[fragments] Pad fragments before decrypting (#1298)

[yt-dlp.git] / yt_dlp / extractor / instagram.py

from __future__ import unicode_literals

import itertools
import hashlib
import json
import re

from .common import InfoExtractor
from ..compat import (
    compat_str,
    compat_HTTPError,
)
from ..utils import (
    ExtractorError,
    float_or_none,
    get_element_by_attribute,
    int_or_none,
    lowercase_escape,
    std_headers,
    try_get,
    url_or_none,
    variadic,
)


class InstagramIE(InfoExtractor):
    _VALID_URL = r'(?P<url>https?://(?:www\.)?instagram\.com/(?:p|tv|reel)/(?P<id>[^/?#&]+))'
    _TESTS = [{
        'url': 'https://instagram.com/p/aye83DjauH/?foo=bar#abc',
        'md5': '0d2da106a9d2631273e192b372806516',
        'info_dict': {
            'id': 'aye83DjauH',
            'ext': 'mp4',
            'title': 'Video by naomipq',
            'description': 'md5:1f17f0ab29bd6fe2bfad705f58de3cb8',
            'thumbnail': r're:^https?://.*\.jpg',
            'duration': 0,
            'timestamp': 1371748545,
            'upload_date': '20130620',
            'uploader_id': 'naomipq',
            'uploader': 'B E A U T Y  F O R  A S H E S',
            'like_count': int,
            'comment_count': int,
            'comments': list,
        },
    }, {
        # missing description
        'url': 'https://www.instagram.com/p/BA-pQFBG8HZ/?taken-by=britneyspears',
        'info_dict': {
            'id': 'BA-pQFBG8HZ',
            'ext': 'mp4',
            'title': 'Video by britneyspears',
            'thumbnail': r're:^https?://.*\.jpg',
            'duration': 0,
            'timestamp': 1453760977,
            'upload_date': '20160125',
            'uploader_id': 'britneyspears',
            'uploader': 'Britney Spears',
            'like_count': int,
            'comment_count': int,
            'comments': list,
        },
        'params': {
            'skip_download': True,
        },
    }, {
        # multi video post
        'url': 'https://www.instagram.com/p/BQ0eAlwhDrw/',
        'playlist': [{
            'info_dict': {
                'id': 'BQ0dSaohpPW',
                'ext': 'mp4',
                'title': 'Video 1',
            },
        }, {
            'info_dict': {
                'id': 'BQ0dTpOhuHT',
                'ext': 'mp4',
                'title': 'Video 2',
            },
        }, {
            'info_dict': {
                'id': 'BQ0dT7RBFeF',
                'ext': 'mp4',
                'title': 'Video 3',
            },
        }],
        'info_dict': {
            'id': 'BQ0eAlwhDrw',
            'title': 'Post by instagram',
            'description': 'md5:0f9203fc6a2ce4d228da5754bcf54957',
        },
    }, {
        # IGTV
        'url': 'https://www.instagram.com/tv/BkfuX9UB-eK/',
        'info_dict': {
            'id': 'BkfuX9UB-eK',
            'ext': 'mp4',
            'title': 'Fingerboarding Tricks with @cass.fb',
            'thumbnail': r're:^https?://.*\.jpg',
            'duration': 53.83,
            'timestamp': 1530032919,
            'upload_date': '20180626',
            'uploader_id': 'instagram',
            'uploader': 'Instagram',
            'like_count': int,
            'comment_count': int,
            'comments': list,
            'description': 'Meet Cass Hirst (@cass.fb), a fingerboarding pro who can perform tiny ollies and kickflips while blindfolded.',
        }
    }, {
        'url': 'https://instagram.com/p/-Cmh1cukG2/',
        'only_matching': True,
    }, {
        'url': 'http://instagram.com/p/9o6LshA7zy/embed/',
        'only_matching': True,
    }, {
        'url': 'https://www.instagram.com/tv/aye83DjauH/',
        'only_matching': True,
    }, {
        'url': 'https://www.instagram.com/reel/CDUMkliABpa/',
        'only_matching': True,
    }]

    @staticmethod
    def _extract_embed_url(webpage):
        mobj = re.search(
            r'<iframe[^>]+src=(["\'])(?P<url>(?:https?:)?//(?:www\.)?instagram\.com/p/[^/]+/embed.*?)\1',
            webpage)
        if mobj:
            return mobj.group('url')

        blockquote_el = get_element_by_attribute(
            'class', 'instagram-media', webpage)
        if blockquote_el is None:
            return

        mobj = re.search(
            r'<a[^>]+href=([\'"])(?P<link>[^\'"]+)\1', blockquote_el)
        if mobj:
            return mobj.group('link')

    def _real_extract(self, url):
        mobj = self._match_valid_url(url)
        video_id = mobj.group('id')
        url = mobj.group('url')

        webpage, urlh = self._download_webpage_handle(url, video_id)
        if 'www.instagram.com/accounts/login' in urlh.geturl().rstrip('/'):
            self.raise_login_required('You need to log in to access this content', method='cookies')

        (media, video_url, description, thumbnail, timestamp, uploader,
         uploader_id, like_count, comment_count, comments, height,
         width) = [None] * 12

        shared_data = self._parse_json(
            self._search_regex(
                r'window\._sharedData\s*=\s*({.+?});',
                webpage, 'shared data', default='{}'),
            video_id, fatal=False)
        if shared_data:
            media = try_get(
                shared_data,
                (lambda x: x['entry_data']['PostPage'][0]['graphql']['shortcode_media'],
                 lambda x: x['entry_data']['PostPage'][0]['media']),
                dict)
        # _sharedData.entry_data.PostPage is empty when authenticated (see
        # https://github.com/ytdl-org/youtube-dl/pull/22880)
        if not media:
            additional_data = self._parse_json(
                self._search_regex(
                    r'window\.__additionalDataLoaded\s*\(\s*[^,]+,\s*({.+?})\s*\)\s*;',
                    webpage, 'additional data', default='{}'),
                video_id, fatal=False)
            if additional_data:
                media = try_get(
                    additional_data, lambda x: x['graphql']['shortcode_media'],
                    dict)
        if media:
            video_url = media.get('video_url')
            height = int_or_none(media.get('dimensions', {}).get('height'))
            width = int_or_none(media.get('dimensions', {}).get('width'))
            description = try_get(
                media, lambda x: x['edge_media_to_caption']['edges'][0]['node']['text'],
                compat_str) or media.get('caption')
            title = media.get('title')
            thumbnail = media.get('display_src') or media.get('display_url')
            duration = float_or_none(media.get('video_duration'))
            timestamp = int_or_none(media.get('taken_at_timestamp') or media.get('date'))
            uploader = media.get('owner', {}).get('full_name')
            uploader_id = media.get('owner', {}).get('username')

            def get_count(keys, kind):
                for key in variadic(keys):
                    count = int_or_none(try_get(
                        media, (lambda x: x['edge_media_%s' % key]['count'],
                                lambda x: x['%ss' % kind]['count'])))
                    if count is not None:
                        return count

            like_count = get_count('preview_like', 'like')
            comment_count = get_count(
                ('preview_comment', 'to_comment', 'to_parent_comment'), 'comment')

            comments = []
            for comment in try_get(media, lambda x: x['edge_media_to_parent_comment']['edges']):
                comment_dict = comment.get('node', {})
                comment_text = comment_dict.get('text')
                if comment_text:
                    comments.append({
                        'author': try_get(comment_dict, lambda x: x['owner']['username']),
                        'author_id': try_get(comment_dict, lambda x: x['owner']['id']),
                        'id': comment_dict.get('id'),
                        'text': comment_text,
                        'timestamp': int_or_none(comment_dict.get('created_at')),
                    })
            if not video_url:
                edges = try_get(
                    media, lambda x: x['edge_sidecar_to_children']['edges'],
                    list) or []
                if edges:
                    entries = []
                    for edge_num, edge in enumerate(edges, start=1):
                        node = try_get(edge, lambda x: x['node'], dict)
                        if not node:
                            continue
                        node_video_url = url_or_none(node.get('video_url'))
                        if not node_video_url:
                            continue
                        entries.append({
                            'id': node.get('shortcode') or node['id'],
                            'title': node.get('title') or 'Video %d' % edge_num,
                            'url': node_video_url,
                            'thumbnail': node.get('display_url'),
                            'duration': float_or_none(node.get('video_duration')),
                            'width': int_or_none(try_get(node, lambda x: x['dimensions']['width'])),
                            'height': int_or_none(try_get(node, lambda x: x['dimensions']['height'])),
                            'view_count': int_or_none(node.get('video_view_count')),
                        })
                    return self.playlist_result(
                        entries, video_id,
                        'Post by %s' % uploader_id if uploader_id else None,
                        description)

        if not video_url:
            video_url = self._og_search_video_url(webpage, secure=False)

        formats = [{
            'url': video_url,
            'width': width,
            'height': height,
        }]

        if not uploader_id:
            uploader_id = self._search_regex(
                r'"owner"\s*:\s*{\s*"username"\s*:\s*"(.+?)"',
                webpage, 'uploader id', fatal=False)

        if not description:
            description = self._search_regex(
                r'"caption"\s*:\s*"(.+?)"', webpage, 'description', default=None)
            if description is not None:
                description = lowercase_escape(description)

        if not thumbnail:
            thumbnail = self._og_search_thumbnail(webpage)

        return {
            'id': video_id,
            'formats': formats,
            'ext': 'mp4',
            'title': title or 'Video by %s' % uploader_id,
            'description': description,
            'duration': duration,
            'thumbnail': thumbnail,
            'timestamp': timestamp,
            'uploader_id': uploader_id,
            'uploader': uploader,
            'like_count': like_count,
            'comment_count': comment_count,
            'comments': comments,
            'http_headers': {
                'Referer': 'https://www.instagram.com/',
            }
        }


class InstagramPlaylistIE(InfoExtractor):
    # A superclass for handling any kind of query based on GraphQL which
    # results in a playlist.

    _gis_tmpl = None  # used to cache GIS request type

    def _parse_graphql(self, webpage, item_id):
        # Reads a webpage and returns its GraphQL data.
        return self._parse_json(
            self._search_regex(
                r'sharedData\s*=\s*({.+?})\s*;\s*[<\n]', webpage, 'data'),
            item_id)

    def _extract_graphql(self, data, url):
        # Parses GraphQL queries containing videos and generates a playlist.
        def get_count(suffix):
            return int_or_none(try_get(
                node, lambda x: x['edge_media_' + suffix]['count']))

        uploader_id = self._match_id(url)
        csrf_token = data['config']['csrf_token']
        rhx_gis = data.get('rhx_gis') or '3c7ca9dcefcf966d11dacf1f151335e8'

        cursor = ''
        for page_num in itertools.count(1):
            variables = {
                'first': 12,
                'after': cursor,
            }
            variables.update(self._query_vars_for(data))
            variables = json.dumps(variables)

            if self._gis_tmpl:
                gis_tmpls = [self._gis_tmpl]
            else:
                gis_tmpls = [
                    '%s' % rhx_gis,
                    '',
                    '%s:%s' % (rhx_gis, csrf_token),
                    '%s:%s:%s' % (rhx_gis, csrf_token, std_headers['User-Agent']),
                ]

            # try all of the ways to generate a GIS query, and not only use the
            # first one that works, but cache it for future requests
            for gis_tmpl in gis_tmpls:
                try:
                    json_data = self._download_json(
                        'https://www.instagram.com/graphql/query/', uploader_id,
                        'Downloading JSON page %d' % page_num, headers={
                            'X-Requested-With': 'XMLHttpRequest',
                            'X-Instagram-GIS': hashlib.md5(
                                ('%s:%s' % (gis_tmpl, variables)).encode('utf-8')).hexdigest(),
                        }, query={
                            'query_hash': self._QUERY_HASH,
                            'variables': variables,
                        })
                    media = self._parse_timeline_from(json_data)
                    self._gis_tmpl = gis_tmpl
                    break
                except ExtractorError as e:
                    # if it's an error caused by a bad query, and there are
                    # more GIS templates to try, ignore it and keep trying
                    if isinstance(e.cause, compat_HTTPError) and e.cause.code == 403:
                        if gis_tmpl != gis_tmpls[-1]:
                            continue
                    raise

            edges = media.get('edges')
            if not edges or not isinstance(edges, list):
                break

            for edge in edges:
                node = edge.get('node')
                if not node or not isinstance(node, dict):
                    continue
                if node.get('__typename') != 'GraphVideo' and node.get('is_video') is not True:
                    continue
                video_id = node.get('shortcode')
                if not video_id:
                    continue

                info = self.url_result(
                    'https://instagram.com/p/%s/' % video_id,
                    ie=InstagramIE.ie_key(), video_id=video_id)

                description = try_get(
                    node, lambda x: x['edge_media_to_caption']['edges'][0]['node']['text'],
                    compat_str)
                thumbnail = node.get('thumbnail_src') or node.get('display_src')
                timestamp = int_or_none(node.get('taken_at_timestamp'))

                comment_count = get_count('to_comment')
                like_count = get_count('preview_like')
                view_count = int_or_none(node.get('video_view_count'))

                info.update({
                    'description': description,
                    'thumbnail': thumbnail,
                    'timestamp': timestamp,
                    'comment_count': comment_count,
                    'like_count': like_count,
                    'view_count': view_count,
                })

                yield info

            page_info = media.get('page_info')
            if not page_info or not isinstance(page_info, dict):
                break

            has_next_page = page_info.get('has_next_page')
            if not has_next_page:
                break

            cursor = page_info.get('end_cursor')
            if not cursor or not isinstance(cursor, compat_str):
                break

    def _real_extract(self, url):
        user_or_tag = self._match_id(url)
        webpage = self._download_webpage(url, user_or_tag)
        data = self._parse_graphql(webpage, user_or_tag)

        self._set_cookie('instagram.com', 'ig_pr', '1')

        return self.playlist_result(
            self._extract_graphql(data, url), user_or_tag, user_or_tag)


class InstagramUserIE(InstagramPlaylistIE):
    _VALID_URL = r'https?://(?:www\.)?instagram\.com/(?P<id>[^/]{2,})/?(?:$|[?#])'
    IE_DESC = 'Instagram user profile'
    IE_NAME = 'instagram:user'
    _TEST = {
        'url': 'https://instagram.com/porsche',
        'info_dict': {
            'id': 'porsche',
            'title': 'porsche',
        },
        'playlist_count': 5,
        'params': {
            'extract_flat': True,
            'skip_download': True,
            'playlistend': 5,
        }
    }

    _QUERY_HASH = '42323d64886122307be10013ad2dcc44',

    @staticmethod
    def _parse_timeline_from(data):
        # extracts the media timeline data from a GraphQL result
        return data['data']['user']['edge_owner_to_timeline_media']

    @staticmethod
    def _query_vars_for(data):
        # returns a dictionary of variables to add to the timeline query based
        # on the GraphQL of the original page
        return {
            'id': data['entry_data']['ProfilePage'][0]['graphql']['user']['id']
        }


class InstagramTagIE(InstagramPlaylistIE):
    _VALID_URL = r'https?://(?:www\.)?instagram\.com/explore/tags/(?P<id>[^/]+)'
    IE_DESC = 'Instagram hashtag search'
    IE_NAME = 'instagram:tag'
    _TEST = {
        'url': 'https://instagram.com/explore/tags/lolcats',
        'info_dict': {
            'id': 'lolcats',
            'title': 'lolcats',
        },
        'playlist_count': 50,
        'params': {
            'extract_flat': True,
            'skip_download': True,
            'playlistend': 50,
        }
    }

    _QUERY_HASH = 'f92f56d47dc7a55b606908374b43a314',

    @staticmethod
    def _parse_timeline_from(data):
        # extracts the media timeline data from a GraphQL result
        return data['data']['hashtag']['edge_hashtag_to_media']

    @staticmethod
    def _query_vars_for(data):
        # returns a dictionary of variables to add to the timeline query based
        # on the GraphQL of the original page
        return {
            'tag_name':
                data['entry_data']['TagPage'][0]['graphql']['hashtag']['name']
        }