]>
Commit | Line | Data |
---|---|---|
315ab3d5 | 1 | import hashlib |
e3e606de | 2 | import itertools |
27b1c73f | 3 | import json |
59fc531f | 4 | import re |
ab2ffab2 | 5 | import time |
59fc531f JMF |
6 | |
7 | from .common import InfoExtractor | |
3d2623a8 | 8 | from ..networking.exceptions import HTTPError |
e1ec9330 | 9 | from ..utils import ( |
238d42cf | 10 | ExtractorError, |
e3e606de PD |
11 | decode_base_n, |
12 | encode_base_n, | |
50eaea9f | 13 | filter_dict, |
cce889b9 | 14 | float_or_none, |
e3e606de | 15 | format_field, |
c4096e8a | 16 | get_element_by_attribute, |
e1ec9330 | 17 | int_or_none, |
87696e78 | 18 | lowercase_escape, |
013322a9 | 19 | str_or_none, |
4e260d1a | 20 | str_to_int, |
eb56d132 | 21 | traverse_obj, |
3052a30d | 22 | url_or_none, |
ab2ffab2 | 23 | urlencode_postdata, |
e1ec9330 | 24 | ) |
59fc531f | 25 | |
e3e606de PD |
26 | _ENCODING_CHARS = 'ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz0123456789-_' |
27 | ||
28 | ||
add96eb9 | 29 | def _pk_to_id(media_id): |
e3e606de | 30 | """Source: https://stackoverflow.com/questions/24437823/getting-instagram-post-url-from-media-id""" |
add96eb9 | 31 | return encode_base_n(int(media_id.split('_')[0]), table=_ENCODING_CHARS) |
e3e606de PD |
32 | |
33 | ||
34 | def _id_to_pk(shortcode): | |
35 | """Covert a shortcode to a numeric value""" | |
36 | return decode_base_n(shortcode[:11], table=_ENCODING_CHARS) | |
37 | ||
0de668af | 38 | |
8dcf65c9 | 39 | class InstagramBaseIE(InfoExtractor): |
ab2ffab2 | 40 | _NETRC_MACHINE = 'instagram' |
8dcf65c9 | 41 | _IS_LOGGED_IN = False |
42 | ||
7d3b98be | 43 | _API_BASE_URL = 'https://i.instagram.com/api/v1' |
44 | _LOGIN_URL = 'https://www.instagram.com/accounts/login' | |
45 | _API_HEADERS = { | |
46 | 'X-IG-App-ID': '936619743392459', | |
47 | 'X-ASBD-ID': '198387', | |
48 | 'X-IG-WWW-Claim': '0', | |
49 | 'Origin': 'https://www.instagram.com', | |
50 | 'Accept': '*/*', | |
51 | 'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/104.0.0.0 Safari/537.36', | |
52 | } | |
53 | ||
52efa4b3 | 54 | def _perform_login(self, username, password): |
55 | if self._IS_LOGGED_IN: | |
8dcf65c9 | 56 | return |
57 | ||
58 | login_webpage = self._download_webpage( | |
7d3b98be | 59 | self._LOGIN_URL, None, note='Downloading login webpage', errnote='Failed to download login webpage') |
8dcf65c9 | 60 | |
7d3b98be | 61 | shared_data = self._parse_json(self._search_regex( |
62 | r'window\._sharedData\s*=\s*({.+?});', login_webpage, 'shared data', default='{}'), None) | |
63 | ||
64 | login = self._download_json( | |
65 | f'{self._LOGIN_URL}/ajax/', None, note='Logging in', headers={ | |
66 | **self._API_HEADERS, | |
67 | 'X-Requested-With': 'XMLHttpRequest', | |
68 | 'X-CSRFToken': shared_data['config']['csrf_token'], | |
69 | 'X-Instagram-AJAX': shared_data['rollout_hash'], | |
70 | 'Referer': 'https://www.instagram.com/', | |
71 | }, data=urlencode_postdata({ | |
72 | 'enc_password': f'#PWD_INSTAGRAM_BROWSER:0:{int(time.time())}:{password}', | |
73 | 'username': username, | |
74 | 'queryParams': '{}', | |
75 | 'optIntoOneTap': 'false', | |
76 | 'stopDeletionNonce': '', | |
77 | 'trustedDeviceRecords': '{}', | |
78 | })) | |
8dcf65c9 | 79 | |
80 | if not login.get('authenticated'): | |
81 | if login.get('message'): | |
82 | raise ExtractorError(f'Unable to login: {login["message"]}') | |
d298d33f M |
83 | elif login.get('user'): |
84 | raise ExtractorError('Unable to login: Sorry, your password was incorrect. Please double-check your password.', expected=True) | |
85 | elif login.get('user') is False: | |
86 | raise ExtractorError('Unable to login: The username you entered doesn\'t belong to an account. Please check your username and try again.', expected=True) | |
8dcf65c9 | 87 | raise ExtractorError('Unable to login') |
88 | InstagramBaseIE._IS_LOGGED_IN = True | |
89 | ||
eb56d132 | 90 | def _get_count(self, media, kind, *keys): |
91 | return traverse_obj( | |
92 | media, (kind, 'count'), *((f'edge_media_{key}', 'count') for key in keys), | |
93 | expected_type=int_or_none) | |
94 | ||
95 | def _get_dimension(self, name, media, webpage=None): | |
96 | return ( | |
97 | traverse_obj(media, ('dimensions', name), expected_type=int_or_none) | |
98 | or int_or_none(self._html_search_meta( | |
99 | (f'og:video:{name}', f'video:{name}'), webpage or '', default=None))) | |
100 | ||
101 | def _extract_nodes(self, nodes, is_direct=False): | |
102 | for idx, node in enumerate(nodes, start=1): | |
103 | if node.get('__typename') != 'GraphVideo' and node.get('is_video') is not True: | |
104 | continue | |
105 | ||
106 | video_id = node.get('shortcode') | |
107 | ||
108 | if is_direct: | |
109 | info = { | |
110 | 'id': video_id or node['id'], | |
111 | 'url': node.get('video_url'), | |
112 | 'width': self._get_dimension('width', node), | |
113 | 'height': self._get_dimension('height', node), | |
114 | 'http_headers': { | |
115 | 'Referer': 'https://www.instagram.com/', | |
add96eb9 | 116 | }, |
eb56d132 | 117 | } |
118 | elif not video_id: | |
119 | continue | |
120 | else: | |
121 | info = { | |
122 | '_type': 'url', | |
123 | 'ie_key': 'Instagram', | |
124 | 'id': video_id, | |
125 | 'url': f'https://instagram.com/p/{video_id}', | |
126 | } | |
127 | ||
128 | yield { | |
129 | **info, | |
130 | 'title': node.get('title') or (f'Video {idx}' if is_direct else None), | |
131 | 'description': traverse_obj( | |
132 | node, ('edge_media_to_caption', 'edges', 0, 'node', 'text'), expected_type=str), | |
133 | 'thumbnail': traverse_obj( | |
134 | node, 'display_url', 'thumbnail_src', 'display_src', expected_type=url_or_none), | |
135 | 'duration': float_or_none(node.get('video_duration')), | |
136 | 'timestamp': int_or_none(node.get('taken_at_timestamp')), | |
137 | 'view_count': int_or_none(node.get('video_view_count')), | |
138 | 'comment_count': self._get_count(node, 'comments', 'preview_comment', 'to_comment', 'to_parent_comment'), | |
139 | 'like_count': self._get_count(node, 'likes', 'preview_like'), | |
140 | } | |
141 | ||
013322a9 | 142 | def _extract_product_media(self, product_media): |
7d3b98be | 143 | media_id = product_media.get('code') or _pk_to_id(product_media.get('pk')) |
013322a9 M |
144 | vcodec = product_media.get('video_codec') |
145 | dash_manifest_raw = product_media.get('video_dash_manifest') | |
146 | videos_list = product_media.get('video_versions') | |
147 | if not (dash_manifest_raw or videos_list): | |
c5332d7f | 148 | return {} |
013322a9 M |
149 | |
150 | formats = [{ | |
add96eb9 | 151 | 'format_id': fmt.get('id'), |
152 | 'url': fmt.get('url'), | |
153 | 'width': fmt.get('width'), | |
154 | 'height': fmt.get('height'), | |
013322a9 | 155 | 'vcodec': vcodec, |
add96eb9 | 156 | } for fmt in videos_list or []] |
013322a9 M |
157 | if dash_manifest_raw: |
158 | formats.extend(self._parse_mpd_formats(self._parse_xml(dash_manifest_raw, media_id), mpd_id='dash')) | |
013322a9 M |
159 | |
160 | thumbnails = [{ | |
161 | 'url': thumbnail.get('url'), | |
162 | 'width': thumbnail.get('width'), | |
add96eb9 | 163 | 'height': thumbnail.get('height'), |
013322a9 M |
164 | } for thumbnail in traverse_obj(product_media, ('image_versions2', 'candidates')) or []] |
165 | return { | |
166 | 'id': media_id, | |
167 | 'duration': float_or_none(product_media.get('video_duration')), | |
168 | 'formats': formats, | |
add96eb9 | 169 | 'thumbnails': thumbnails, |
013322a9 M |
170 | } |
171 | ||
172 | def _extract_product(self, product_info): | |
173 | if isinstance(product_info, list): | |
174 | product_info = product_info[0] | |
175 | ||
176 | user_info = product_info.get('user') or {} | |
177 | info_dict = { | |
2e767548 | 178 | 'id': _pk_to_id(traverse_obj(product_info, 'pk', 'id', expected_type=str_or_none)[:19]), |
013322a9 M |
179 | 'title': product_info.get('title') or f'Video by {user_info.get("username")}', |
180 | 'description': traverse_obj(product_info, ('caption', 'text'), expected_type=str_or_none), | |
181 | 'timestamp': int_or_none(product_info.get('taken_at')), | |
182 | 'channel': user_info.get('username'), | |
183 | 'uploader': user_info.get('full_name'), | |
184 | 'uploader_id': str_or_none(user_info.get('pk')), | |
185 | 'view_count': int_or_none(product_info.get('view_count')), | |
186 | 'like_count': int_or_none(product_info.get('like_count')), | |
187 | 'comment_count': int_or_none(product_info.get('comment_count')), | |
2e767548 | 188 | '__post_extractor': self.extract_comments(_pk_to_id(product_info.get('pk'))), |
013322a9 M |
189 | 'http_headers': { |
190 | 'Referer': 'https://www.instagram.com/', | |
add96eb9 | 191 | }, |
013322a9 M |
192 | } |
193 | carousel_media = product_info.get('carousel_media') | |
194 | if carousel_media: | |
195 | return { | |
196 | '_type': 'playlist', | |
197 | **info_dict, | |
198 | 'title': f'Post by {user_info.get("username")}', | |
199 | 'entries': [{ | |
200 | **info_dict, | |
201 | **self._extract_product_media(product_media), | |
202 | } for product_media in carousel_media], | |
203 | } | |
204 | ||
205 | return { | |
206 | **info_dict, | |
add96eb9 | 207 | **self._extract_product_media(product_info), |
013322a9 M |
208 | } |
209 | ||
2e767548 PD |
210 | def _get_comments(self, video_id): |
211 | comments_info = self._download_json( | |
212 | f'{self._API_BASE_URL}/media/{_id_to_pk(video_id)}/comments/?can_support_threading=true&permalink_enabled=false', video_id, | |
213 | fatal=False, errnote='Comments extraction failed', note='Downloading comments info', headers=self._API_HEADERS) or {} | |
214 | ||
215 | comment_data = traverse_obj(comments_info, ('edge_media_to_parent_comment', 'edges'), 'comments') | |
216 | for comment_dict in comment_data or []: | |
217 | yield { | |
218 | 'author': traverse_obj(comment_dict, ('node', 'owner', 'username'), ('user', 'username')), | |
219 | 'author_id': traverse_obj(comment_dict, ('node', 'owner', 'id'), ('user', 'pk')), | |
220 | 'author_thumbnail': traverse_obj(comment_dict, ('node', 'owner', 'profile_pic_url'), ('user', 'profile_pic_url'), expected_type=url_or_none), | |
221 | 'id': traverse_obj(comment_dict, ('node', 'id'), 'pk'), | |
222 | 'text': traverse_obj(comment_dict, ('node', 'text'), 'text'), | |
223 | 'like_count': traverse_obj(comment_dict, ('node', 'edge_liked_by', 'count'), 'comment_like_count', expected_type=int_or_none), | |
224 | 'timestamp': traverse_obj(comment_dict, ('node', 'created_at'), 'created_at', expected_type=int_or_none), | |
225 | } | |
226 | ||
8dcf65c9 | 227 | |
fb2d1ee6 | 228 | class InstagramIOSIE(InfoExtractor): |
c586f9e8 | 229 | IE_DESC = 'IOS instagram:// URL' |
fb2d1ee6 | 230 | _VALID_URL = r'instagram://media\?id=(?P<id>[\d_]+)' |
231 | _TESTS = [{ | |
232 | 'url': 'instagram://media?id=482584233761418119', | |
233 | 'md5': '0d2da106a9d2631273e192b372806516', | |
234 | 'info_dict': { | |
235 | 'id': 'aye83DjauH', | |
236 | 'ext': 'mp4', | |
237 | 'title': 'Video by naomipq', | |
238 | 'description': 'md5:1f17f0ab29bd6fe2bfad705f58de3cb8', | |
239 | 'thumbnail': r're:^https?://.*\.jpg', | |
240 | 'duration': 0, | |
241 | 'timestamp': 1371748545, | |
242 | 'upload_date': '20130620', | |
243 | 'uploader_id': 'naomipq', | |
244 | 'uploader': 'B E A U T Y F O R A S H E S', | |
245 | 'like_count': int, | |
246 | 'comment_count': int, | |
247 | 'comments': list, | |
248 | }, | |
add96eb9 | 249 | 'add_ie': ['Instagram'], |
fb2d1ee6 | 250 | }] |
251 | ||
fb2d1ee6 | 252 | def _real_extract(self, url): |
e3e606de PD |
253 | video_id = _pk_to_id(self._match_id(url)) |
254 | return self.url_result(f'http://instagram.com/tv/{video_id}', InstagramIE, video_id) | |
fb2d1ee6 | 255 | |
256 | ||
8dcf65c9 | 257 | class InstagramIE(InstagramBaseIE): |
06cb0638 | 258 | _VALID_URL = r'(?P<url>https?://(?:www\.)?instagram\.com(?:/[^/]+)?/(?:p|tv|reels?(?!/audio/))/(?P<id>[^/?#&]+))' |
bfd973ec | 259 | _EMBED_REGEX = [r'<iframe[^>]+src=(["\'])(?P<url>(?:https?:)?//(?:www\.)?instagram\.com/p/[^/]+/embed.*?)\1'] |
4479600d | 260 | _TESTS = [{ |
fc6e75dd | 261 | 'url': 'https://instagram.com/p/aye83DjauH/?foo=bar#abc', |
0de668af JMF |
262 | 'md5': '0d2da106a9d2631273e192b372806516', |
263 | 'info_dict': { | |
264 | 'id': 'aye83DjauH', | |
265 | 'ext': 'mp4', | |
0de668af JMF |
266 | 'title': 'Video by naomipq', |
267 | 'description': 'md5:1f17f0ab29bd6fe2bfad705f58de3cb8', | |
ec85ded8 | 268 | 'thumbnail': r're:^https?://.*\.jpg', |
2e767548 | 269 | 'duration': 8.747, |
98960c91 S |
270 | 'timestamp': 1371748545, |
271 | 'upload_date': '20130620', | |
013322a9 | 272 | 'uploader_id': '2815873', |
29f7c58a | 273 | 'uploader': 'B E A U T Y F O R A S H E S', |
013322a9 | 274 | 'channel': 'naomipq', |
98960c91 S |
275 | 'like_count': int, |
276 | 'comment_count': int, | |
a56e74e2 | 277 | 'comments': list, |
98960c91 | 278 | }, |
2e767548 PD |
279 | 'expected_warnings': [ |
280 | 'General metadata extraction failed', | |
281 | 'Main webpage is locked behind the login page', | |
282 | ], | |
fb4b3458 | 283 | }, { |
2e767548 PD |
284 | # reel |
285 | 'url': 'https://www.instagram.com/reel/Chunk8-jurw/', | |
286 | 'md5': 'f6d8277f74515fa3ff9f5791426e42b1', | |
fb4b3458 | 287 | 'info_dict': { |
2e767548 | 288 | 'id': 'Chunk8-jurw', |
fb4b3458 | 289 | 'ext': 'mp4', |
2e767548 PD |
290 | 'title': 'Video by instagram', |
291 | 'description': 'md5:c9cde483606ed6f80fbe9283a6a2b290', | |
ec85ded8 | 292 | 'thumbnail': r're:^https?://.*\.jpg', |
2e767548 PD |
293 | 'duration': 5.016, |
294 | 'timestamp': 1661529231, | |
295 | 'upload_date': '20220826', | |
296 | 'uploader_id': '25025320', | |
297 | 'uploader': 'Instagram', | |
298 | 'channel': 'instagram', | |
98960c91 S |
299 | 'like_count': int, |
300 | 'comment_count': int, | |
a56e74e2 | 301 | 'comments': list, |
fb4b3458 | 302 | }, |
2e767548 PD |
303 | 'expected_warnings': [ |
304 | 'General metadata extraction failed', | |
305 | 'Main webpage is locked behind the login page', | |
306 | ], | |
ada77fa5 S |
307 | }, { |
308 | # multi video post | |
309 | 'url': 'https://www.instagram.com/p/BQ0eAlwhDrw/', | |
310 | 'playlist': [{ | |
311 | 'info_dict': { | |
312 | 'id': 'BQ0dSaohpPW', | |
313 | 'ext': 'mp4', | |
314 | 'title': 'Video 1', | |
2e767548 PD |
315 | 'thumbnail': r're:^https?://.*\.jpg', |
316 | 'view_count': int, | |
ada77fa5 S |
317 | }, |
318 | }, { | |
319 | 'info_dict': { | |
320 | 'id': 'BQ0dTpOhuHT', | |
321 | 'ext': 'mp4', | |
322 | 'title': 'Video 2', | |
2e767548 PD |
323 | 'thumbnail': r're:^https?://.*\.jpg', |
324 | 'view_count': int, | |
ada77fa5 S |
325 | }, |
326 | }, { | |
327 | 'info_dict': { | |
328 | 'id': 'BQ0dT7RBFeF', | |
329 | 'ext': 'mp4', | |
330 | 'title': 'Video 3', | |
2e767548 PD |
331 | 'thumbnail': r're:^https?://.*\.jpg', |
332 | 'view_count': int, | |
ada77fa5 S |
333 | }, |
334 | }], | |
335 | 'info_dict': { | |
336 | 'id': 'BQ0eAlwhDrw', | |
337 | 'title': 'Post by instagram', | |
338 | 'description': 'md5:0f9203fc6a2ce4d228da5754bcf54957', | |
339 | }, | |
2e767548 PD |
340 | 'expected_warnings': [ |
341 | 'General metadata extraction failed', | |
342 | 'Main webpage is locked behind the login page', | |
343 | ], | |
cce889b9 | 344 | }, { |
345 | # IGTV | |
346 | 'url': 'https://www.instagram.com/tv/BkfuX9UB-eK/', | |
347 | 'info_dict': { | |
348 | 'id': 'BkfuX9UB-eK', | |
349 | 'ext': 'mp4', | |
350 | 'title': 'Fingerboarding Tricks with @cass.fb', | |
351 | 'thumbnail': r're:^https?://.*\.jpg', | |
352 | 'duration': 53.83, | |
353 | 'timestamp': 1530032919, | |
354 | 'upload_date': '20180626', | |
013322a9 | 355 | 'uploader_id': '25025320', |
cce889b9 | 356 | 'uploader': 'Instagram', |
013322a9 | 357 | 'channel': 'instagram', |
cce889b9 | 358 | 'like_count': int, |
359 | 'comment_count': int, | |
360 | 'comments': list, | |
361 | 'description': 'Meet Cass Hirst (@cass.fb), a fingerboarding pro who can perform tiny ollies and kickflips while blindfolded.', | |
2e767548 PD |
362 | }, |
363 | 'expected_warnings': [ | |
364 | 'General metadata extraction failed', | |
365 | 'Main webpage is locked behind the login page', | |
366 | ], | |
4479600d S |
367 | }, { |
368 | 'url': 'https://instagram.com/p/-Cmh1cukG2/', | |
369 | 'only_matching': True, | |
0dafea02 S |
370 | }, { |
371 | 'url': 'http://instagram.com/p/9o6LshA7zy/embed/', | |
372 | 'only_matching': True, | |
edb2820c RA |
373 | }, { |
374 | 'url': 'https://www.instagram.com/tv/aye83DjauH/', | |
375 | 'only_matching': True, | |
29f7c58a | 376 | }, { |
377 | 'url': 'https://www.instagram.com/reel/CDUMkliABpa/', | |
378 | 'only_matching': True, | |
8b688881 | 379 | }, { |
380 | 'url': 'https://www.instagram.com/marvelskies.fc/reel/CWqAgUZgCku/', | |
381 | 'only_matching': True, | |
06cb0638 AP |
382 | }, { |
383 | 'url': 'https://www.instagram.com/reels/Cop84x6u7CP/', | |
384 | 'only_matching': True, | |
4479600d | 385 | }] |
59fc531f | 386 | |
bfd973ec | 387 | @classmethod |
388 | def _extract_embed_urls(cls, url, webpage): | |
389 | res = tuple(super()._extract_embed_urls(url, webpage)) | |
390 | if res: | |
391 | return res | |
c4096e8a | 392 | |
bfd973ec | 393 | mobj = re.search(r'<a[^>]+href=([\'"])(?P<link>[^\'"]+)\1', |
394 | get_element_by_attribute('class', 'instagram-media', webpage) or '') | |
c4096e8a | 395 | if mobj: |
bfd973ec | 396 | return [mobj.group('link')] |
c4096e8a | 397 | |
59fc531f | 398 | def _real_extract(self, url): |
eb56d132 | 399 | video_id, url = self._match_valid_url(url).group('id', 'url') |
7d3b98be | 400 | media, webpage = {}, '' |
401 | ||
2e767548 PD |
402 | if self._get_cookies(url).get('sessionid'): |
403 | info = traverse_obj(self._download_json( | |
404 | f'{self._API_BASE_URL}/media/{_id_to_pk(video_id)}/info/', video_id, | |
405 | fatal=False, errnote='Video info extraction failed', | |
406 | note='Downloading video info', headers=self._API_HEADERS), ('items', 0)) | |
407 | if info: | |
408 | media.update(info) | |
409 | return self._extract_product(media) | |
410 | ||
7d3b98be | 411 | api_check = self._download_json( |
412 | f'{self._API_BASE_URL}/web/get_ruling_for_content/?content_type=MEDIA&target_id={_id_to_pk(video_id)}', | |
413 | video_id, headers=self._API_HEADERS, fatal=False, note='Setting up session', errnote=False) or {} | |
414 | csrf_token = self._get_cookies('https://www.instagram.com').get('csrftoken') | |
415 | ||
416 | if not csrf_token: | |
417 | self.report_warning('No csrf token set by Instagram API', video_id) | |
7d3b98be | 418 | else: |
2e767548 PD |
419 | csrf_token = csrf_token.value if api_check.get('status') == 'ok' else None |
420 | if not csrf_token: | |
421 | self.report_warning('Instagram API is not granting access', video_id) | |
422 | ||
423 | variables = { | |
424 | 'shortcode': video_id, | |
425 | 'child_comment_count': 3, | |
426 | 'fetch_comment_count': 40, | |
427 | 'parent_comment_count': 24, | |
428 | 'has_threaded_comments': True, | |
429 | } | |
430 | general_info = self._download_json( | |
431 | 'https://www.instagram.com/graphql/query/', video_id, fatal=False, errnote=False, | |
432 | headers={ | |
433 | **self._API_HEADERS, | |
434 | 'X-CSRFToken': csrf_token or '', | |
435 | 'X-Requested-With': 'XMLHttpRequest', | |
436 | 'Referer': url, | |
437 | }, query={ | |
438 | 'query_hash': '9f8827793ef34641b2fb195d4d41151c', | |
439 | 'variables': json.dumps(variables, separators=(',', ':')), | |
440 | }) | |
441 | media.update(traverse_obj(general_info, ('data', 'shortcode_media')) or {}) | |
442 | ||
443 | if not general_info: | |
7d3b98be | 444 | self.report_warning('General metadata extraction failed (some metadata might be missing).', video_id) |
445 | webpage, urlh = self._download_webpage_handle(url, video_id) | |
446 | shared_data = self._search_json( | |
8a3da4c6 | 447 | r'window\._sharedData\s*=', webpage, 'shared data', video_id, fatal=False) or {} |
7d3b98be | 448 | |
3d2623a8 | 449 | if shared_data and self._LOGIN_URL not in urlh.url: |
7d3b98be | 450 | media.update(traverse_obj( |
451 | shared_data, ('entry_data', 'PostPage', 0, 'graphql', 'shortcode_media'), | |
452 | ('entry_data', 'PostPage', 0, 'media'), expected_type=dict) or {}) | |
453 | else: | |
2e767548 | 454 | self.report_warning('Main webpage is locked behind the login page. Retrying with embed webpage (some metadata might be missing).') |
7d3b98be | 455 | webpage = self._download_webpage( |
6aaf96a3 | 456 | f'{url}/embed/', video_id, note='Downloading embed webpage', fatal=False) or '' |
7d3b98be | 457 | additional_data = self._search_json( |
304ad45a | 458 | r'window\.__additionalDataLoaded\s*\(\s*[^,]+,', webpage, 'additional data', video_id, fatal=False) |
2e767548 | 459 | if not additional_data and not media: |
8a3da4c6 | 460 | self.raise_login_required('Requested content is not available, rate-limit reached or login required') |
7d3b98be | 461 | |
462 | product_item = traverse_obj(additional_data, ('items', 0), expected_type=dict) | |
463 | if product_item: | |
464 | media.update(product_item) | |
465 | return self._extract_product(media) | |
466 | ||
467 | media.update(traverse_obj( | |
468 | additional_data, ('graphql', 'shortcode_media'), 'shortcode_media', expected_type=dict) or {}) | |
eb56d132 | 469 | |
013322a9 M |
470 | username = traverse_obj(media, ('owner', 'username')) or self._search_regex( |
471 | r'"owner"\s*:\s*{\s*"username"\s*:\s*"(.+?)"', webpage, 'username', fatal=False) | |
eb56d132 | 472 | |
473 | description = ( | |
474 | traverse_obj(media, ('edge_media_to_caption', 'edges', 0, 'node', 'text'), expected_type=str) | |
475 | or media.get('caption')) | |
476 | if not description: | |
477 | description = self._search_regex( | |
478 | r'"caption"\s*:\s*"(.+?)"', webpage, 'description', default=None) | |
479 | if description is not None: | |
480 | description = lowercase_escape(description) | |
98960c91 | 481 | |
eb56d132 | 482 | video_url = media.get('video_url') |
98960c91 | 483 | if not video_url: |
eb56d132 | 484 | nodes = traverse_obj(media, ('edge_sidecar_to_children', 'edges', ..., 'node'), expected_type=dict) or [] |
485 | if nodes: | |
486 | return self.playlist_result( | |
487 | self._extract_nodes(nodes, True), video_id, | |
a70635b8 | 488 | format_field(username, None, 'Post by %s'), description) |
eb56d132 | 489 | |
98960c91 S |
490 | video_url = self._og_search_video_url(webpage, secure=False) |
491 | ||
16097822 DR |
492 | formats = [{ |
493 | 'url': video_url, | |
eb56d132 | 494 | 'width': self._get_dimension('width', media, webpage), |
495 | 'height': self._get_dimension('height', media, webpage), | |
16097822 | 496 | }] |
eb56d132 | 497 | dash = traverse_obj(media, ('dash_info', 'video_dash_manifest')) |
cd9ea410 | 498 | if dash: |
499 | formats.extend(self._parse_mpd_formats(self._parse_xml(dash, video_id), mpd_id='dash')) | |
16097822 | 500 | |
4e260d1a | 501 | comment_data = traverse_obj(media, ('edge_media_to_parent_comment', 'edges')) |
eb56d132 | 502 | comments = [{ |
503 | 'author': traverse_obj(comment_dict, ('node', 'owner', 'username')), | |
504 | 'author_id': traverse_obj(comment_dict, ('node', 'owner', 'id')), | |
505 | 'id': traverse_obj(comment_dict, ('node', 'id')), | |
506 | 'text': traverse_obj(comment_dict, ('node', 'text')), | |
507 | 'timestamp': traverse_obj(comment_dict, ('node', 'created_at'), expected_type=int_or_none), | |
4e260d1a | 508 | } for comment_dict in comment_data] if comment_data else None |
eb56d132 | 509 | |
510 | display_resources = ( | |
511 | media.get('display_resources') | |
512 | or [{'src': media.get(key)} for key in ('display_src', 'display_url')] | |
513 | or [{'src': self._og_search_thumbnail(webpage)}]) | |
514 | thumbnails = [{ | |
515 | 'url': thumbnail['src'], | |
516 | 'width': thumbnail.get('config_width'), | |
517 | 'height': thumbnail.get('config_height'), | |
518 | } for thumbnail in display_resources if thumbnail.get('src')] | |
59fc531f | 519 | |
0de668af JMF |
520 | return { |
521 | 'id': video_id, | |
16097822 | 522 | 'formats': formats, |
add96eb9 | 523 | 'title': media.get('title') or f'Video by {username}', |
98960c91 | 524 | 'description': description, |
eb56d132 | 525 | 'duration': float_or_none(media.get('video_duration')), |
526 | 'timestamp': traverse_obj(media, 'taken_at_timestamp', 'date', expected_type=int_or_none), | |
013322a9 | 527 | 'uploader_id': traverse_obj(media, ('owner', 'id')), |
eb56d132 | 528 | 'uploader': traverse_obj(media, ('owner', 'full_name')), |
013322a9 | 529 | 'channel': username, |
4e260d1a M |
530 | 'like_count': self._get_count(media, 'likes', 'preview_like') or str_to_int(self._search_regex( |
531 | r'data-log-event="likeCountClick"[^>]*>[^\d]*([\d,\.]+)', webpage, 'like count', fatal=False)), | |
eb56d132 | 532 | 'comment_count': self._get_count(media, 'comments', 'preview_comment', 'to_comment', 'to_parent_comment'), |
a56e74e2 | 533 | 'comments': comments, |
eb56d132 | 534 | 'thumbnails': thumbnails, |
3dd39c5f S |
535 | 'http_headers': { |
536 | 'Referer': 'https://www.instagram.com/', | |
add96eb9 | 537 | }, |
0de668af | 538 | } |
ea38e55f PH |
539 | |
540 | ||
8dcf65c9 | 541 | class InstagramPlaylistBaseIE(InstagramBaseIE): |
31fbedc0 | 542 | _gis_tmpl = None # used to cache GIS request type |
ea38e55f | 543 | |
31fbedc0 | 544 | def _parse_graphql(self, webpage, item_id): |
545 | # Reads a webpage and returns its GraphQL data. | |
546 | return self._parse_json( | |
547 | self._search_regex( | |
548 | r'sharedData\s*=\s*({.+?})\s*;\s*[<\n]', webpage, 'data'), | |
549 | item_id) | |
238d42cf | 550 | |
31fbedc0 | 551 | def _extract_graphql(self, data, url): |
552 | # Parses GraphQL queries containing videos and generates a playlist. | |
31fbedc0 | 553 | uploader_id = self._match_id(url) |
dd9aea8c S |
554 | csrf_token = data['config']['csrf_token'] |
555 | rhx_gis = data.get('rhx_gis') or '3c7ca9dcefcf966d11dacf1f151335e8' | |
556 | ||
cba5d1b6 S |
557 | cursor = '' |
558 | for page_num in itertools.count(1): | |
31fbedc0 | 559 | variables = { |
9b3036bd | 560 | 'first': 12, |
dd9aea8c | 561 | 'after': cursor, |
31fbedc0 | 562 | } |
563 | variables.update(self._query_vars_for(data)) | |
564 | variables = json.dumps(variables) | |
238d42cf S |
565 | |
566 | if self._gis_tmpl: | |
567 | gis_tmpls = [self._gis_tmpl] | |
568 | else: | |
569 | gis_tmpls = [ | |
add96eb9 | 570 | f'{rhx_gis}', |
238d42cf | 571 | '', |
add96eb9 | 572 | f'{rhx_gis}:{csrf_token}', |
573 | '{}:{}:{}'.format(rhx_gis, csrf_token, self.get_param('http_headers')['User-Agent']), | |
238d42cf S |
574 | ] |
575 | ||
31fbedc0 | 576 | # try all of the ways to generate a GIS query, and not only use the |
577 | # first one that works, but cache it for future requests | |
238d42cf S |
578 | for gis_tmpl in gis_tmpls: |
579 | try: | |
31fbedc0 | 580 | json_data = self._download_json( |
238d42cf | 581 | 'https://www.instagram.com/graphql/query/', uploader_id, |
add96eb9 | 582 | f'Downloading JSON page {page_num}', headers={ |
238d42cf S |
583 | 'X-Requested-With': 'XMLHttpRequest', |
584 | 'X-Instagram-GIS': hashlib.md5( | |
add96eb9 | 585 | (f'{gis_tmpl}:{variables}').encode()).hexdigest(), |
238d42cf | 586 | }, query={ |
31fbedc0 | 587 | 'query_hash': self._QUERY_HASH, |
238d42cf | 588 | 'variables': variables, |
31fbedc0 | 589 | }) |
590 | media = self._parse_timeline_from(json_data) | |
238d42cf S |
591 | self._gis_tmpl = gis_tmpl |
592 | break | |
593 | except ExtractorError as e: | |
31fbedc0 | 594 | # if it's an error caused by a bad query, and there are |
595 | # more GIS templates to try, ignore it and keep trying | |
3d2623a8 | 596 | if isinstance(e.cause, HTTPError) and e.cause.status == 403: |
238d42cf S |
597 | if gis_tmpl != gis_tmpls[-1]: |
598 | continue | |
599 | raise | |
cba5d1b6 | 600 | |
eb56d132 | 601 | nodes = traverse_obj(media, ('edges', ..., 'node'), expected_type=dict) or [] |
602 | if not nodes: | |
cba5d1b6 | 603 | break |
eb56d132 | 604 | yield from self._extract_nodes(nodes) |
cba5d1b6 | 605 | |
eb56d132 | 606 | has_next_page = traverse_obj(media, ('page_info', 'has_next_page')) |
607 | cursor = traverse_obj(media, ('page_info', 'end_cursor'), expected_type=str) | |
608 | if not has_next_page or not cursor: | |
cba5d1b6 | 609 | break |
5fc12b95 S |
610 | |
611 | def _real_extract(self, url): | |
31fbedc0 | 612 | user_or_tag = self._match_id(url) |
613 | webpage = self._download_webpage(url, user_or_tag) | |
614 | data = self._parse_graphql(webpage, user_or_tag) | |
dd9aea8c | 615 | |
31fbedc0 | 616 | self._set_cookie('instagram.com', 'ig_pr', '1') |
dd9aea8c | 617 | |
5fc12b95 | 618 | return self.playlist_result( |
31fbedc0 | 619 | self._extract_graphql(data, url), user_or_tag, user_or_tag) |
620 | ||
621 | ||
8dcf65c9 | 622 | class InstagramUserIE(InstagramPlaylistBaseIE): |
df773c3d | 623 | _WORKING = False |
31fbedc0 | 624 | _VALID_URL = r'https?://(?:www\.)?instagram\.com/(?P<id>[^/]{2,})/?(?:$|[?#])' |
625 | IE_DESC = 'Instagram user profile' | |
626 | IE_NAME = 'instagram:user' | |
8dcf65c9 | 627 | _TESTS = [{ |
31fbedc0 | 628 | 'url': 'https://instagram.com/porsche', |
629 | 'info_dict': { | |
630 | 'id': 'porsche', | |
631 | 'title': 'porsche', | |
632 | }, | |
633 | 'playlist_count': 5, | |
634 | 'params': { | |
635 | 'extract_flat': True, | |
636 | 'skip_download': True, | |
637 | 'playlistend': 5, | |
add96eb9 | 638 | }, |
8dcf65c9 | 639 | }] |
31fbedc0 | 640 | |
add96eb9 | 641 | _QUERY_HASH = ('42323d64886122307be10013ad2dcc44',) |
31fbedc0 | 642 | |
643 | @staticmethod | |
644 | def _parse_timeline_from(data): | |
645 | # extracts the media timeline data from a GraphQL result | |
646 | return data['data']['user']['edge_owner_to_timeline_media'] | |
647 | ||
648 | @staticmethod | |
649 | def _query_vars_for(data): | |
650 | # returns a dictionary of variables to add to the timeline query based | |
651 | # on the GraphQL of the original page | |
652 | return { | |
add96eb9 | 653 | 'id': data['entry_data']['ProfilePage'][0]['graphql']['user']['id'], |
31fbedc0 | 654 | } |
655 | ||
656 | ||
8dcf65c9 | 657 | class InstagramTagIE(InstagramPlaylistBaseIE): |
31fbedc0 | 658 | _VALID_URL = r'https?://(?:www\.)?instagram\.com/explore/tags/(?P<id>[^/]+)' |
f304da8a | 659 | IE_DESC = 'Instagram hashtag search URLs' |
31fbedc0 | 660 | IE_NAME = 'instagram:tag' |
8dcf65c9 | 661 | _TESTS = [{ |
31fbedc0 | 662 | 'url': 'https://instagram.com/explore/tags/lolcats', |
663 | 'info_dict': { | |
664 | 'id': 'lolcats', | |
665 | 'title': 'lolcats', | |
666 | }, | |
667 | 'playlist_count': 50, | |
668 | 'params': { | |
669 | 'extract_flat': True, | |
670 | 'skip_download': True, | |
671 | 'playlistend': 50, | |
add96eb9 | 672 | }, |
8dcf65c9 | 673 | }] |
31fbedc0 | 674 | |
add96eb9 | 675 | _QUERY_HASH = ('f92f56d47dc7a55b606908374b43a314',) |
31fbedc0 | 676 | |
677 | @staticmethod | |
678 | def _parse_timeline_from(data): | |
679 | # extracts the media timeline data from a GraphQL result | |
680 | return data['data']['hashtag']['edge_hashtag_to_media'] | |
681 | ||
682 | @staticmethod | |
683 | def _query_vars_for(data): | |
684 | # returns a dictionary of variables to add to the timeline query based | |
685 | # on the GraphQL of the original page | |
686 | return { | |
687 | 'tag_name': | |
add96eb9 | 688 | data['entry_data']['TagPage'][0]['graphql']['hashtag']['name'], |
31fbedc0 | 689 | } |
dd5e60b1 | 690 | |
691 | ||
692 | class InstagramStoryIE(InstagramBaseIE): | |
693 | _VALID_URL = r'https?://(?:www\.)?instagram\.com/stories/(?P<user>[^/]+)/(?P<id>\d+)' | |
694 | IE_NAME = 'instagram:story' | |
695 | ||
696 | _TESTS = [{ | |
697 | 'url': 'https://www.instagram.com/stories/highlights/18090946048123978/', | |
698 | 'info_dict': { | |
699 | 'id': '18090946048123978', | |
700 | 'title': 'Rare', | |
701 | }, | |
add96eb9 | 702 | 'playlist_mincount': 50, |
dd5e60b1 | 703 | }] |
704 | ||
705 | def _real_extract(self, url): | |
706 | username, story_id = self._match_valid_url(url).groups() | |
e3e606de PD |
707 | story_info = self._download_webpage(url, story_id) |
708 | user_info = self._search_json(r'"user":', story_info, 'user info', story_id, fatal=False) | |
709 | if not user_info: | |
710 | self.raise_login_required('This content is unreachable') | |
dd5e60b1 | 711 | |
50eaea9f | 712 | user_id = traverse_obj(user_info, 'pk', 'id', expected_type=str) |
dd5e60b1 | 713 | story_info_url = user_id if username != 'highlights' else f'highlight:{story_id}' |
50eaea9f | 714 | if not story_info_url: # user id is only mandatory for non-highlights |
715 | raise ExtractorError('Unable to extract user id') | |
716 | ||
e3e606de | 717 | videos = traverse_obj(self._download_json( |
7d3b98be | 718 | f'{self._API_BASE_URL}/feed/reels_media/?reel_ids={story_info_url}', |
719 | story_id, errnote=False, fatal=False, headers=self._API_HEADERS), 'reels') | |
e3e606de PD |
720 | if not videos: |
721 | self.raise_login_required('You need to log in to access this content') | |
f7085283 | 722 | |
50eaea9f | 723 | full_name = traverse_obj(videos, (f'highlight:{story_id}', 'user', 'full_name'), (user_id, 'user', 'full_name')) |
e3e606de PD |
724 | story_title = traverse_obj(videos, (f'highlight:{story_id}', 'title')) |
725 | if not story_title: | |
726 | story_title = f'Story by {username}' | |
f7085283 | 727 | |
50eaea9f | 728 | highlights = traverse_obj(videos, (f'highlight:{story_id}', 'items'), (user_id, 'items')) |
e3e606de PD |
729 | info_data = [] |
730 | for highlight in highlights: | |
731 | highlight_data = self._extract_product(highlight) | |
732 | if highlight_data.get('formats'): | |
733 | info_data.append({ | |
e3e606de PD |
734 | 'uploader': full_name, |
735 | 'uploader_id': user_id, | |
50eaea9f | 736 | **filter_dict(highlight_data), |
e3e606de PD |
737 | }) |
738 | return self.playlist_result(info_data, playlist_id=story_id, playlist_title=story_title) |