | 22 | parse_iso8601, |
adc74b3c | 23 | try_get, |
bd8f48c7 | 24 | smuggle_url, |
4bc15a68 | 25 | str_or_none, |
bd8f48c7 | 26 | strip_jsonp, |
04b32c8f | 27 | unified_timestamp, |
bd8f48c7 | 28 | unsmuggle_url, |
1f85029d | 29 | urlencode_postdata, |
c34f505b | 30 | OnDemandPagedList |
28746fbd PH |
31 | ) |
32 | ||
33 | ||
34 | class BiliBiliIE(InfoExtractor): | |
b4eb08bb S |
35 | _VALID_URL = r'''(?x) |
36 | https?:// | |
37 | (?:(?:www|bangumi)\.)? | |
38 | bilibili\.(?:tv|com)/ | |
39 | (?: | |
40 | (?: | |
41 | video/[aA][vV]| | |
42 | anime/(?P<anime_id>\d+)/play\# | |
06167fbb | 43 | )(?P<id>\d+)| |
9536bc07 | 44 | (s/)?video/[bB][vV](?P<id_bv>[^/?#&]+) |
b4eb08bb | 45 | ) |
06167fbb | 46 | (?:/?\?p=(?P<page>\d+))? |
b4eb08bb | 47 | ''' |
28746fbd | 48 | |
bd8f48c7 | 49 | _TESTS = [{ |
06167fbb | 50 | 'url': 'http://www.bilibili.com/video/av1074402/', |
3526c304 | 51 | 'md5': '5f7d29e1a2872f3df0cf76b1f87d3788', |
28746fbd | 52 | 'info_dict': { |
04b32c8f | 53 | 'id': '1074402', |
3526c304 | 54 | 'ext': 'flv', |
28746fbd | 55 | 'title': '【金坷垃】金泡沫', |
6461f2b7 | 56 | 'description': 'md5:ce18c2a2d2193f0df2917d270f2e5923', |
3526c304 S |
57 | 'duration': 308.067, |
58 | 'timestamp': 1398012678, | |
28746fbd | 59 | 'upload_date': '20140420', |
ec85ded8 | 60 | 'thumbnail': r're:^https?://.+\.jpg', |
d90e4030 | 61 | 'uploader': '菊子桑', |
6461f2b7 | 62 | 'uploader_id': '156160', |
28746fbd | 63 | }, |
bd8f48c7 YCH |
64 | }, { |
65 | # Tested in BiliBiliBangumiIE | |
66 | 'url': 'http://bangumi.bilibili.com/anime/1869/play#40062', | |
67 | 'only_matching': True, | |
06167fbb | 68 | }, { |
69 | # bilibili.tv | |
70 | 'url': 'http://www.bilibili.tv/video/av1074402/', | |
71 | 'only_matching': True, | |
bd8f48c7 YCH |
72 | }, { |
73 | 'url': 'http://bangumi.bilibili.com/anime/5802/play#100643', | |
74 | 'md5': '3f721ad1e75030cc06faf73587cfec57', | |
75 | 'info_dict': { | |
76 | 'id': '100643', | |
77 | 'ext': 'mp4', | |
78 | 'title': 'CHAOS;CHILD', | |
79 | 'description': '如果你是神明,并且能够让妄想成为现实。那你会进行怎么样的妄想?是淫靡的世界?独裁社会?毁灭性的制裁?还是……2015年,涩谷。从6年前发生的大灾害“涩谷地震”之后复兴了的这个街区里新设立的私立高中...', | |
80 | }, | |
81 | 'skip': 'Geo-restricted to China', | |
ca270371 YCH |
82 | }, { |
83 | # Title with double quotes | |
84 | 'url': 'http://www.bilibili.com/video/av8903802/', | |
85 | 'info_dict': { | |
86 | 'id': '8903802', | |
ca270371 YCH |
87 | 'title': '阿滴英文|英文歌分享#6 "Closer', |
88 | 'description': '滴妹今天唱Closer給你聽! 有史以来,被推最多次也是最久的歌曲,其实歌词跟我原本想像差蛮多的,不过还是好听! 微博@阿滴英文', | |
ca270371 | 89 | }, |
3526c304 S |
90 | 'playlist': [{ |
91 | 'info_dict': { | |
92 | 'id': '8903802_part1', | |
93 | 'ext': 'flv', | |
94 | 'title': '阿滴英文|英文歌分享#6 "Closer', | |
95 | 'description': 'md5:3b1b9e25b78da4ef87e9b548b88ee76a', | |
96 | 'uploader': '阿滴英文', | |
97 | 'uploader_id': '65880958', | |
98 | 'timestamp': 1488382634, | |
99 | 'upload_date': '20170301', | |
100 | }, | |
101 | 'params': { | |
102 | 'skip_download': True, # Test metadata only | |
103 | }, | |
104 | }, { | |
105 | 'info_dict': { | |
106 | 'id': '8903802_part2', | |
107 | 'ext': 'flv', | |
108 | 'title': '阿滴英文|英文歌分享#6 "Closer', | |
109 | 'description': 'md5:3b1b9e25b78da4ef87e9b548b88ee76a', | |
110 | 'uploader': '阿滴英文', | |
111 | 'uploader_id': '65880958', | |
112 | 'timestamp': 1488382634, | |
113 | 'upload_date': '20170301', | |
114 | }, | |
115 | 'params': { | |
116 | 'skip_download': True, # Test metadata only | |
117 | }, | |
118 | }] | |
b4eb08bb S |
119 | }, { |
120 | # new BV video id format | |
121 | 'url': 'https://www.bilibili.com/video/BV1JE411F741', | |
122 | 'only_matching': True, | |
adc74b3c | 123 | }, { |
124 | # Anthology | |
125 | 'url': 'https://www.bilibili.com/video/BV1bK411W797', | |
126 | 'info_dict': { | |
127 | 'id': 'BV1bK411W797', | |
7e60c069 | 128 | 'title': '物语中的人物是如何吐槽自己的OP的' |
adc74b3c | 129 | }, |
130 | 'playlist_count': 17, | |
bd8f48c7 | 131 | }] |
28746fbd | 132 | |
c9a0ea6e S |
133 | _APP_KEY = 'iVGUTjsxvpLeuDCf' |
134 | _BILIBILI_KEY = 'aHRmhWMLkdeMuILqORnYZocwMBpMEOdt' | |
6461f2b7 | 135 | |
bd8f48c7 YCH |
136 | def _report_error(self, result): |
137 | if 'message' in result: | |
138 | raise ExtractorError('%s said: %s' % (self.IE_NAME, result['message']), expected=True) | |
139 | elif 'code' in result: | |
140 | raise ExtractorError('%s returns error %d' % (self.IE_NAME, result['code']), expected=True) | |
141 | else: | |
142 | raise ExtractorError('Can\'t extract Bangumi episode ID') | |
143 | ||
520e7533 | 144 | def _real_extract(self, url): |
bd8f48c7 YCH |
145 | url, smuggled_data = unsmuggle_url(url, {}) |
146 | ||
5ad28e7f | 147 | mobj = self._match_valid_url(url) |
06167fbb | 148 | video_id = mobj.group('id_bv') or mobj.group('id') |
149 | ||
150 | av_id, bv_id = self._get_video_id_set(video_id, mobj.group('id_bv') is not None) | |
151 | video_id = av_id | |
152 | ||
bd8f48c7 | 153 | anime_id = mobj.group('anime_id') |
06167fbb | 154 | page_id = mobj.group('page') |
6461f2b7 YCH |
155 | webpage = self._download_webpage(url, video_id) |
156 | ||
adc74b3c | 157 | # Bilibili anthologies are similar to playlists but all videos share the same video ID as the anthology itself. |
158 | # If the video has no page argument, check to see if it's an anthology | |
159 | if page_id is None: | |
a06916d9 | 160 | if not self.get_param('noplaylist'): |
adc74b3c | 161 | r = self._extract_anthology_entries(bv_id, video_id, webpage) |
162 | if r is not None: | |
163 | self.to_screen('Downloading anthology %s - add --no-playlist to just download video' % video_id) | |
164 | return r | |
7e60c069 | 165 | else: |
166 | self.to_screen('Downloading just video %s because of --no-playlist' % video_id) | |
adc74b3c | 167 | |
bd8f48c7 | 168 | if 'anime/' not in url: |
3526c304 | 169 | cid = self._search_regex( |
adc74b3c | 170 | r'\bcid(?:["\']:|=)(\d+),["\']page(?:["\']:|=)' + compat_str(page_id), webpage, 'cid', |
06167fbb | 171 | default=None |
172 | ) or self._search_regex( | |
61cb6683 | 173 | r'\bcid(?:["\']:|=)(\d+)', webpage, 'cid', |
3526c304 S |
174 | default=None |
175 | ) or compat_parse_qs(self._search_regex( | |
95a1322b S |
176 | [r'EmbedPlayer\([^)]+,\s*"([^"]+)"\)', |
177 | r'EmbedPlayer\([^)]+,\s*\\"([^"]+)\\"\)', | |
178 | r'<iframe[^>]+src="https://secure\.bilibili\.com/secure,([^"]+)"'], | |
7be15d40 P |
179 | webpage, 'player parameters'))['cid'][0] |
180 | else: | |
bd8f48c7 | 181 | if 'no_bangumi_tip' not in smuggled_data: |
7a5c1cfe | 182 | self.to_screen('Downloading episode %s. To download all videos in anime %s, re-run yt-dlp with %s' % ( |
bd8f48c7 | 183 | video_id, anime_id, compat_urlparse.urljoin(url, '//bangumi.bilibili.com/anime/%s' % anime_id))) |
10db0d2f | 184 | headers = { |
185 | 'Content-Type': 'application/x-www-form-urlencoded; charset=UTF-8', | |
186 | 'Referer': url | |
187 | } | |
188 | headers.update(self.geo_verification_headers()) | |
bd8f48c7 | 189 | |
1f85029d YCH |
190 | js = self._download_json( |
191 | 'http://bangumi.bilibili.com/web_api/get_source', video_id, | |
192 | data=urlencode_postdata({'episode_id': video_id}), | |
bd8f48c7 YCH |
193 | headers=headers) |
194 | if 'result' not in js: | |
195 | self._report_error(js) | |
7be15d40 | 196 | cid = js['result']['cid'] |
04b32c8f | 197 | |
10db0d2f | 198 | headers = { |
199 | 'Accept': 'application/json', | |
200 | 'Referer': url | |
201 | } | |
202 | headers.update(self.geo_verification_headers()) | |
203 | ||
d90e4030 | 204 | entries = [] |
c4a21bc9 | 205 | |
3526c304 S |
206 | RENDITIONS = ('qn=80&quality=80&type=', 'quality=2&type=mp4') |
207 | for num, rendition in enumerate(RENDITIONS, start=1): | |
208 | payload = 'appkey=%s&cid=%s&otype=json&%s' % (self._APP_KEY, cid, rendition) | |
209 | sign = hashlib.md5((payload + self._BILIBILI_KEY).encode('utf-8')).hexdigest() | |
210 | ||
211 | video_info = self._download_json( | |
d1239608 | 212 | 'http://interface.bilibili.com/v2/playurl?%s&sign=%s' % (payload, sign), |
3526c304 S |
213 | video_id, note='Downloading video info page', |
214 | headers=headers, fatal=num == len(RENDITIONS)) | |
215 | ||
216 | if not video_info: | |
217 | continue | |
218 | ||
219 | if 'durl' not in video_info: | |
220 | if num < len(RENDITIONS): | |
221 | continue | |
222 | self._report_error(video_info) | |
223 | ||
224 | for idx, durl in enumerate(video_info['durl']): | |
225 | formats = [{ | |
226 | 'url': durl['url'], | |
227 | 'filesize': int_or_none(durl['size']), | |
228 | }] | |
229 | for backup_url in durl.get('backup_url', []): | |
230 | formats.append({ | |
231 | 'url': backup_url, | |
232 | # backup URLs have lower priorities | |
f983b875 | 233 | 'quality': -2 if 'hd.mp4' in backup_url else -3, |
3526c304 S |
234 | }) |
235 | ||
236 | for a_format in formats: | |
237 | a_format.setdefault('http_headers', {}).update({ | |
238 | 'Referer': url, | |
239 | }) | |
240 | ||
241 | self._sort_formats(formats) | |
242 | ||
243 | entries.append({ | |
244 | 'id': '%s_part%s' % (video_id, idx), | |
245 | 'duration': float_or_none(durl.get('length'), 1000), | |
246 | 'formats': formats, | |
6461f2b7 | 247 | }) |
3526c304 | 248 | break |
6461f2b7 | 249 | |
3526c304 | 250 | title = self._html_search_regex( |
06167fbb | 251 | (r'<h1[^>]+\btitle=(["\'])(?P<title>(?:(?!\1).)+)\1', |
252 | r'(?s)<h1[^>]*>(?P<title>.+?)</h1>'), webpage, 'title', | |
adc74b3c | 253 | group='title') |
254 | ||
255 | # Get part title for anthologies | |
256 | if page_id is not None: | |
257 | # TODO: The json is already downloaded by _extract_anthology_entries. Don't redownload for each video | |
258 | part_title = try_get( | |
259 | self._download_json( | |
260 | "https://api.bilibili.com/x/player/pagelist?bvid=%s&jsonp=jsonp" % bv_id, | |
261 | video_id, note='Extracting videos in anthology'), | |
262 | lambda x: x['data'][int(page_id) - 1]['part']) | |
263 | title = part_title or title | |
264 | ||
6461f2b7 | 265 | description = self._html_search_meta('description', webpage) |
04b32c8f | 266 | timestamp = unified_timestamp(self._html_search_regex( |
3526c304 S |
267 | r'<time[^>]+datetime="([^"]+)"', webpage, 'upload time', |
268 | default=None) or self._html_search_meta( | |
269 | 'uploadDate', webpage, 'timestamp', default=None)) | |
1f85029d | 270 | thumbnail = self._html_search_meta(['og:image', 'thumbnailUrl'], webpage) |
6461f2b7 YCH |
271 | |
272 | # TODO 'view_count' requires deobfuscating Javascript | |
d90e4030 | 273 | info = { |
adc74b3c | 274 | 'id': compat_str(video_id) if page_id is None else '%s_p%s' % (video_id, page_id), |
06167fbb | 275 | 'cid': cid, |
d90e4030 | 276 | 'title': title, |
6461f2b7 YCH |
277 | 'description': description, |
278 | 'timestamp': timestamp, | |
7be15d40 | 279 | 'thumbnail': thumbnail, |
04b32c8f | 280 | 'duration': float_or_none(video_info.get('timelength'), scale=1000), |
28746fbd | 281 | } |
d90e4030 | 282 | |
6461f2b7 | 283 | uploader_mobj = re.search( |
7e60c069 | 284 | r'<a[^>]+href="(?:https?:)?//space\.bilibili\.com/(?P<id>\d+)"[^>]*>\s*(?P<name>[^<]+?)\s*<', |
6461f2b7 YCH |
285 | webpage) |
286 | if uploader_mobj: | |
287 | info.update({ | |
ed807c18 | 288 | 'uploader': uploader_mobj.group('name').strip(), |
6461f2b7 YCH |
289 | 'uploader_id': uploader_mobj.group('id'), |
290 | }) | |
06167fbb | 291 | |
3526c304 S |
292 | if not info.get('uploader'): |
293 | info['uploader'] = self._html_search_meta( | |
294 | 'author', webpage, 'uploader', default=None) | |
6461f2b7 | 295 | |
06167fbb | 296 | raw_danmaku = self._get_raw_danmaku(video_id, cid) |
297 | ||
298 | raw_tags = self._get_tags(video_id) | |
299 | tags = list(map(lambda x: x['tag_name'], raw_tags)) | |
300 | ||
301 | top_level_info = { | |
302 | 'raw_danmaku': raw_danmaku, | |
06167fbb | 303 | 'tags': tags, |
304 | 'raw_tags': raw_tags, | |
305 | } | |
a06916d9 | 306 | if self.get_param('getcomments', False): |
277d6ff5 | 307 | def get_comments(): |
308 | comments = self._get_all_comment_pages(video_id) | |
309 | return { | |
310 | 'comments': comments, | |
311 | 'comment_count': len(comments) | |
312 | } | |
313 | ||
314 | top_level_info['__post_extractor'] = get_comments | |
06167fbb | 315 | |
316 | ''' | |
317 | # Requires https://github.com/m13253/danmaku2ass which is licenced under GPL3 | |
318 | # See https://github.com/animelover1984/youtube-dl | |
319 | danmaku = NiconicoIE.CreateDanmaku(raw_danmaku, commentType='Bilibili', x=1024, y=576) | |
320 | entries[0]['subtitles'] = { | |
321 | 'danmaku': [{ | |
322 | 'ext': 'ass', | |
323 | 'data': danmaku | |
324 | }] | |
325 | } | |
326 | ''' | |
327 | ||
6461f2b7 YCH |
328 | for entry in entries: |
329 | entry.update(info) | |
330 | ||
d90e4030 | 331 | if len(entries) == 1: |
06167fbb | 332 | entries[0].update(top_level_info) |
d90e4030 | 333 | return entries[0] |
334 | else: | |
ad73083f YCH |
335 | for idx, entry in enumerate(entries): |
336 | entry['id'] = '%s_part%d' % (video_id, (idx + 1)) | |
337 | ||
06167fbb | 338 | global_info = { |
d90e4030 | 339 | '_type': 'multi_video', |
adc74b3c | 340 | 'id': compat_str(video_id), |
06167fbb | 341 | 'bv_id': bv_id, |
6461f2b7 YCH |
342 | 'title': title, |
343 | 'description': description, | |
d90e4030 | 344 | 'entries': entries, |
6461f2b7 | 345 | } |
bd8f48c7 | 346 | |
06167fbb | 347 | global_info.update(info) |
348 | global_info.update(top_level_info) | |
349 | ||
350 | return global_info | |
351 | ||
adc74b3c | 352 | def _extract_anthology_entries(self, bv_id, video_id, webpage): |
353 | title = self._html_search_regex( | |
354 | (r'<h1[^>]+\btitle=(["\'])(?P<title>(?:(?!\1).)+)\1', | |
355 | r'(?s)<h1[^>]*>(?P<title>.+?)</h1>'), webpage, 'title', | |
356 | group='title') | |
357 | json_data = self._download_json( | |
358 | "https://api.bilibili.com/x/player/pagelist?bvid=%s&jsonp=jsonp" % bv_id, | |
359 | video_id, note='Extracting videos in anthology') | |
360 | ||
361 | if len(json_data['data']) > 1: | |
362 | return self.playlist_from_matches( | |
363 | json_data['data'], bv_id, title, ie=BiliBiliIE.ie_key(), | |
364 | getter=lambda entry: 'https://www.bilibili.com/video/%s?p=%d' % (bv_id, entry['page'])) | |
365 | ||
06167fbb | 366 | def _get_video_id_set(self, id, is_bv): |
367 | query = {'bvid': id} if is_bv else {'aid': id} | |
368 | response = self._download_json( | |
369 | "http://api.bilibili.cn/x/web-interface/view", | |
370 | id, query=query, | |
371 | note='Grabbing original ID via API') | |
372 | ||
373 | if response['code'] == -400: | |
374 | raise ExtractorError('Video ID does not exist', expected=True, video_id=id) | |
375 | elif response['code'] != 0: | |
376 | raise ExtractorError('Unknown error occurred during API check (code %s)' % response['code'], expected=True, video_id=id) | |
377 | return (response['data']['aid'], response['data']['bvid']) | |
378 | ||
379 | # recursive solution to getting every page of comments for the video | |
380 | # we can stop when we reach a page without any comments | |
381 | def _get_all_comment_pages(self, video_id, commentPageNumber=0): | |
382 | comment_url = "https://api.bilibili.com/x/v2/reply?jsonp=jsonp&pn=%s&type=1&oid=%s&sort=2&_=1567227301685" % (commentPageNumber, video_id) | |
383 | json_str = self._download_webpage( | |
384 | comment_url, video_id, | |
385 | note='Extracting comments from page %s' % (commentPageNumber)) | |
386 | replies = json.loads(json_str)['data']['replies'] | |
387 | if replies is None: | |
388 | return [] | |
389 | return self._get_all_children(replies) + self._get_all_comment_pages(video_id, commentPageNumber + 1) | |
390 | ||
391 | # extracts all comments in the tree | |
392 | def _get_all_children(self, replies): | |
393 | if replies is None: | |
394 | return [] | |
395 | ||
396 | ret = [] | |
397 | for reply in replies: | |
398 | author = reply['member']['uname'] | |
399 | author_id = reply['member']['mid'] | |
400 | id = reply['rpid'] | |
401 | text = reply['content']['message'] | |
402 | timestamp = reply['ctime'] | |
403 | parent = reply['parent'] if reply['parent'] != 0 else 'root' | |
404 | ||
405 | comment = { | |
406 | "author": author, | |
407 | "author_id": author_id, | |
408 | "id": id, | |
409 | "text": text, | |
410 | "timestamp": timestamp, | |
411 | "parent": parent, | |
412 | } | |
413 | ret.append(comment) | |
414 | ||
415 | # from the JSON, the comment structure seems arbitrarily deep, but I could be wrong. | |
416 | # Regardless, this should work. | |
417 | ret += self._get_all_children(reply['replies']) | |
418 | ||
419 | return ret | |
420 | ||
421 | def _get_raw_danmaku(self, video_id, cid): | |
422 | # This will be useful if I decide to scrape all pages instead of doing them individually | |
423 | # cid_url = "https://www.bilibili.com/widget/getPageList?aid=%s" % (video_id) | |
424 | # cid_str = self._download_webpage(cid_url, video_id, note=False) | |
425 | # cid = json.loads(cid_str)[0]['cid'] | |
426 | ||
427 | danmaku_url = "https://comment.bilibili.com/%s.xml" % (cid) | |
428 | danmaku = self._download_webpage(danmaku_url, video_id, note='Downloading danmaku comments') | |
429 | return danmaku | |
430 | ||
431 | def _get_tags(self, video_id): | |
432 | tags_url = "https://api.bilibili.com/x/tag/archive/tags?aid=%s" % (video_id) | |
433 | tags_json = self._download_json(tags_url, video_id, note='Downloading tags') | |
434 | return tags_json['data'] | |
435 | ||
bd8f48c7 YCH |
436 | |
437 | class BiliBiliBangumiIE(InfoExtractor): | |
438 | _VALID_URL = r'https?://bangumi\.bilibili\.com/anime/(?P<id>\d+)' | |
439 | ||
440 | IE_NAME = 'bangumi.bilibili.com' | |
441 | IE_DESC = 'BiliBili番剧' | |
442 | ||
443 | _TESTS = [{ | |
444 | 'url': 'http://bangumi.bilibili.com/anime/1869', | |
445 | 'info_dict': { | |
446 | 'id': '1869', | |
447 | 'title': '混沌武士', | |
448 | 'description': 'md5:6a9622b911565794c11f25f81d6a97d2', | |
449 | }, | |
450 | 'playlist_count': 26, | |
451 | }, { | |
452 | 'url': 'http://bangumi.bilibili.com/anime/1869', | |
453 | 'info_dict': { | |
454 | 'id': '1869', | |
455 | 'title': '混沌武士', | |
456 | 'description': 'md5:6a9622b911565794c11f25f81d6a97d2', | |
457 | }, | |
458 | 'playlist': [{ | |
459 | 'md5': '91da8621454dd58316851c27c68b0c13', | |
460 | 'info_dict': { | |
461 | 'id': '40062', | |
462 | 'ext': 'mp4', | |
463 | 'title': '混沌武士', | |
464 | 'description': '故事发生在日本的江户时代。风是一个小酒馆的打工女。一日,酒馆里来了一群恶霸,虽然他们的举动令风十分不满,但是毕竟风只是一届女流,无法对他们采取什么行动,只能在心里嘟哝。这时,酒家里又进来了个“不良份子...', | |
465 | 'timestamp': 1414538739, | |
466 | 'upload_date': '20141028', | |
467 | 'episode': '疾风怒涛 Tempestuous Temperaments', | |
468 | 'episode_number': 1, | |
469 | }, | |
470 | }], | |
471 | 'params': { | |
472 | 'playlist_items': '1', | |
473 | }, | |
474 | }] | |
475 | ||
476 | @classmethod | |
477 | def suitable(cls, url): | |
478 | return False if BiliBiliIE.suitable(url) else super(BiliBiliBangumiIE, cls).suitable(url) | |
479 | ||
480 | def _real_extract(self, url): | |
481 | bangumi_id = self._match_id(url) | |
482 | ||
483 | # Sometimes this API returns a JSONP response | |
484 | season_info = self._download_json( | |
485 | 'http://bangumi.bilibili.com/jsonp/seasoninfo/%s.ver' % bangumi_id, | |
486 | bangumi_id, transform_source=strip_jsonp)['result'] | |
487 | ||
488 | entries = [{ | |
489 | '_type': 'url_transparent', | |
490 | 'url': smuggle_url(episode['webplay_url'], {'no_bangumi_tip': 1}), | |
491 | 'ie_key': BiliBiliIE.ie_key(), | |
492 | 'timestamp': parse_iso8601(episode.get('update_time'), delimiter=' '), | |
493 | 'episode': episode.get('index_title'), | |
494 | 'episode_number': int_or_none(episode.get('index')), | |
495 | } for episode in season_info['episodes']] | |
496 | ||
497 | entries = sorted(entries, key=lambda entry: entry.get('episode_number')) | |
498 | ||
499 | return self.playlist_result( | |
500 | entries, bangumi_id, | |
501 | season_info.get('bangumi_title'), season_info.get('evaluate')) | |
4bc15a68 RA |
502 | |
503 | ||
06167fbb | 504 | class BilibiliChannelIE(InfoExtractor): |
505 | _VALID_URL = r'https?://space.bilibili\.com/(?P<id>\d+)' | |
6efb0711 | 506 | _API_URL = "https://api.bilibili.com/x/space/arc/search?mid=%s&pn=%d&jsonp=jsonp" |
507 | _TESTS = [{ | |
508 | 'url': 'https://space.bilibili.com/3985676/video', | |
509 | 'info_dict': {}, | |
510 | 'playlist_mincount': 112, | |
511 | }] | |
512 | ||
513 | def _entries(self, list_id): | |
514 | count, max_count = 0, None | |
515 | ||
516 | for page_num in itertools.count(1): | |
517 | data = self._parse_json( | |
518 | self._download_webpage( | |
519 | self._API_URL % (list_id, page_num), list_id, | |
520 | note='Downloading page %d' % page_num), | |
521 | list_id)['data'] | |
522 | ||
523 | max_count = max_count or try_get(data, lambda x: x['page']['count']) | |
524 | ||
525 | entries = try_get(data, lambda x: x['list']['vlist']) | |
526 | if not entries: | |
527 | return | |
528 | for entry in entries: | |
529 | yield self.url_result( | |
530 | 'https://www.bilibili.com/video/%s' % entry['bvid'], | |
531 | BiliBiliIE.ie_key(), entry['bvid']) | |
532 | ||
533 | count += len(entries) | |
534 | if max_count and count >= max_count: | |
535 | return | |
06167fbb | 536 | |
537 | def _real_extract(self, url): | |
538 | list_id = self._match_id(url) | |
6efb0711 | 539 | return self.playlist_result(self._entries(list_id), list_id) |
06167fbb | 540 | |
541 | ||
c34f505b | 542 | class BilibiliCategoryIE(InfoExtractor): |
543 | IE_NAME = 'Bilibili category extractor' | |
544 | _MAX_RESULTS = 1000000 | |
545 | _VALID_URL = r'https?://www\.bilibili\.com/v/[a-zA-Z]+\/[a-zA-Z]+' | |
546 | _TESTS = [{ | |
547 | 'url': 'https://www.bilibili.com/v/kichiku/mad', | |
548 | 'info_dict': { | |
549 | 'id': 'kichiku: mad', | |
550 | 'title': 'kichiku: mad' | |
551 | }, | |
552 | 'playlist_mincount': 45, | |
553 | 'params': { | |
554 | 'playlistend': 45 | |
555 | } | |
556 | }] | |
557 | ||
558 | def _fetch_page(self, api_url, num_pages, query, page_num): | |
559 | parsed_json = self._download_json( | |
560 | api_url, query, query={'Search_key': query, 'pn': page_num}, | |
561 | note='Extracting results from page %s of %s' % (page_num, num_pages)) | |
562 | ||
563 | video_list = try_get(parsed_json, lambda x: x['data']['archives'], list) | |
564 | if not video_list: | |
565 | raise ExtractorError('Failed to retrieve video list for page %d' % page_num) | |
566 | ||
567 | for video in video_list: | |
568 | yield self.url_result( | |
569 | 'https://www.bilibili.com/video/%s' % video['bvid'], 'BiliBili', video['bvid']) | |
570 | ||
571 | def _entries(self, category, subcategory, query): | |
572 | # map of categories : subcategories : RIDs | |
573 | rid_map = { | |
574 | 'kichiku': { | |
575 | 'mad': 26, | |
576 | 'manual_vocaloid': 126, | |
577 | 'guide': 22, | |
578 | 'theatre': 216, | |
579 | 'course': 127 | |
580 | }, | |
581 | } | |
582 | ||
583 | if category not in rid_map: | |
584 | raise ExtractorError('The supplied category, %s, is not supported. List of supported categories: %s' % (category, list(rid_map.keys()))) | |
585 | ||
586 | if subcategory not in rid_map[category]: | |
587 | raise ExtractorError('The subcategory, %s, isn\'t supported for this category. Supported subcategories: %s' % (subcategory, list(rid_map[category].keys()))) | |
588 | ||
589 | rid_value = rid_map[category][subcategory] | |
590 | ||
591 | api_url = 'https://api.bilibili.com/x/web-interface/newlist?rid=%d&type=1&ps=20&jsonp=jsonp' % rid_value | |
592 | page_json = self._download_json(api_url, query, query={'Search_key': query, 'pn': '1'}) | |
593 | page_data = try_get(page_json, lambda x: x['data']['page'], dict) | |
594 | count, size = int_or_none(page_data.get('count')), int_or_none(page_data.get('size')) | |
595 | if count is None or not size: | |
596 | raise ExtractorError('Failed to calculate either page count or size') | |
597 | ||
598 | num_pages = math.ceil(count / size) | |
599 | ||
600 | return OnDemandPagedList(functools.partial( | |
601 | self._fetch_page, api_url, num_pages, query), size) | |
602 | ||
603 | def _real_extract(self, url): | |
604 | u = compat_urllib_parse_urlparse(url) | |
605 | category, subcategory = u.path.split('/')[2:4] | |
606 | query = '%s: %s' % (category, subcategory) | |
607 | ||
608 | return self.playlist_result(self._entries(category, subcategory, query), query, query) | |
609 | ||
610 | ||
06167fbb | 611 | class BiliBiliSearchIE(SearchInfoExtractor): |
612 | IE_DESC = 'Bilibili video search, "bilisearch" keyword' | |
613 | _MAX_RESULTS = 100000 | |
614 | _SEARCH_KEY = 'bilisearch' | |
615 | MAX_NUMBER_OF_RESULTS = 1000 | |
616 | ||
617 | def _get_n_results(self, query, n): | |
618 | """Get a specified number of results for a query""" | |
619 | ||
620 | entries = [] | |
621 | pageNumber = 0 | |
622 | while True: | |
623 | pageNumber += 1 | |
624 | # FIXME | |
625 | api_url = "https://api.bilibili.com/x/web-interface/search/type?context=&page=%s&order=pubdate&keyword=%s&duration=0&tids_2=&__refresh__=true&search_type=video&tids=0&highlight=1" % (pageNumber, query) | |
626 | json_str = self._download_webpage( | |
627 | api_url, "None", query={"Search_key": query}, | |
628 | note='Extracting results from page %s' % pageNumber) | |
629 | data = json.loads(json_str)['data'] | |
630 | ||
631 | # FIXME: this is hideous | |
632 | if "result" not in data: | |
633 | return { | |
634 | '_type': 'playlist', | |
635 | 'id': query, | |
636 | 'entries': entries[:n] | |
637 | } | |
638 | ||
639 | videos = data['result'] | |
640 | for video in videos: | |
adc74b3c | 641 | e = self.url_result(video['arcurl'], 'BiliBili', compat_str(video['aid'])) |
06167fbb | 642 | entries.append(e) |
643 | ||
644 | if(len(entries) >= n or len(videos) >= BiliBiliSearchIE.MAX_NUMBER_OF_RESULTS): | |
645 | return { | |
646 | '_type': 'playlist', | |
647 | 'id': query, | |
648 | 'entries': entries[:n] | |
649 | } | |
650 | ||
651 | ||
4bc15a68 RA |
652 | class BilibiliAudioBaseIE(InfoExtractor): |
653 | def _call_api(self, path, sid, query=None): | |
654 | if not query: | |
655 | query = {'sid': sid} | |
656 | return self._download_json( | |
657 | 'https://www.bilibili.com/audio/music-service-c/web/' + path, | |
658 | sid, query=query)['data'] | |
659 | ||
660 | ||
661 | class BilibiliAudioIE(BilibiliAudioBaseIE): | |
662 | _VALID_URL = r'https?://(?:www\.)?bilibili\.com/audio/au(?P<id>\d+)' | |
663 | _TEST = { | |
664 | 'url': 'https://www.bilibili.com/audio/au1003142', | |
665 | 'md5': 'fec4987014ec94ef9e666d4d158ad03b', | |
666 | 'info_dict': { | |
667 | 'id': '1003142', | |
668 | 'ext': 'm4a', | |
669 | 'title': '【tsukimi】YELLOW / 神山羊', | |
670 | 'artist': 'tsukimi', | |
671 | 'comment_count': int, | |
672 | 'description': 'YELLOW的mp3版!', | |
673 | 'duration': 183, | |
674 | 'subtitles': { | |
675 | 'origin': [{ | |
676 | 'ext': 'lrc', | |
677 | }], | |
678 | }, | |
679 | 'thumbnail': r're:^https?://.+\.jpg', | |
680 | 'timestamp': 1564836614, | |
681 | 'upload_date': '20190803', | |
682 | 'uploader': 'tsukimi-つきみぐー', | |
683 | 'view_count': int, | |
684 | }, | |
685 | } | |
686 | ||
687 | def _real_extract(self, url): | |
688 | au_id = self._match_id(url) | |
689 | ||
690 | play_data = self._call_api('url', au_id) | |
691 | formats = [{ | |
692 | 'url': play_data['cdns'][0], | |
693 | 'filesize': int_or_none(play_data.get('size')), | |
f0884c8b | 694 | 'vcodec': 'none' |
4bc15a68 RA |
695 | }] |
696 | ||
697 | song = self._call_api('song/info', au_id) | |
698 | title = song['title'] | |
699 | statistic = song.get('statistic') or {} | |
700 | ||
701 | subtitles = None | |
702 | lyric = song.get('lyric') | |
703 | if lyric: | |
704 | subtitles = { | |
705 | 'origin': [{ | |
706 | 'url': lyric, | |
707 | }] | |
708 | } | |
709 | ||
710 | return { | |
711 | 'id': au_id, | |
712 | 'title': title, | |
713 | 'formats': formats, | |
714 | 'artist': song.get('author'), | |
715 | 'comment_count': int_or_none(statistic.get('comment')), | |
716 | 'description': song.get('intro'), | |
717 | 'duration': int_or_none(song.get('duration')), | |
718 | 'subtitles': subtitles, | |
719 | 'thumbnail': song.get('cover'), | |
720 | 'timestamp': int_or_none(song.get('passtime')), | |
721 | 'uploader': song.get('uname'), | |
722 | 'view_count': int_or_none(statistic.get('play')), | |
723 | } | |
724 | ||
725 | ||
726 | class BilibiliAudioAlbumIE(BilibiliAudioBaseIE): | |
727 | _VALID_URL = r'https?://(?:www\.)?bilibili\.com/audio/am(?P<id>\d+)' | |
728 | _TEST = { | |
729 | 'url': 'https://www.bilibili.com/audio/am10624', | |
730 | 'info_dict': { | |
731 | 'id': '10624', | |
732 | 'title': '每日新曲推荐(每日11:00更新)', | |
733 | 'description': '每天11:00更新,为你推送最新音乐', | |
734 | }, | |
735 | 'playlist_count': 19, | |
736 | } | |
737 | ||
738 | def _real_extract(self, url): | |
739 | am_id = self._match_id(url) | |
740 | ||
741 | songs = self._call_api( | |
742 | 'song/of-menu', am_id, {'sid': am_id, 'pn': 1, 'ps': 100})['data'] | |
743 | ||
744 | entries = [] | |
745 | for song in songs: | |
746 | sid = str_or_none(song.get('id')) | |
747 | if not sid: | |
748 | continue | |
749 | entries.append(self.url_result( | |
750 | 'https://www.bilibili.com/audio/au' + sid, | |
751 | BilibiliAudioIE.ie_key(), sid)) | |
752 | ||
753 | if entries: | |
754 | album_data = self._call_api('menu/info', am_id) or {} | |
755 | album_title = album_data.get('title') | |
756 | if album_title: | |
757 | for entry in entries: | |
758 | entry['album'] = album_title | |
759 | return self.playlist_result( | |
760 | entries, am_id, album_title, album_data.get('intro')) | |
761 | ||
762 | return self.playlist_result(entries, am_id) | |
63dce309 S |
763 | |
764 | ||
765 | class BiliBiliPlayerIE(InfoExtractor): | |
766 | _VALID_URL = r'https?://player\.bilibili\.com/player\.html\?.*?\baid=(?P<id>\d+)' | |
767 | _TEST = { | |
768 | 'url': 'http://player.bilibili.com/player.html?aid=92494333&cid=157926707&page=1', | |
769 | 'only_matching': True, | |
770 | } | |
771 | ||
772 | def _real_extract(self, url): | |
773 | video_id = self._match_id(url) | |
774 | return self.url_result( | |
775 | 'http://www.bilibili.tv/video/av%s/' % video_id, | |
776 | ie=BiliBiliIE.ie_key(), video_id=video_id) |