]>
Commit | Line | Data |
---|---|---|
c5e8d7af | 1 | # coding: utf-8 |
c5e8d7af PH |
2 | |
3 | import json | |
4 | import netrc | |
5 | import re | |
6 | import socket | |
7 | ||
b05654f0 | 8 | from .common import InfoExtractor, SearchInfoExtractor |
c5e8d7af PH |
9 | from ..utils import ( |
10 | compat_http_client, | |
11 | compat_parse_qs, | |
12 | compat_urllib_error, | |
13 | compat_urllib_parse, | |
14 | compat_urllib_request, | |
15 | compat_str, | |
16 | ||
17 | clean_html, | |
18 | get_element_by_id, | |
19 | ExtractorError, | |
20 | unescapeHTML, | |
21 | unified_strdate, | |
22 | ) | |
23 | ||
24 | ||
25 | class YoutubeIE(InfoExtractor): | |
26 | """Information extractor for youtube.com.""" | |
27 | ||
28 | _VALID_URL = r"""^ | |
29 | ( | |
30 | (?:https?://)? # http(s):// (optional) | |
31 | (?:youtu\.be/|(?:\w+\.)?youtube(?:-nocookie)?\.com/| | |
32 | tube\.majestyc\.net/) # the various hostnames, with wildcard subdomains | |
33 | (?:.*?\#/)? # handle anchor (#/) redirect urls | |
34 | (?: # the various things that can precede the ID: | |
35 | (?:(?:v|embed|e)/) # v/ or embed/ or e/ | |
36 | |(?: # or the v= param in all its forms | |
37 | (?:watch(?:_popup)?(?:\.php)?)? # preceding watch(_popup|.php) or nothing (like /?v=xxxx) | |
38 | (?:\?|\#!?) # the params delimiter ? or # or #! | |
39 | (?:.*?&)? # any other preceding param (like /?s=tuff&v=xxxx) | |
40 | v= | |
41 | ) | |
42 | )? # optional -> youtube.com/xxxx is OK | |
43 | )? # all until now is optional -> you can pass the naked ID | |
44 | ([0-9A-Za-z_-]+) # here is it! the YouTube video ID | |
45 | (?(1).+)? # if we found the ID, everything can follow | |
46 | $""" | |
47 | _LANG_URL = r'https://www.youtube.com/?hl=en&persist_hl=1&gl=US&persist_gl=1&opt_out_ackd=1' | |
48 | _LOGIN_URL = 'https://accounts.google.com/ServiceLogin' | |
49 | _AGE_URL = 'http://www.youtube.com/verify_age?next_url=/&gl=US&hl=en' | |
50 | _NEXT_URL_RE = r'[\?&]next_url=([^&]+)' | |
51 | _NETRC_MACHINE = 'youtube' | |
52 | # Listed in order of quality | |
53 | _available_formats = ['38', '37', '46', '22', '45', '35', '44', '34', '18', '43', '6', '5', '17', '13'] | |
54 | _available_formats_prefer_free = ['38', '46', '37', '45', '22', '44', '35', '43', '34', '18', '6', '5', '17', '13'] | |
55 | _video_extensions = { | |
56 | '13': '3gp', | |
57 | '17': 'mp4', | |
58 | '18': 'mp4', | |
59 | '22': 'mp4', | |
60 | '37': 'mp4', | |
d69cf69a | 61 | '38': 'mp4', |
c5e8d7af PH |
62 | '43': 'webm', |
63 | '44': 'webm', | |
64 | '45': 'webm', | |
65 | '46': 'webm', | |
66 | } | |
67 | _video_dimensions = { | |
68 | '5': '240x400', | |
69 | '6': '???', | |
70 | '13': '???', | |
71 | '17': '144x176', | |
72 | '18': '360x640', | |
73 | '22': '720x1280', | |
74 | '34': '360x640', | |
75 | '35': '480x854', | |
76 | '37': '1080x1920', | |
77 | '38': '3072x4096', | |
78 | '43': '360x640', | |
79 | '44': '480x854', | |
80 | '45': '720x1280', | |
81 | '46': '1080x1920', | |
82 | } | |
83 | IE_NAME = u'youtube' | |
2eb88d95 PH |
84 | _TESTS = [ |
85 | { | |
86 | "url": "http://www.youtube.com/watch?v=BaW_jenozKc", | |
87 | "file": "BaW_jenozKc.mp4", | |
88 | "info_dict": { | |
89 | "title": "youtube-dl test video \"'/\\ä↭𝕐", | |
90 | "uploader": "Philipp Hagemeister", | |
91 | "uploader_id": "phihag", | |
92 | "upload_date": "20121002", | |
93 | "description": "test chars: \"'/\\ä↭𝕐\n\nThis is a test video for youtube-dl.\n\nFor more information, contact phihag@phihag.de ." | |
94 | } | |
95 | }, | |
96 | { | |
97 | "url": "http://www.youtube.com/watch?v=1ltcDfZMA3U", | |
98 | "file": "1ltcDfZMA3U.flv", | |
99 | "note": "Test VEVO video (#897)", | |
100 | "info_dict": { | |
101 | "upload_date": "20070518", | |
102 | "title": "Maps - It Will Find You", | |
103 | "description": "Music video by Maps performing It Will Find You.", | |
104 | "uploader": "MuteUSA", | |
105 | "uploader_id": "MuteUSA" | |
106 | } | |
107 | }, | |
108 | { | |
109 | "url": "http://www.youtube.com/watch?v=UxxajLWwzqY", | |
110 | "file": "UxxajLWwzqY.mp4", | |
111 | "note": "Test generic use_cipher_signature video (#897)", | |
112 | "info_dict": { | |
113 | "upload_date": "20120506", | |
114 | "title": "Icona Pop - I Love It (feat. Charli XCX) [OFFICIAL VIDEO]", | |
115 | "description": "md5:b085c9804f5ab69f4adea963a2dceb3c", | |
116 | "uploader": "IconaPop", | |
117 | "uploader_id": "IconaPop" | |
118 | } | |
119 | } | |
120 | ] | |
121 | ||
c5e8d7af PH |
122 | |
123 | @classmethod | |
124 | def suitable(cls, url): | |
125 | """Receives a URL and returns True if suitable for this IE.""" | |
126 | if YoutubePlaylistIE.suitable(url): return False | |
127 | return re.match(cls._VALID_URL, url, re.VERBOSE) is not None | |
128 | ||
129 | def report_lang(self): | |
130 | """Report attempt to set language.""" | |
131 | self.to_screen(u'Setting language') | |
132 | ||
133 | def report_login(self): | |
134 | """Report attempt to log in.""" | |
135 | self.to_screen(u'Logging in') | |
136 | ||
137 | def report_video_webpage_download(self, video_id): | |
138 | """Report attempt to download video webpage.""" | |
139 | self.to_screen(u'%s: Downloading video webpage' % video_id) | |
140 | ||
141 | def report_video_info_webpage_download(self, video_id): | |
142 | """Report attempt to download video info webpage.""" | |
143 | self.to_screen(u'%s: Downloading video info webpage' % video_id) | |
144 | ||
145 | def report_video_subtitles_download(self, video_id): | |
146 | """Report attempt to download video info webpage.""" | |
147 | self.to_screen(u'%s: Checking available subtitles' % video_id) | |
148 | ||
149 | def report_video_subtitles_request(self, video_id, sub_lang, format): | |
150 | """Report attempt to download video info webpage.""" | |
151 | self.to_screen(u'%s: Downloading video subtitles for %s.%s' % (video_id, sub_lang, format)) | |
152 | ||
153 | def report_video_subtitles_available(self, video_id, sub_lang_list): | |
154 | """Report available subtitles.""" | |
155 | sub_lang = ",".join(list(sub_lang_list.keys())) | |
156 | self.to_screen(u'%s: Available subtitles for video: %s' % (video_id, sub_lang)) | |
157 | ||
158 | def report_information_extraction(self, video_id): | |
159 | """Report attempt to extract video information.""" | |
160 | self.to_screen(u'%s: Extracting video information' % video_id) | |
161 | ||
162 | def report_unavailable_format(self, video_id, format): | |
163 | """Report extracted video URL.""" | |
164 | self.to_screen(u'%s: Format %s not available' % (video_id, format)) | |
165 | ||
166 | def report_rtmp_download(self): | |
167 | """Indicate the download will use the RTMP protocol.""" | |
168 | self.to_screen(u'RTMP download detected') | |
169 | ||
98bcd283 | 170 | def _decrypt_signature(self, s): |
c90f13d1 | 171 | """Decrypt the key""" |
6b37f0be PH |
172 | |
173 | if len(s) == 88: | |
ee313cdc | 174 | return s[48] + s[81:67:-1] + s[82] + s[66:62:-1] + s[85] + s[61:48:-1] + s[67] + s[47:12:-1] + s[3] + s[11:3:-1] + s[2] + s[12] |
6b37f0be | 175 | elif len(s) == 87: |
ee313cdc | 176 | return s[62] + s[82:62:-1] + s[83] + s[61:52:-1] + s[0] + s[51:2:-1] |
6b37f0be | 177 | elif len(s) == 86: |
23300d71 | 178 | return s[2:63] + s[82] + s[64:82] + s[63] |
6b37f0be | 179 | elif len(s) == 85: |
ee313cdc | 180 | return s[76] + s[82:76:-1] + s[83] + s[75:60:-1] + s[0] + s[59:50:-1] + s[1] + s[49:2:-1] |
6b37f0be | 181 | elif len(s) == 84: |
ee313cdc | 182 | return s[83:36:-1] + s[2] + s[35:26:-1] + s[3] + s[25:3:-1] + s[26] |
6b37f0be | 183 | elif len(s) == 83: |
ee313cdc | 184 | return s[52] + s[81:55:-1] + s[2] + s[54:52:-1] + s[82] + s[51:36:-1] + s[55] + s[35:2:-1] + s[36] |
6b37f0be | 185 | elif len(s) == 82: |
ee313cdc PH |
186 | return s[36] + s[79:67:-1] + s[81] + s[66:40:-1] + s[33] + s[39:36:-1] + s[40] + s[35] + s[0] + s[67] + s[32:0:-1] + s[34] |
187 | ||
6b37f0be | 188 | else: |
c90f13d1 | 189 | raise ExtractorError(u'Unable to decrypt signature, key length %d not supported; retrying might work' % (len(s))) |
c5e8d7af PH |
190 | |
191 | def _get_available_subtitles(self, video_id): | |
192 | self.report_video_subtitles_download(video_id) | |
193 | request = compat_urllib_request.Request('http://video.google.com/timedtext?hl=en&type=list&v=%s' % video_id) | |
194 | try: | |
195 | sub_list = compat_urllib_request.urlopen(request).read().decode('utf-8') | |
196 | except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err: | |
197 | return (u'unable to download video subtitles: %s' % compat_str(err), None) | |
198 | sub_lang_list = re.findall(r'name="([^"]*)"[^>]+lang_code="([\w\-]+)"', sub_list) | |
199 | sub_lang_list = dict((l[1], l[0]) for l in sub_lang_list) | |
200 | if not sub_lang_list: | |
201 | return (u'video doesn\'t have subtitles', None) | |
202 | return sub_lang_list | |
203 | ||
204 | def _list_available_subtitles(self, video_id): | |
205 | sub_lang_list = self._get_available_subtitles(video_id) | |
206 | self.report_video_subtitles_available(video_id, sub_lang_list) | |
207 | ||
208 | def _request_subtitle(self, sub_lang, sub_name, video_id, format): | |
209 | """ | |
210 | Return tuple: | |
211 | (error_message, sub_lang, sub) | |
212 | """ | |
213 | self.report_video_subtitles_request(video_id, sub_lang, format) | |
214 | params = compat_urllib_parse.urlencode({ | |
215 | 'lang': sub_lang, | |
216 | 'name': sub_name, | |
217 | 'v': video_id, | |
218 | 'fmt': format, | |
219 | }) | |
220 | url = 'http://www.youtube.com/api/timedtext?' + params | |
221 | try: | |
222 | sub = compat_urllib_request.urlopen(url).read().decode('utf-8') | |
223 | except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err: | |
224 | return (u'unable to download video subtitles: %s' % compat_str(err), None, None) | |
225 | if not sub: | |
226 | return (u'Did not fetch video subtitles', None, None) | |
227 | return (None, sub_lang, sub) | |
228 | ||
229 | def _request_automatic_caption(self, video_id, webpage): | |
230 | """We need the webpage for getting the captions url, pass it as an | |
231 | argument to speed up the process.""" | |
232 | sub_lang = self._downloader.params.get('subtitleslang') or 'en' | |
233 | sub_format = self._downloader.params.get('subtitlesformat') | |
234 | self.to_screen(u'%s: Looking for automatic captions' % video_id) | |
235 | mobj = re.search(r';ytplayer.config = ({.*?});', webpage) | |
236 | err_msg = u'Couldn\'t find automatic captions for "%s"' % sub_lang | |
237 | if mobj is None: | |
238 | return [(err_msg, None, None)] | |
239 | player_config = json.loads(mobj.group(1)) | |
240 | try: | |
241 | args = player_config[u'args'] | |
242 | caption_url = args[u'ttsurl'] | |
243 | timestamp = args[u'timestamp'] | |
244 | params = compat_urllib_parse.urlencode({ | |
245 | 'lang': 'en', | |
246 | 'tlang': sub_lang, | |
247 | 'fmt': sub_format, | |
248 | 'ts': timestamp, | |
249 | 'kind': 'asr', | |
250 | }) | |
251 | subtitles_url = caption_url + '&' + params | |
252 | sub = self._download_webpage(subtitles_url, video_id, u'Downloading automatic captions') | |
253 | return [(None, sub_lang, sub)] | |
254 | except KeyError: | |
255 | return [(err_msg, None, None)] | |
256 | ||
257 | def _extract_subtitle(self, video_id): | |
258 | """ | |
259 | Return a list with a tuple: | |
260 | [(error_message, sub_lang, sub)] | |
261 | """ | |
262 | sub_lang_list = self._get_available_subtitles(video_id) | |
263 | sub_format = self._downloader.params.get('subtitlesformat') | |
264 | if isinstance(sub_lang_list,tuple): #There was some error, it didn't get the available subtitles | |
265 | return [(sub_lang_list[0], None, None)] | |
266 | if self._downloader.params.get('subtitleslang', False): | |
267 | sub_lang = self._downloader.params.get('subtitleslang') | |
268 | elif 'en' in sub_lang_list: | |
269 | sub_lang = 'en' | |
270 | else: | |
271 | sub_lang = list(sub_lang_list.keys())[0] | |
272 | if not sub_lang in sub_lang_list: | |
273 | return [(u'no closed captions found in the specified language "%s"' % sub_lang, None, None)] | |
274 | ||
275 | subtitle = self._request_subtitle(sub_lang, sub_lang_list[sub_lang].encode('utf-8'), video_id, sub_format) | |
276 | return [subtitle] | |
277 | ||
278 | def _extract_all_subtitles(self, video_id): | |
279 | sub_lang_list = self._get_available_subtitles(video_id) | |
280 | sub_format = self._downloader.params.get('subtitlesformat') | |
281 | if isinstance(sub_lang_list,tuple): #There was some error, it didn't get the available subtitles | |
282 | return [(sub_lang_list[0], None, None)] | |
283 | subtitles = [] | |
284 | for sub_lang in sub_lang_list: | |
285 | subtitle = self._request_subtitle(sub_lang, sub_lang_list[sub_lang].encode('utf-8'), video_id, sub_format) | |
286 | subtitles.append(subtitle) | |
287 | return subtitles | |
288 | ||
289 | def _print_formats(self, formats): | |
290 | print('Available formats:') | |
291 | for x in formats: | |
292 | print('%s\t:\t%s\t[%s]' %(x, self._video_extensions.get(x, 'flv'), self._video_dimensions.get(x, '???'))) | |
293 | ||
294 | def _real_initialize(self): | |
295 | if self._downloader is None: | |
296 | return | |
297 | ||
298 | username = None | |
299 | password = None | |
300 | downloader_params = self._downloader.params | |
301 | ||
302 | # Attempt to use provided username and password or .netrc data | |
303 | if downloader_params.get('username', None) is not None: | |
304 | username = downloader_params['username'] | |
305 | password = downloader_params['password'] | |
306 | elif downloader_params.get('usenetrc', False): | |
307 | try: | |
308 | info = netrc.netrc().authenticators(self._NETRC_MACHINE) | |
309 | if info is not None: | |
310 | username = info[0] | |
311 | password = info[2] | |
312 | else: | |
313 | raise netrc.NetrcParseError('No authenticators for %s' % self._NETRC_MACHINE) | |
314 | except (IOError, netrc.NetrcParseError) as err: | |
315 | self._downloader.report_warning(u'parsing .netrc: %s' % compat_str(err)) | |
316 | return | |
317 | ||
318 | # Set language | |
319 | request = compat_urllib_request.Request(self._LANG_URL) | |
320 | try: | |
321 | self.report_lang() | |
322 | compat_urllib_request.urlopen(request).read() | |
323 | except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err: | |
324 | self._downloader.report_warning(u'unable to set language: %s' % compat_str(err)) | |
325 | return | |
326 | ||
327 | # No authentication to be performed | |
328 | if username is None: | |
329 | return | |
330 | ||
331 | request = compat_urllib_request.Request(self._LOGIN_URL) | |
332 | try: | |
333 | login_page = compat_urllib_request.urlopen(request).read().decode('utf-8') | |
334 | except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err: | |
335 | self._downloader.report_warning(u'unable to fetch login page: %s' % compat_str(err)) | |
336 | return | |
337 | ||
338 | galx = None | |
339 | dsh = None | |
340 | match = re.search(re.compile(r'<input.+?name="GALX".+?value="(.+?)"', re.DOTALL), login_page) | |
341 | if match: | |
342 | galx = match.group(1) | |
343 | ||
344 | match = re.search(re.compile(r'<input.+?name="dsh".+?value="(.+?)"', re.DOTALL), login_page) | |
345 | if match: | |
346 | dsh = match.group(1) | |
347 | ||
348 | # Log in | |
349 | login_form_strs = { | |
350 | u'continue': u'https://www.youtube.com/signin?action_handle_signin=true&feature=sign_in_button&hl=en_US&nomobiletemp=1', | |
351 | u'Email': username, | |
352 | u'GALX': galx, | |
353 | u'Passwd': password, | |
354 | u'PersistentCookie': u'yes', | |
355 | u'_utf8': u'霱', | |
356 | u'bgresponse': u'js_disabled', | |
357 | u'checkConnection': u'', | |
358 | u'checkedDomains': u'youtube', | |
359 | u'dnConn': u'', | |
360 | u'dsh': dsh, | |
361 | u'pstMsg': u'0', | |
362 | u'rmShown': u'1', | |
363 | u'secTok': u'', | |
364 | u'signIn': u'Sign in', | |
365 | u'timeStmp': u'', | |
366 | u'service': u'youtube', | |
367 | u'uilel': u'3', | |
368 | u'hl': u'en_US', | |
369 | } | |
370 | # Convert to UTF-8 *before* urlencode because Python 2.x's urlencode | |
371 | # chokes on unicode | |
372 | login_form = dict((k.encode('utf-8'), v.encode('utf-8')) for k,v in login_form_strs.items()) | |
373 | login_data = compat_urllib_parse.urlencode(login_form).encode('ascii') | |
374 | request = compat_urllib_request.Request(self._LOGIN_URL, login_data) | |
375 | try: | |
376 | self.report_login() | |
377 | login_results = compat_urllib_request.urlopen(request).read().decode('utf-8') | |
378 | if re.search(r'(?i)<form[^>]* id="gaia_loginform"', login_results) is not None: | |
379 | self._downloader.report_warning(u'unable to log in: bad username or password') | |
380 | return | |
381 | except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err: | |
382 | self._downloader.report_warning(u'unable to log in: %s' % compat_str(err)) | |
383 | return | |
384 | ||
385 | # Confirm age | |
386 | age_form = { | |
387 | 'next_url': '/', | |
388 | 'action_confirm': 'Confirm', | |
389 | } | |
390 | request = compat_urllib_request.Request(self._AGE_URL, compat_urllib_parse.urlencode(age_form)) | |
391 | try: | |
392 | self.report_age_confirmation() | |
93d3a642 | 393 | compat_urllib_request.urlopen(request).read().decode('utf-8') |
c5e8d7af PH |
394 | except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err: |
395 | raise ExtractorError(u'Unable to confirm age: %s' % compat_str(err)) | |
396 | ||
397 | def _extract_id(self, url): | |
398 | mobj = re.match(self._VALID_URL, url, re.VERBOSE) | |
399 | if mobj is None: | |
400 | raise ExtractorError(u'Invalid URL: %s' % url) | |
401 | video_id = mobj.group(2) | |
402 | return video_id | |
403 | ||
404 | def _real_extract(self, url): | |
405 | # Extract original video URL from URL with redirection, like age verification, using next_url parameter | |
406 | mobj = re.search(self._NEXT_URL_RE, url) | |
407 | if mobj: | |
408 | url = 'https://www.youtube.com/' + compat_urllib_parse.unquote(mobj.group(1)).lstrip('/') | |
409 | video_id = self._extract_id(url) | |
410 | ||
411 | # Get video webpage | |
412 | self.report_video_webpage_download(video_id) | |
413 | url = 'https://www.youtube.com/watch?v=%s&gl=US&hl=en&has_verified=1' % video_id | |
414 | request = compat_urllib_request.Request(url) | |
415 | try: | |
416 | video_webpage_bytes = compat_urllib_request.urlopen(request).read() | |
417 | except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err: | |
418 | raise ExtractorError(u'Unable to download video webpage: %s' % compat_str(err)) | |
419 | ||
420 | video_webpage = video_webpage_bytes.decode('utf-8', 'ignore') | |
421 | ||
422 | # Attempt to extract SWF player URL | |
423 | mobj = re.search(r'swfConfig.*?"(http:\\/\\/.*?watch.*?-.*?\.swf)"', video_webpage) | |
424 | if mobj is not None: | |
425 | player_url = re.sub(r'\\(.)', r'\1', mobj.group(1)) | |
426 | else: | |
427 | player_url = None | |
428 | ||
429 | # Get video info | |
430 | self.report_video_info_webpage_download(video_id) | |
431 | for el_type in ['&el=embedded', '&el=detailpage', '&el=vevo', '']: | |
432 | video_info_url = ('https://www.youtube.com/get_video_info?&video_id=%s%s&ps=default&eurl=&gl=US&hl=en' | |
433 | % (video_id, el_type)) | |
434 | video_info_webpage = self._download_webpage(video_info_url, video_id, | |
435 | note=False, | |
436 | errnote='unable to download video info webpage') | |
437 | video_info = compat_parse_qs(video_info_webpage) | |
438 | if 'token' in video_info: | |
439 | break | |
440 | if 'token' not in video_info: | |
441 | if 'reason' in video_info: | |
442 | raise ExtractorError(u'YouTube said: %s' % video_info['reason'][0]) | |
443 | else: | |
444 | raise ExtractorError(u'"token" parameter not in video info for unknown reason') | |
445 | ||
446 | # Check for "rental" videos | |
447 | if 'ypc_video_rental_bar_text' in video_info and 'author' not in video_info: | |
448 | raise ExtractorError(u'"rental" videos not supported') | |
449 | ||
450 | # Start extracting information | |
451 | self.report_information_extraction(video_id) | |
452 | ||
453 | # uploader | |
454 | if 'author' not in video_info: | |
455 | raise ExtractorError(u'Unable to extract uploader name') | |
456 | video_uploader = compat_urllib_parse.unquote_plus(video_info['author'][0]) | |
457 | ||
458 | # uploader_id | |
459 | video_uploader_id = None | |
460 | mobj = re.search(r'<link itemprop="url" href="http://www.youtube.com/(?:user|channel)/([^"]+)">', video_webpage) | |
461 | if mobj is not None: | |
462 | video_uploader_id = mobj.group(1) | |
463 | else: | |
464 | self._downloader.report_warning(u'unable to extract uploader nickname') | |
465 | ||
466 | # title | |
467 | if 'title' not in video_info: | |
468 | raise ExtractorError(u'Unable to extract video title') | |
469 | video_title = compat_urllib_parse.unquote_plus(video_info['title'][0]) | |
470 | ||
471 | # thumbnail image | |
472 | if 'thumbnail_url' not in video_info: | |
473 | self._downloader.report_warning(u'unable to extract video thumbnail') | |
474 | video_thumbnail = '' | |
475 | else: # don't panic if we can't find it | |
476 | video_thumbnail = compat_urllib_parse.unquote_plus(video_info['thumbnail_url'][0]) | |
477 | ||
478 | # upload date | |
479 | upload_date = None | |
480 | mobj = re.search(r'id="eow-date.*?>(.*?)</span>', video_webpage, re.DOTALL) | |
481 | if mobj is not None: | |
482 | upload_date = ' '.join(re.sub(r'[/,-]', r' ', mobj.group(1)).split()) | |
483 | upload_date = unified_strdate(upload_date) | |
484 | ||
485 | # description | |
486 | video_description = get_element_by_id("eow-description", video_webpage) | |
487 | if video_description: | |
488 | video_description = clean_html(video_description) | |
489 | else: | |
490 | fd_mobj = re.search(r'<meta name="description" content="([^"]+)"', video_webpage) | |
491 | if fd_mobj: | |
492 | video_description = unescapeHTML(fd_mobj.group(1)) | |
493 | else: | |
494 | video_description = u'' | |
495 | ||
496 | # subtitles | |
497 | video_subtitles = None | |
498 | ||
499 | if self._downloader.params.get('writesubtitles', False): | |
500 | video_subtitles = self._extract_subtitle(video_id) | |
501 | if video_subtitles: | |
502 | (sub_error, sub_lang, sub) = video_subtitles[0] | |
503 | if sub_error: | |
b004821f JMF |
504 | self._downloader.report_warning(sub_error) |
505 | ||
506 | if self._downloader.params.get('writeautomaticsub', False): | |
507 | video_subtitles = self._request_automatic_caption(video_id, video_webpage) | |
508 | (sub_error, sub_lang, sub) = video_subtitles[0] | |
509 | if sub_error: | |
510 | self._downloader.report_warning(sub_error) | |
c5e8d7af PH |
511 | |
512 | if self._downloader.params.get('allsubtitles', False): | |
513 | video_subtitles = self._extract_all_subtitles(video_id) | |
514 | for video_subtitle in video_subtitles: | |
515 | (sub_error, sub_lang, sub) = video_subtitle | |
516 | if sub_error: | |
517 | self._downloader.report_warning(sub_error) | |
518 | ||
519 | if self._downloader.params.get('listsubtitles', False): | |
93d3a642 | 520 | self._list_available_subtitles(video_id) |
c5e8d7af PH |
521 | return |
522 | ||
523 | if 'length_seconds' not in video_info: | |
524 | self._downloader.report_warning(u'unable to extract video duration') | |
525 | video_duration = '' | |
526 | else: | |
527 | video_duration = compat_urllib_parse.unquote_plus(video_info['length_seconds'][0]) | |
528 | ||
c5e8d7af PH |
529 | # Decide which formats to download |
530 | req_format = self._downloader.params.get('format', None) | |
531 | ||
532 | try: | |
533 | mobj = re.search(r';ytplayer.config = ({.*?});', video_webpage) | |
50be92c1 PH |
534 | if not mobj: |
535 | raise ValueError('Could not find vevo ID') | |
c5e8d7af PH |
536 | info = json.loads(mobj.group(1)) |
537 | args = info['args'] | |
7ce7e394 JMF |
538 | # Easy way to know if the 's' value is in url_encoded_fmt_stream_map |
539 | # this signatures are encrypted | |
540 | m_s = re.search(r'[&,]s=', args['url_encoded_fmt_stream_map']) | |
541 | if m_s is not None: | |
542 | self.to_screen(u'%s: Encrypted signatures detected.' % video_id) | |
c5e8d7af PH |
543 | video_info['url_encoded_fmt_stream_map'] = [args['url_encoded_fmt_stream_map']] |
544 | except ValueError: | |
545 | pass | |
546 | ||
547 | if 'conn' in video_info and video_info['conn'][0].startswith('rtmp'): | |
548 | self.report_rtmp_download() | |
549 | video_url_list = [(None, video_info['conn'][0])] | |
550 | elif 'url_encoded_fmt_stream_map' in video_info and len(video_info['url_encoded_fmt_stream_map']) >= 1: | |
551 | url_map = {} | |
552 | for url_data_str in video_info['url_encoded_fmt_stream_map'][0].split(','): | |
553 | url_data = compat_parse_qs(url_data_str) | |
554 | if 'itag' in url_data and 'url' in url_data: | |
555 | url = url_data['url'][0] | |
556 | if 'sig' in url_data: | |
557 | url += '&signature=' + url_data['sig'][0] | |
558 | elif 's' in url_data: | |
769fda3c FV |
559 | if self._downloader.params.get('verbose'): |
560 | s = url_data['s'][0] | |
561 | player = self._search_regex(r'html5player-(.+?)\.js', video_webpage, | |
562 | 'html5 player', fatal=False) | |
563 | self.to_screen('encrypted signature length %d (%d.%d), itag %s, html5 player %s' % | |
564 | (len(s), len(s.split('.')[0]), len(s.split('.')[1]), url_data['itag'][0], player)) | |
c5e8d7af PH |
565 | signature = self._decrypt_signature(url_data['s'][0]) |
566 | url += '&signature=' + signature | |
567 | if 'ratebypass' not in url: | |
568 | url += '&ratebypass=yes' | |
569 | url_map[url_data['itag'][0]] = url | |
570 | ||
571 | format_limit = self._downloader.params.get('format_limit', None) | |
572 | available_formats = self._available_formats_prefer_free if self._downloader.params.get('prefer_free_formats', False) else self._available_formats | |
573 | if format_limit is not None and format_limit in available_formats: | |
574 | format_list = available_formats[available_formats.index(format_limit):] | |
575 | else: | |
576 | format_list = available_formats | |
577 | existing_formats = [x for x in format_list if x in url_map] | |
578 | if len(existing_formats) == 0: | |
579 | raise ExtractorError(u'no known formats available for video') | |
580 | if self._downloader.params.get('listformats', None): | |
581 | self._print_formats(existing_formats) | |
582 | return | |
583 | if req_format is None or req_format == 'best': | |
584 | video_url_list = [(existing_formats[0], url_map[existing_formats[0]])] # Best quality | |
585 | elif req_format == 'worst': | |
586 | video_url_list = [(existing_formats[len(existing_formats)-1], url_map[existing_formats[len(existing_formats)-1]])] # worst quality | |
587 | elif req_format in ('-1', 'all'): | |
588 | video_url_list = [(f, url_map[f]) for f in existing_formats] # All formats | |
589 | else: | |
590 | # Specific formats. We pick the first in a slash-delimeted sequence. | |
591 | # For example, if '1/2/3/4' is requested and '2' and '4' are available, we pick '2'. | |
592 | req_formats = req_format.split('/') | |
593 | video_url_list = None | |
594 | for rf in req_formats: | |
595 | if rf in url_map: | |
596 | video_url_list = [(rf, url_map[rf])] | |
597 | break | |
598 | if video_url_list is None: | |
599 | raise ExtractorError(u'requested format not available') | |
600 | else: | |
601 | raise ExtractorError(u'no conn or url_encoded_fmt_stream_map information found in video info') | |
602 | ||
603 | results = [] | |
604 | for format_param, video_real_url in video_url_list: | |
605 | # Extension | |
606 | video_extension = self._video_extensions.get(format_param, 'flv') | |
607 | ||
608 | video_format = '{0} - {1}'.format(format_param if format_param else video_extension, | |
609 | self._video_dimensions.get(format_param, '???')) | |
610 | ||
611 | results.append({ | |
612 | 'id': video_id, | |
613 | 'url': video_real_url, | |
614 | 'uploader': video_uploader, | |
615 | 'uploader_id': video_uploader_id, | |
616 | 'upload_date': upload_date, | |
617 | 'title': video_title, | |
618 | 'ext': video_extension, | |
619 | 'format': video_format, | |
620 | 'thumbnail': video_thumbnail, | |
621 | 'description': video_description, | |
622 | 'player_url': player_url, | |
623 | 'subtitles': video_subtitles, | |
624 | 'duration': video_duration | |
625 | }) | |
626 | return results | |
627 | ||
628 | class YoutubePlaylistIE(InfoExtractor): | |
629 | """Information Extractor for YouTube playlists.""" | |
630 | ||
631 | _VALID_URL = r"""(?: | |
632 | (?:https?://)? | |
633 | (?:\w+\.)? | |
634 | youtube\.com/ | |
635 | (?: | |
636 | (?:course|view_play_list|my_playlists|artist|playlist|watch) | |
637 | \? (?:.*?&)*? (?:p|a|list)= | |
638 | | p/ | |
639 | ) | |
640 | ((?:PL|EC|UU)?[0-9A-Za-z-_]{10,}) | |
641 | .* | |
642 | | | |
643 | ((?:PL|EC|UU)[0-9A-Za-z-_]{10,}) | |
644 | )""" | |
645 | _TEMPLATE_URL = 'https://gdata.youtube.com/feeds/api/playlists/%s?max-results=%i&start-index=%i&v=2&alt=json&safeSearch=none' | |
646 | _MAX_RESULTS = 50 | |
647 | IE_NAME = u'youtube:playlist' | |
648 | ||
649 | @classmethod | |
650 | def suitable(cls, url): | |
651 | """Receives a URL and returns True if suitable for this IE.""" | |
652 | return re.match(cls._VALID_URL, url, re.VERBOSE) is not None | |
653 | ||
654 | def _real_extract(self, url): | |
655 | # Extract playlist id | |
656 | mobj = re.match(self._VALID_URL, url, re.VERBOSE) | |
657 | if mobj is None: | |
658 | raise ExtractorError(u'Invalid URL: %s' % url) | |
659 | ||
660 | # Download playlist videos from API | |
661 | playlist_id = mobj.group(1) or mobj.group(2) | |
662 | page_num = 1 | |
663 | videos = [] | |
664 | ||
665 | while True: | |
666 | url = self._TEMPLATE_URL % (playlist_id, self._MAX_RESULTS, self._MAX_RESULTS * (page_num - 1) + 1) | |
667 | page = self._download_webpage(url, playlist_id, u'Downloading page #%s' % page_num) | |
668 | ||
669 | try: | |
670 | response = json.loads(page) | |
671 | except ValueError as err: | |
672 | raise ExtractorError(u'Invalid JSON in API response: ' + compat_str(err)) | |
673 | ||
674 | if 'feed' not in response: | |
675 | raise ExtractorError(u'Got a malformed response from YouTube API') | |
676 | playlist_title = response['feed']['title']['$t'] | |
677 | if 'entry' not in response['feed']: | |
678 | # Number of videos is a multiple of self._MAX_RESULTS | |
679 | break | |
680 | ||
681 | for entry in response['feed']['entry']: | |
682 | index = entry['yt$position']['$t'] | |
683 | if 'media$group' in entry and 'media$player' in entry['media$group']: | |
684 | videos.append((index, entry['media$group']['media$player']['url'])) | |
685 | ||
686 | if len(response['feed']['entry']) < self._MAX_RESULTS: | |
687 | break | |
688 | page_num += 1 | |
689 | ||
690 | videos = [v[1] for v in sorted(videos)] | |
691 | ||
692 | url_results = [self.url_result(url, 'Youtube') for url in videos] | |
693 | return [self.playlist_result(url_results, playlist_id, playlist_title)] | |
694 | ||
695 | ||
696 | class YoutubeChannelIE(InfoExtractor): | |
697 | """Information Extractor for YouTube channels.""" | |
698 | ||
699 | _VALID_URL = r"^(?:https?://)?(?:youtu\.be|(?:\w+\.)?youtube(?:-nocookie)?\.com)/channel/([0-9A-Za-z_-]+)" | |
700 | _TEMPLATE_URL = 'http://www.youtube.com/channel/%s/videos?sort=da&flow=list&view=0&page=%s&gl=US&hl=en' | |
701 | _MORE_PAGES_INDICATOR = 'yt-uix-load-more' | |
702 | _MORE_PAGES_URL = 'http://www.youtube.com/channel_ajax?action_load_more_videos=1&flow=list&paging=%s&view=0&sort=da&channel_id=%s' | |
703 | IE_NAME = u'youtube:channel' | |
704 | ||
705 | def extract_videos_from_page(self, page): | |
706 | ids_in_page = [] | |
707 | for mobj in re.finditer(r'href="/watch\?v=([0-9A-Za-z_-]+)&?', page): | |
708 | if mobj.group(1) not in ids_in_page: | |
709 | ids_in_page.append(mobj.group(1)) | |
710 | return ids_in_page | |
711 | ||
712 | def _real_extract(self, url): | |
713 | # Extract channel id | |
714 | mobj = re.match(self._VALID_URL, url) | |
715 | if mobj is None: | |
716 | raise ExtractorError(u'Invalid URL: %s' % url) | |
717 | ||
718 | # Download channel page | |
719 | channel_id = mobj.group(1) | |
720 | video_ids = [] | |
721 | pagenum = 1 | |
722 | ||
723 | url = self._TEMPLATE_URL % (channel_id, pagenum) | |
724 | page = self._download_webpage(url, channel_id, | |
725 | u'Downloading page #%s' % pagenum) | |
726 | ||
727 | # Extract video identifiers | |
728 | ids_in_page = self.extract_videos_from_page(page) | |
729 | video_ids.extend(ids_in_page) | |
730 | ||
731 | # Download any subsequent channel pages using the json-based channel_ajax query | |
732 | if self._MORE_PAGES_INDICATOR in page: | |
733 | while True: | |
734 | pagenum = pagenum + 1 | |
735 | ||
736 | url = self._MORE_PAGES_URL % (pagenum, channel_id) | |
737 | page = self._download_webpage(url, channel_id, | |
738 | u'Downloading page #%s' % pagenum) | |
739 | ||
740 | page = json.loads(page) | |
741 | ||
742 | ids_in_page = self.extract_videos_from_page(page['content_html']) | |
743 | video_ids.extend(ids_in_page) | |
744 | ||
745 | if self._MORE_PAGES_INDICATOR not in page['load_more_widget_html']: | |
746 | break | |
747 | ||
748 | self._downloader.to_screen(u'[youtube] Channel %s: Found %i videos' % (channel_id, len(video_ids))) | |
749 | ||
750 | urls = ['http://www.youtube.com/watch?v=%s' % id for id in video_ids] | |
751 | url_entries = [self.url_result(url, 'Youtube') for url in urls] | |
752 | return [self.playlist_result(url_entries, channel_id)] | |
753 | ||
754 | ||
755 | class YoutubeUserIE(InfoExtractor): | |
756 | """Information Extractor for YouTube users.""" | |
757 | ||
758 | _VALID_URL = r'(?:(?:(?:https?://)?(?:\w+\.)?youtube\.com/user/)|ytuser:)([A-Za-z0-9_-]+)' | |
759 | _TEMPLATE_URL = 'http://gdata.youtube.com/feeds/api/users/%s' | |
760 | _GDATA_PAGE_SIZE = 50 | |
761 | _GDATA_URL = 'http://gdata.youtube.com/feeds/api/users/%s/uploads?max-results=%d&start-index=%d' | |
762 | _VIDEO_INDICATOR = r'/watch\?v=(.+?)[\<&]' | |
763 | IE_NAME = u'youtube:user' | |
764 | ||
765 | def _real_extract(self, url): | |
766 | # Extract username | |
767 | mobj = re.match(self._VALID_URL, url) | |
768 | if mobj is None: | |
769 | raise ExtractorError(u'Invalid URL: %s' % url) | |
770 | ||
771 | username = mobj.group(1) | |
772 | ||
773 | # Download video ids using YouTube Data API. Result size per | |
774 | # query is limited (currently to 50 videos) so we need to query | |
775 | # page by page until there are no video ids - it means we got | |
776 | # all of them. | |
777 | ||
778 | video_ids = [] | |
779 | pagenum = 0 | |
780 | ||
781 | while True: | |
782 | start_index = pagenum * self._GDATA_PAGE_SIZE + 1 | |
783 | ||
784 | gdata_url = self._GDATA_URL % (username, self._GDATA_PAGE_SIZE, start_index) | |
785 | page = self._download_webpage(gdata_url, username, | |
786 | u'Downloading video ids from %d to %d' % (start_index, start_index + self._GDATA_PAGE_SIZE)) | |
787 | ||
788 | # Extract video identifiers | |
789 | ids_in_page = [] | |
790 | ||
791 | for mobj in re.finditer(self._VIDEO_INDICATOR, page): | |
792 | if mobj.group(1) not in ids_in_page: | |
793 | ids_in_page.append(mobj.group(1)) | |
794 | ||
795 | video_ids.extend(ids_in_page) | |
796 | ||
797 | # A little optimization - if current page is not | |
798 | # "full", ie. does not contain PAGE_SIZE video ids then | |
799 | # we can assume that this page is the last one - there | |
800 | # are no more ids on further pages - no need to query | |
801 | # again. | |
802 | ||
803 | if len(ids_in_page) < self._GDATA_PAGE_SIZE: | |
804 | break | |
805 | ||
806 | pagenum += 1 | |
807 | ||
808 | urls = ['http://www.youtube.com/watch?v=%s' % video_id for video_id in video_ids] | |
809 | url_results = [self.url_result(url, 'Youtube') for url in urls] | |
810 | return [self.playlist_result(url_results, playlist_title = username)] | |
b05654f0 PH |
811 | |
812 | class YoutubeSearchIE(SearchInfoExtractor): | |
813 | """Information Extractor for YouTube search queries.""" | |
814 | _API_URL = 'https://gdata.youtube.com/feeds/api/videos?q=%s&start-index=%i&max-results=50&v=2&alt=jsonc' | |
815 | _MAX_RESULTS = 1000 | |
816 | IE_NAME = u'youtube:search' | |
817 | _SEARCH_KEY = 'ytsearch' | |
818 | ||
819 | def report_download_page(self, query, pagenum): | |
820 | """Report attempt to download search page with given number.""" | |
821 | self._downloader.to_screen(u'[youtube] query "%s": Downloading page %s' % (query, pagenum)) | |
822 | ||
823 | def _get_n_results(self, query, n): | |
824 | """Get a specified number of results for a query""" | |
825 | ||
826 | video_ids = [] | |
827 | pagenum = 0 | |
828 | limit = n | |
829 | ||
830 | while (50 * pagenum) < limit: | |
831 | self.report_download_page(query, pagenum+1) | |
832 | result_url = self._API_URL % (compat_urllib_parse.quote_plus(query), (50*pagenum)+1) | |
833 | request = compat_urllib_request.Request(result_url) | |
834 | try: | |
835 | data = compat_urllib_request.urlopen(request).read().decode('utf-8') | |
836 | except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err: | |
837 | raise ExtractorError(u'Unable to download API page: %s' % compat_str(err)) | |
838 | api_response = json.loads(data)['data'] | |
839 | ||
840 | if not 'items' in api_response: | |
841 | raise ExtractorError(u'[youtube] No video results') | |
842 | ||
843 | new_ids = list(video['id'] for video in api_response['items']) | |
844 | video_ids += new_ids | |
845 | ||
846 | limit = min(n, api_response['totalItems']) | |
847 | pagenum += 1 | |
848 | ||
849 | if len(video_ids) > n: | |
850 | video_ids = video_ids[:n] | |
851 | videos = [self.url_result('http://www.youtube.com/watch?v=%s' % id, 'Youtube') for id in video_ids] | |
852 | return self.playlist_result(videos, query) |