Skip to content

TikTok Scraper. Download video posts, collect user/trend/hashtag/music feed metadata, sign URL and etc.

Notifications You must be signed in to change notification settings

Tom-A-Lynch/tiktok-scraper

 
 

Repository files navigation

TikTok Scraper & Downloader

NPM npm Codacy grade CI

Scrape and download useful information from TikTok.

No login or password are required

This is not an official API support and etc. This is just a scraper that is using TikTok Web API to scrape media and related meta information.


Buy Me A Coffee

Discord Server


Content

Important notes

  • As of right now it is NOT possible to download video without the watermark

Features

  • Download unlimited post metadata from the User, Hashtag, Trends, or Music-Id pages
  • Save post metadata to the JSON/CSV files
  • Download media with and without the watermark and save to the ZIP file
  • Download single video without the watermark from the CLI
  • Sign URL to make custom request to the TikTok API
  • Extract metadata from the User, Hashtag and Single Video pages
  • Save previous progress and download only new videos that weren't downloaded before. This feature only works from the CLI and only if download flag is on.
  • View and manage previously downloaded posts history in the CLI
  • Scrape and download user, hashtag, music feeds and single videos specified in the file in batch mode

To Do

  • CLI: save progress to avoid downloading same videos
  • Rewrite everything in TypeScript
  • Improve proxy support
  • Add tests
  • Download video without the watermark
  • Indicate in the output file(csv/json) if the video was downloaded or not
  • Build and run from Docker
  • CLI: Scrape and download in batch
  • CLI: Load proxies from a file
  • CLI: Optional ZIP
  • Renew API
  • Set WebHook URL (CLI)
  • Add new method to collect music metadata
  • Add Manual Pagination
  • Improve documentation
  • Download audio files
  • Web interface

Contribution

  • Don't forget about tests
yarn test
yarn build

Installation

tiktok-scraper requires Node.js v10+ to run.

Install from NPM

npm i -g tiktok-scraper

Install from YARN

yarn global add tiktok-scraper

USAGE

In Terminal

$ tiktok-scraper --help

Usage: tiktok-scraper <command> [options]

Commands:
  tiktok-scraper user [id]     Scrape videos from username. Enter only username
  tiktok-scraper hashtag [id]  Scrape videos from hashtag. Enter hashtag without #
  tiktok-scraper trend         Scrape posts from current trends
  tiktok-scraper music [id]    Scrape posts from a music id number
  tiktok-scraper video [id]    Download single video without the watermark
  tiktok-scraper history       View previous download history
  tiktok-scraper from-file [file] [async]  Scrape users, hashtags, music, videos mentioned
                                in a file. 1 value per 1 line

Options:
  --version            Show version number                             [boolean]
  --session            Set session cookie value. Sometimes session can be
                       helpful when scraping data from any method  [default: ""]
  --session-file       Set path to the file with list of active sessions. One
                       session per line!                           [default: ""]
  --timeout            Set timeout between requests. Timeout is in Milliseconds:
                       1000 mls = 1 s                               [default: 0]
  --number, -n         Number of posts to scrape. If you will set 0 then all
                       posts will be scraped                        [default: 0]
  --since              Scrape no posts published before this date (timestamp).
                       If set to 0 the filter is deactived          [default: 0]
  --proxy, -p          Set single proxy                            [default: ""]
  --proxy-file         Use proxies from a file. Scraper will use random proxies
                       from the file per each request. 1 line 1 proxy.
                                                                   [default: ""]
  --download, -d       Download video posts to the folder with the name input
                       [id]                           [boolean] [default: false]
  --asyncDownload, -a  Number of concurrent downloads               [default: 5]
  --hd                 Download video in HD. Video size will be x5-x10 times
                       larger and this will affect scraper execution speed. This
                       option only works in combination with -w flag
                                                      [boolean] [default: false]
  --zip, -z            ZIP all downloaded video posts [boolean] [default: false]
  --filepath           File path to save all output files.
      [default: "/Users/karl.wint/Documents/projects/javascript/tiktok-scraper"]
  --filetype, -t       Type of the output file where post information will be
                       saved. 'all' - save information about all posts to the`
                       'json' and 'csv'
                               [choices: "csv", "json", "all", ""] [default: ""]
  --filename, -f       Set custom filename for the output files    [default: ""]
  --noWaterMark, -w    Download video without the watermark. NOTE: With the
                       recent update you only need to use this option if you are
                       scraping Hashtag Feed. User/Trend/Music feeds will have
                       this url by default            [boolean] [default: false]
  --store, -s          Scraper will save the progress in the OS TMP or Custom
                       folder and in the future usage will only download new
                       videos avoiding duplicates     [boolean] [default: false]
  --historypath        Set custom path where history file/files will be stored
                   [default: "/var/folders/d5/fyh1_f2926q7c65g7skc0qh80000gn/T"]
  --remove, -r         Delete the history record by entering "TYPE:INPUT" or
                       "all" to clean all the history. For example: user:bob
                                                                   [default: ""]
  --webHookUrl         Set webhook url to receive scraper result as HTTP
                       requests. For example to your own API       [default: ""]
  --method             Receive data to your webhook url as POST or GET request
                                      [choices: "GET", "POST"] [default: "POST"]
  --help               Show help                                       [boolean]

Examples:
  tiktok-scraper user USERNAME -d -n 100 --session sid_tt=dae32131231
  tiktok-scraper trend -d -n 100 --session sid_tt=dae32131231
  tiktok-scraper hashtag HASHTAG_NAME -d -n 100 --session sid_tt=dae32131231
  tiktok-scraper music MUSIC_ID -d -n 50 --session sid_tt=dae32131231
  tiktok-scraper video https://www.tiktok.com/@tiktok/video/6807491984882765062 -d
  tiktok-scraper history
  tiktok-scraper history -r user:bob
  tiktok-scraper history -r all
  tiktok-scraper from-file BATCH_FILE ASYNC_TASKS -d

Output File Example

Demo

Docker

By using docker you won't be able to use --filepath and --historypath , but you can set volume(host path where all files will be saved) by using -v

Build
docker build . -t tiktok-scraper
Run

Example 1: All files including history file will be saved in the directory($pwd) where you running the docker from

docker run -v $(pwd):/usr/app/files tiktok-scraper user tiktok -d -n 5 -s

Example 2: All files including history file will be saved in /User/blah/downloads

docker run -v /User/blah/downloads:/usr/app/files tiktok-scraper user tiktok -d -n 5 -s

Module

Methods

.user(id, options) //Scrape posts from a specific user (Promise)
.hashtag(id, options) //Scrape posts from hashtag section (Promise)
.trend('', options) // Scrape posts from a trends section (Promise)
.music(id, options) // Scrape posts by music id (Promise)

.userEvent(id, options) //Scrape posts from a specific user (Event)
.hashtagEvent(id, options) //Scrape posts from hashtag section (Event)
.trendEvent('', options) // Scrape posts from a trends section (Event)
.musicEvent(id, options) // Scrape posts by music id (Event)

.getUserProfileInfo('USERNAME', options) // Get user profile information
.getHashtagInfo('HASHTAG', options) // Get hashtag information
.signUrl('URL', options) // Get signature for the request
.getVideoMeta('WEB_VIDEO_URL', options) // Get video meta info, including video url without the watermark
.getMusicInfo('https://www.tiktok.com/music/original-sound-6801885499343571718', options) // Get music metadata

Options

const options = {
    // Number of posts to scrape: {int default: 20}
    number: 50,

    // Scrape posts published since this date: { int default: 0}
    since: 0,

    // Set session: {string[] default: ['']}
    // Authenticated session cookie value is required to scrape user/trending/music/hashtag feed
    // You can put here any number of sessions, each request will select random session from the list
    sessionList: ['sid_tt=21312213'],

    // Set proxy {string[] | string default: ''}
    // http proxy: 127.0.0.1:8080
    // socks proxy: socks5://127.0.0.1:8080
    // You can pass proxies as an array and scraper will randomly select a proxy from the array to execute the requests
    proxy: '',

    // Set to {true} to search by user id: {boolean default: false}
    by_user_id: false,

    // How many post should be downloaded asynchronously. Only if {download:true}: {int default: 5}
    asyncDownload: 5,

    // How many post should be scraped asynchronously: {int default: 3}
    // Current option will be applied only with current types: music and hashtag
    // With other types it is always 1 because every request response to the TikTok API is providing the "maxCursor" value
    // that is required to send the next request
    asyncScraping: 3,

    // File path where all files will be saved: {string default: 'CURRENT_DIR'}
    filepath: `CURRENT_DIR`,

    // Custom file name for the output files: {string default: ''}
    fileName: `CURRENT_DIR`,

    // Output with information can be saved to a CSV or JSON files: {string default: 'na'}
    // 'csv' to save in csv
    // 'json' to save in json
    // 'all' to save in json and csv
    // 'na' to skip this step
    filetype: `na`,

    // Set custom headers: user-agent, cookie and etc
    // NOTE: When you parse video feed or single video metadata then in return you will receive {headers} object
    // that was used to extract the information and in order to access and download video through received {videoUrl} value you need to use same headers
    headers: {
        'user-agent': "BLAH",
        referer: 'https://www.tiktok.com/',
        cookie: `tt_webid_v2=68dssds`,
    },

    // Download video without the watermark: {boolean default: false}
    // Set to true to download without the watermark
    // This option will affect the execution speed
    noWaterMark: false,

    // Create link to HD video: {boolean default: false}
    // This option will only work if {noWaterMark} is set to {true}
    hdVideo: false,

    // verifyFp is used to verify the request and avoid captcha
    // When you are using proxy then there are high chances that the request will be
    // blocked with captcha
    // You can set your own verifyFp value or default(hardcoded) will be used
    verifyFp: '',

    // Switch main host to Tiktok test enpoint.
    // When your requests are blocked by captcha you can try to use Tiktok test endpoints.
    useTestEndpoints: false
};

Don't forget to check the examples folder

Promise

const TikTokScraper = require('tiktok-scraper');

// User feed by username
(async () => {
    try {
        const posts = await TikTokScraper.user('USERNAME', {
            number: 100,
            sessionList: ['sid_tt=58ba9e34431774703d3c34e60d584475;']
        });
        console.log(posts);
    } catch (error) {
        console.log(error);
    }
})();

// User feed by user id
// Some TikTok user id's are larger then MAX_SAFE_INTEGER, you need to pass user id as a string
(async () => {
    try {
        const posts = await TikTokScraper.user(`USER_ID`, {
            number: 100,
            by_user_id: true,
            sessionList: ['sid_tt=58ba9e34431774703d3c34e60d584475;']
        });
        console.log(posts);
    } catch (error) {
        console.log(error);
    }
})();

// Trending feed
(async () => {
    try {
        const posts = await TikTokScraper.trend('', {
            number: 100,
            sessionList: ['sid_tt=58ba9e34431774703d3c34e60d584475;']
        });
        console.log(posts);
    } catch (error) {
        console.log(error);
    }
})();

// Hashtag feed
(async () => {
    try {
        const posts = await TikTokScraper.hashtag('HASHTAG', {
            number: 100,
            sessionList: ['sid_tt=58ba9e34431774703d3c34e60d584475;']
        });
        console.log(posts);
    } catch (error) {
        console.log(error);
    }
})();

// Get single user profile information: Number of followers and etc
// input - USERNAME
// options - not required
(async () => {
    try {
        const user = await TikTokScraper.getUserProfileInfo('USERNAME', options);
        console.log(user);
    } catch (error) {
        console.log(error);
    }
})();

// Get single hashtag information: Number of views and etc
// input - HASHTAG NAME
// options - not required
(async () => {
    try {
        const hashtag = await TikTokScraper.getHashtagInfo('HASHTAG', options);
        console.log(hashtag);
    } catch (error) {
        console.log(error);
    }
})();


// Get single video metadata
// input - WEB_VIDEO_URL
// For example: https://www.tiktok.com/@tiktok/video/6807491984882765062
// options - not required
(async () => {
    try {
        const videoMeta = await TikTokScraper.getVideoMeta('https://www.tiktok.com/@tiktok/video/6807491984882765062', options);
        console.log(videoMeta);
    } catch (error) {
        console.log(error);
    }
})();

Event

const TikTokScraper = require('tiktok-scraper');

const users = TikTokScraper.userEvent("tiktok", { number: 30 });
users.on('data', json => {
    //data in JSON format
});
users.on('done', () => {
    //completed
});
users.on('error', error => {
    //error message
});
users.scrape();

const hashtag = TikTokScraper.hashtagEvent("summer", { number: 250, proxy: 'socks5://1.1.1.1:90' });
hashtag.on('data', json => {
    //data in JSON format
});
hashtag.on('done', () => {
    //completed
});
hashtag.on('error', error => {
    //error message
});
hashtag.scrape();

Get Set Session

NOT REQUIRED

Very common problem is when tiktok is blacklisting your IP/PROXY and in such case you can try to set session and there will be higher chances for success

Get the session:

  • Open https://www.tiktok.com/ in any browser
  • Login in to your account
  • Right click -> inspector -> networking
  • Refresh page -> select any request that was made to the tiktok -> go to the Request Header sections -> Cookies
  • Find in cookies sid_tt value. It usually looks like that: sid_tt=521kkadkasdaskdj4j213j12j312;
  • sid_tt=521kkadkasdaskdj4j213j12j312; - this will be your authenticated session cookie value that should be used to scrape user/hashtag/music/trending feed

Set the session:

  • CLI:

    • Set single session by using option --session. For example --session sid_tt=521kkadkasdaskdj4j213j12j312;
    • Set path to the file with the list of sessions by using option --session-file. For example --session-file /var/bob/sessionList.txt
      • Example content /var/bob/sessionList.txt:
      sid_tt=521kkadkasdaskdj4j213j12j312;
      sid_tt=521kkadkasdaskdj4j213j12j312;
      sid_tt=521kkadkasdaskdj4j213j12j312;
      sid_tt=521kkadkasdaskdj4j213j12j312;
      
  • In the MODULE you can set session by setting the option value sessionList . For example sessionList:["sid_tt=521kkadkasdaskdj4j213j12j312;", "sid_tt=12312312312312;"]

Download Video

This part is related to the MODULE usage (NOT THE CLI)

The {videoUrl} value is binded to the cookie value {tt_webid_v2} that can contain any value

Method 1: default headers

When you extract videos from the user, hashtag, music, trending feed or single video then in response besides the video metadata you will receive headers object that will contain params that were used to extract the data. Here is the important part, in order to access/download video through {videoUrl} value you need to use same {headers} values.

    headers: {
        "user-agent": "Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_5) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/86.0.4240.80 Safari/537.36",
        "referer": "https://www.tiktok.com/",
        "cookie": "tt_webid_v2=689854141086886123"
    },

Method 2: custom headers

You can pass your own headers with the {options}.

const headers = {
    "user-agent": "BOB",
    "referer": "https://www.tiktok.com/",
    "cookie": "tt_webid_v2=BOB"
}
getVideoMeta('WEB_VIDEO_URL', {headers})
user('WEB_VIDEO_URL', {headers})
hashtag('WEB_VIDEO_URL', {headers})
trend('WEB_VIDEO_URL', {headers})
music('WEB_VIDEO_URL', {headers})
// And after you can access video through {videoUrl} value by using same custom headers

Json Output Example

Video Feed

Example output for the methods: user, hashtag, trend, music, userEvent, hashtagEvent, musicEvent, trendEvent

{
    headers: {
        'user-agent': 'Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_5) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/86.0.4240.80 Safari/537.36',
        referer: 'https://www.tiktok.com/',
        cookie: 'tt_webid_v2=689854141086886123'
    },
    collector:[{
        id: 'VIDEO_ID',
        text: 'CAPTION',
        createTime: '1583870600',
        authorMeta:{
            id: 'USER ID',
            name: 'USERNAME',
            following: 195,
            fans: 43500,
            heart: '1093998',
            video: 3,
            digg: 95,
            verified: false,
            private: false,
            signature: 'USER BIO',
            avatar:'AVATAR_URL'
        },
        musicMeta:{
            musicId: '6808098113188120838',
            musicName: 'blah blah',
            musicAuthor: 'blah',
            musicOriginal: true,
            playUrl: 'SOUND/MUSIC_URL',
        },
        covers:{
            default: 'COVER_URL',
            origin: 'COVER_URL',
            dynamic: 'COVER_URL'
        },
        imageUrl:'IMAGE_URL',
        videoUrl:'VIDEO_URL',
        videoUrlNoWaterMark:'VIDEO_URL_WITHOUT_THE_WATERMARK',
        videoMeta: { width: 480, height: 864, ratio: 14, duration: 14 },
        diggCount: 2104,
        shareCount: 1,
        playCount: 9007,
        commentCount: 50,
        mentions: ['@bob', '@sam', '@bob_again', '@and_sam_again'],
        hashtags:
        [{
            id: '69573911',
            name: 'PlayWithLife',
            title: 'HASHTAG_TITLE',
            cover: [Array]
        }...],
        downloaded: true
    }...],
    //If {filetype} and {download} options are enbabled then:
    zip: '/{CURRENT_PATH}/user_1552963581094.zip',
    json: '/{CURRENT_PATH}/user_1552963581094.json',
    csv: '/{CURRENT_PATH}/user_1552963581094.csv'
}
getUserProfileInfo
{
    secUid: 'MS4wLjABAAAAv7iSuuXDJGDvJkmH_vz1qkDZYo1apxgzaxdBSeIuPiM',
    userId: '107955',
    isSecret: false,
    uniqueId: 'tiktok',
    nickName: 'TikTok',
    signature: 'Make Your Day',
    covers: ['COVER_URL'],
    coversMedium: ['COVER_URL'],
    following: 490,
    fans: 38040567,
    heart: '211522962',
    video: 93,
    verified: true,
    digg: 29,
}
getHashtagInfo
{
    challengeId: '4231',
    challengeName: 'love',
    text: '',
    covers: [],
    coversMedium: [],
    posts: 66904972,
    views: '194557706433',
    isCommerce: false,
    splitTitle: ''
}
getVideoMeta
{
    headers: {
        'user-agent': 'Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_5) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/86.0.4240.80 Safari/537.36',
        referer: 'https://www.tiktok.com/',
        cookie: 'tt_webid_v2=689854141086886123'
    },
    collector:[{
        id: '6807491984882765062',
        text: 'We’re kicking off the #happyathome live stream series today at 5pm PT!',
        createTime: '1584992742',
        authorMeta: { id: '6812221792183403526', name: 'blah' },
        musicMeta:{
            musicId: '6822233276137213677',
            musicName: 'blah',
            musicAuthor: 'blah'
        },
        imageUrl: 'IMAGE_URL',
        videoUrl: 'VIDEO_URL',
        videoUrlNoWaterMark: 'VIDEO_URL_WITHOUT_THE_WATERMARK',
        videoMeta: { width: 480, height: 864, ratio: 14, duration: 14 },
        covers:{
            default: 'COVER_URL',
            origin: 'COVER_URL'
        },
        diggCount: 49292,
        shareCount: 339,
        playCount: 614678,
        commentCount: 4023,
        downloaded: false,
        hashtags: [],
    }]
}
getMusicInfo
{
    music: {
        id: '6882925279036066566',
        title: 'doja x calabria',
        playUrl: 'dfdfdfdf',
        coverThumb:
            'dfdfdf',
        coverMedium:
            'dfdfdf',
        coverLarge:
            'fdfdf',
        authorName: 'bryce',
        original: true,
        playToken:
            'ffdfdf',
        keyToken: 'dfdfdfd',
        audioURLWithcookie: false,
        private: false,
        duration: 46,
        album: '',
    },
    author: {
        id: '6835300004094166021',
        uniqueId: 'mashupsbybryce',
        nickname: 'bryce',
        avatarThumb:
            'dfdfd',
        avatarMedium:
            'dfdfdf',
        avatarLarger:
            'dfdfdf',
        signature: 'hi ily :)\n70k sounds cool tbh\n👇follow my soundcloud & insta👇',
        verified: false,
        secUid: 'MS4wLjABAAAA1_5bjLAamayD4rv3q49qJGa_7dZ5jzExTO0ozOybqIwwhw5TAg_iM25lkO94DM3K',
        secret: false,
        ftc: false,
        relation: 0,
        openFavorite: false,
        commentSetting: 0,
        duetSetting: 0,
        stitchSetting: 0,
        privateAccount: false,
    },
    stats: { videoCount: 361700 },
    shareMeta: {
        title: 'bryceyouloser | ♬ doja x calabria | on TikTok',
        desc: '361.0k videos - Watch awesome short ' + 'videos created with ♬ doja x calabria',
    },
};

Buy Me A Coffee


License


MIT

Free Software

About

TikTok Scraper. Download video posts, collect user/trend/hashtag/music feed metadata, sign URL and etc.

Resources

Stars

Watchers

Forks

Packages

No packages published

Languages

  • TypeScript 96.3%
  • JavaScript 3.5%
  • Dockerfile 0.2%