drkain / scrape-youtube

A lightning fast package to scrape YouTube search results

A lightning fast package to scrape YouTube search results. This was made for Discord Bots.

Install

npm install scrape-youtube --save

Example

import { youtube } from 'scrape-youtube';
// const { youtube } = require('scrape-youtube');

youtube.search('Short Change Hero').then((results) => {
    // Unless you specify a custom type you will only receive 'video' results
    console.log(results.videos);
});
// or
const { videos } = await youtube.search('Short Change Hero');

Custom Types

Supported types are: video, live, movie, channel, playlist and any. These types use the sp parameter. If you pass your own sp value it will overwrite the types. See the "custom filters" section for more information.

youtube.search('lofi hip hop beats to relax/study to', { type: 'live' }).then((results) => {
    console.log(results.streams);
});

Examples

Please see the examples directory for examples on what data you will receive from each search.
Your IDE should have autocompletion that works with the interface files, but these are here just in case.

If you want help using this in Discord Bot, please visit the wiki page for TS/JS examples.

Playlists

Please note that the "videos" listed in playlist responses will only contain 1-2 items. This is what is available from the search results.
If you wish to load an entire playlist, consider using ytdl-core.

Custom Filters

You can pass { sp: 'ABC' } as the second parameter to use custom filters like upload date, duration, features ect.
You will need to fetch the SP parameter yourself from youtube. Please see this image for an example.

Request Options

You can pass { request: { } } as the second parameter to use a proxy, custom headers, agents ect.
See http.request for more information.

const options = {
    type: 'video',
    request: {
        headers: {
            Cookie: 'PREF=f2=8000000',
            'Accept-Language': 'de'
        }
    }
};

youtube.search('Poets of the fall', options);

For example, using Cookie: 'PREF=f2=8000000' will enable restricted mode to filter out videos with bad language or adult themes.
Additionally, 'Accept-Language': 'de' will load YouTube in German, sometimes resulting in different titles and content responses.

Extra Info

You can use ytdl-core by fent to load extra information like exact upload dates, full descriptions, like/dislike ratio, video ads ect.
Please see this example in the wiki to see how.

Subscriber count

As mentioned in issue #57 some channels appear to have the subscriber count hidden. In this case the package will default to 0.

Handles

Channel handles will be used if available. Some channels do not have handles, if this is the case the channel link will use the old ID, EG: /channel/12345 vs /@user I recommend sticking to the ID as handles are unreliable.

Quick Info

• Multiple pages currently can not be loaded. This may change in the future, but currently you will be limited to 20 results.
• This package will only make 1 request per search, even when using any types.
• If this package stops working, please open an issue on GitHub. I will fix it within a day.

👤 Author

This project was made by Kain (ksir.pw) Feel free to contact me if you have any trouble with this package.

• Website: ksir.pw
• Github: @DrKain
• Discord: drkain

🤝 Contributing

Contributions, issues and feature requests are welcome! Feel free to check issues page.

Show your support

Give a ⭐️ if this project helped you!

📝 License

Copyright © 2018 Kain (ksir.pw). This project is MIT licensed.

scrape-youtube's Issues

Scrape available channels

Is there a away to scrape available youtube channels ?

YouTube Playlist Search always only results in two videos

Describe the bug
When I'm trying to search for a playlist by given ID the right playlist is found but always only has two videos although it should have 15.

YT-Playlist I used: PLcpBBg8UA1Oxz7973OVcCzyBR-caH7hCQ
Result:

{
  id: 'PLcpBBg8UA1Oxz7973OVcCzyBR-caH7hCQ',
  title: "Quentin Tarantino's DJANGO UNCHAINED Official Soundtrack",
  link: 'https://www.youtube.com/playlist?list=PLcpBBg8UA1Oxz7973OVcCzyBR-caH7hCQ',
  thumbnail: 'https://i.ytimg.com/vi/OhlVBpEnjig/hqdefault.jpg',
    name: 'Django Unchained OST',
    link: 'https://www.youtube.com/user/UnchainedSoundtrack',
    verified: false,
    thumbnail: 'https://www.gstatic.com/youtube/img/originals/promo/ytr-logo-for-search_160x160.png'
  },
  videoCount: 15,
  videos: [
    {
      id: 'OhlVBpEnjig',
      title: 'Django (Luis Bacalov)',
      link: 'https://youtu.be/OhlVBpEnjig',
      duration: 174,
      thumbnail: 'https://i.ytimg.com/vi/OhlVBpEnjig/hqdefault.jpg'
    },
    {
      id: 'XAWpIQATcRE',
      title: 'The Braying Mule (Ennio Morricone)',
      link: 'https://youtu.be/XAWpIQATcRE',
      duration: 154,
      thumbnail: 'https://i.ytimg.com/vi/XAWpIQATcRE/hqdefault.jpg'
    }
  ]
}

Nevertheless great library by the way! I really like it 👍

[Question] Fetching Video ID

is there any way to fetch the video id of the first result? i JUST wanna log the ID of the first video result

TypeError when using exported search function

const youtube = require('scrape-youtube');

Perhaps obvious but youtube.search is no longer a function this way; any means of updating the example to help others understand I do enjoy this quite a bit! :)

const youtube = require('scrape-youtube').youtube;
youtube.search("hello-world").then((results) => {
    console.log(results);
});

No results for certain short search terms

Describe the bug
On a UK located server, the search term 'duggee' doesn't pull through any results. (same issue for many other terms, lego, cats, 'kids toys play' )

note: safesearch is enabled through the header options: { Cookie: "PREF=f2=8000000" }

To Reproduce
Steps to reproduce the behavior:

1. On UK server, search 'duggee'
2. No returned results available

Expected behavior
Duggee should pull through the 20 results listed here: https://www.youtube.com/results?search_query=duggee

Screenshots

Versions:

• Package: 0.2.6
• Node: 12.20.1

Additional context
'duggee' results pull through on a machine located in Pakistan.

How to pass maxResult with TS?

How can I pass the number of videos length count in youtube.search()?

Convert subscriber count to number

Is your feature request related to a problem? Please describe.
The channel subscriber count is displayed as 381K instead in the search results. Some smaller channels will have the full number, IE: 127
This isn't a huge problem but it would be a nice feature to have.

Describe the solution you'd like
The string should be converted to a number. This should be included along with the original value in the results.

1.2M -> 1200000
381K -> 381000
1.5K -> 1500

Additional context
Example response:

{
    "id": "UC0hNui8bT7yV0Xb8w8YxjHw",
    "name": "Poets of the Fall (Official)",
    "link": "https://www.youtube.com/channel/UC0hNui8bT7yV0Xb8w8YxjHw",
    "verified": true,
    "thumbnail": "https://yt3.ggpht.com/ytc/AAUvwniodp2yJktm7UlbHChoA_yqHNDEAUUZlJOKj6Ltxw=s0?imgmax=0",
    "description": "Welcome to the official Poets of the Fall YouTube channel! Finnish rockers Poets of the Fall - singer Marko, guitarist Olli and ...",
    "videoCount": 98,
    "subscribers": "381K",
    "subscriberCount": 381000
}

Search Filter - Is it possible to search for YouTube videos that has subtitles/CC ?

In YouTube you can filter out the results based on subtitle feature.
Is it possible to add subtitle filter?

Certain searches yield no results

Describe the bug
Hello, following on from closed issue #33 - update seemed to work, but now I'm getting no results again.
'duggee' never works, 'lego' and 'cats' are intermittent.
UK server, using safe serach.

To Reproduce
App code here:

const express = require("express");
const { default: youtube } = require("scrape-youtube");

const cache = require("../helpers/cache");

const Router = express.Router();

Router.get("/search", async (req, res) => {
 try {
   let { q, cached } = req.query;
   const allowCached = cached !== "false";
   if (!q) return res.send({ error: "Invalid query!" });
   q = q.trim();
   if (allowCached) {
     const cachedResults = await cache.getResults(q);
     if (cachedResults && cachedResults.length !== 0) {
       return res.send({ results: cachedResults, cached: true });
     }
   }
   let freshResults = await youtube.search(
     q,
     { safeSearch: true },
     { safeSearch: true, headers: { Cookie: "PREF=f2=8000000" } }
   );
   freshResults = freshResults.videos.map((video) => ({
     id: video.id,
     title: video.title,
   }));
   res.send({ results: freshResults, length: freshResults.length });
   if (freshResults.length > 0) await cache.saveResults(q, freshResults);
 } catch (error) {
   console.error(error);
   res.send({ error: "Backend error." });
 }
});

module.exports = Router;

Please let me know if you require any further info.
Originally posted by @accesstechnology-mike in #33 (comment)

Versions:

• Package: 2.0.9
• Node: 12.20.1

Livestream and Premiere support

Add a feature to find livestreams and premieres by channel id.

Can I search by channel?

I want to use this to scrape all the videos from a channel, but there doesnt seem to be any option to select a channel? Searching using the channel name doesnt return all videos on that channel.

Invalid subscriber count / video count

For some reason YouTube is returning incorrectly labelled data.
The videoCountText contains subscriber count and subscriberCountText contains the channel handle.
I don't modify this data so this is what's being returned by YouTube.

  videoCountText: {
    accessibility: { accessibilityData: [Object] },
    simpleText: '21 subscribers'
  },
  subscriptionButton: { subscribed: false },
  subscriberCountText: { simpleText: '@lespetitesmouettes8996' },

URL Support

Add URL Support
My friend said this module doesn't support url query. So, add query support for url.

Filter videos by hashtag

I know that you can filter whether the video is live or not and so on but it will be helpful to add features to filter only shorts too.

it will be easy since the shorts video has shorts in their url

Feature: Test runner

Preferably something like ava. Doesn't need to be too complex, just a quick search of each type to verify the response is correct.

This issue is open for anyone to add if they feel like contributing, otherwise I'll add it when I get the time.

No results for many short terms

These terms are failing regularly on my app;

'cat'
'cats'
'hey duggee'
'duggee'
'kids toys play'
'toys'
'roblox'
'lego'

Any idea would these might fail, yet most requests work fine?

[Feature request] Support for new channel handles

Currently the old channel ID/URL is returned (still working) but YouTube is using @ handles.
This should be included in the 'channel' object along with the old ID.
Example:

{
    "id": "UCYDD7WruLEgEBfjxeor48aw",
    "name": "The Heavy",
    "link": "https://www.youtube.com/channel/UCYDD7WruLEgEBfjxeor48aw",
    "verified": true,
    "thumbnail": "https://yt3.ggpht.com/oMj9hZlJsbS-JI4MUZnwPDcfwIW36KhYeA5CnOG3_5lwHPnmaY_FL5hjBuUFBo32esrBmoOiZKI=s0?imgmax=0"
}

..will become:

{
    "id": "UCYDD7WruLEgEBfjxeor48aw",
    "name": "The Heavy",
    "handle": "@TheHeavyOfficial",
    "link": "https://www.youtube.com/@TheHeavyOfficial",
    "verified": true,
    "thumbnail": "https://yt3.ggpht.com/oMj9hZlJsbS-JI4MUZnwPDcfwIW36KhYeA5CnOG3_5lwHPnmaY_FL5hjBuUFBo32esrBmoOiZKI=s0?imgmax=0"
}

ID should remain as it is unique/locked, unlike handles.

Returns empty array instead of results

Describe the bug
When calling multiple times with any search query, it returns an empty array sometimes.

To Reproduce
Steps to reproduce the behavior:

1. Try calling the example in readme multiple times

Invalid JSON returned on empty searches where itemSectionRenderer does not exist

Describe the bug
Hi there,

This seems to happen when no results are returned:
eg. https://www.youtube.com/results?search_query=Benzinga+Live&sp=EgJAAQ%253D%253D

To Reproduce
Steps to reproduce the behavior:

1. Make a search for which no results are returned
2. See the error TypeError: Cannot read property 'itemSectionRenderer' of undefined

Expected behavior
The search should return empty

Screenshots
The Bezinga search is the one that isn't returning at this time, and this is where the error is thrown.

Versions:

• Package: 2.1.3
• Node: 14.16

Additional context
I guess youtube changed their styling 😢

Unable to fetch results client-side

Describe the bug
I am having problems on the client side.
Access to fetch at 'https://www.youtube.com/results?search_query=aaaaaa&page=0&sp=EgIQAQ%253D%253D' from origin 'http://localhost:3001' has been blocked by CORS policy: No 'Access-Control-Allow-Origin' header is present on the requested resource. If an opaque response serves your needs, set the request's mode to 'no-cors' to fetch the resource with CORS disabled.

Error: Failed to extract InitialData. The request may have been blocked.

Describe the bug
When searching for a video on YouTube it returns the below error:

Error: Failed to extract InitialData. The request may have been blocked.
    at /usr/src/app/node_modules/scrape-youtube/lib/index.js:252:27
    at new Promise (<anonymous>)
    at Youtube.extractInitialData (/usr/src/app/node_modules/scrape-youtube/lib/index.js:228:16)
    at Youtube.<anonymous> (/usr/src/app/node_modules/scrape-youtube/lib/index.js:187:55)
    at step (/usr/src/app/node_modules/scrape-youtube/lib/index.js:44:23)
    at Object.next (/usr/src/app/node_modules/scrape-youtube/lib/index.js:25:53)
    at /usr/src/app/node_modules/scrape-youtube/lib/index.js:19:71
    at new Promise (<anonymous>)
    at __awaiter (/usr/src/app/node_modules/scrape-youtube/lib/index.js:15:12)
    at Request._callback (/usr/src/app/node_modules/scrape-youtube/lib/index.js:174:112)
    at Request.self.callback (/usr/src/app/node_modules/request/request.js:185:22)
    at Request.emit (events.js:314:20)
    at Request.<anonymous> (/usr/src/app/node_modules/request/request.js:1154:10)
    at Request.emit (events.js:314:20)
    at IncomingMessage.<anonymous> (/usr/src/app/node_modules/request/request.js:1076:12)
    at Object.onceWrapper (events.js:420:28)
    at IncomingMessage.emit (events.js:326:22)
    at endReadableNT (_stream_readable.js:1252:12)
    at processTicksAndRejections (internal/process/task_queues.js:80:21)

Before I could fix this by retrying searching.

To Reproduce
Use the package as you would normally.

Expected behavior
The package should not fail permanently on such requests.

Screenshots
https://prnt.sc/v7ysxj

Versions:

• Package: 0.2.4
• Node: 14.11.0

Additional context
none

Reintroduce support for playlists, live streams, ect

Support will need to be added to search for channels, playlists, movies and live streams.
Each type should have their own interface in the interfaces file, Extending the default Result interface when needed.

Channels

• name
• link
• verified
  • If the channel has a verified badge
• thumbnail

Live Streams

• watching
  • Number of users watching the stream
• duration
  • How long (in seconds) the stream has been running for

Playlists

• videoCount
  • The number of videos in the playlist
• videos
  • An array of videos in the playlist. If possible the Video interface should be used.

Movies

These are just the Result interface, but the channel data should be static. All YouTube movies have the same channel information.

Progress:

• Channels
• Playlists
• Live Streams
• Movies

Allow scrapping trending videos

It would be good to have the capability to scrape trending videos and it's related information

Sometimes search incorrectly returns no results

Describe the bug
Hy, I don't know if it's really a bug or if there is any reason to that but sometime there is no video in the result, but for the same search in youtube, there is some result

To Reproduce
Steps to reproduce the behavior:

youtube.search("l'enfant sauvage").then(results => {
    console.log(results)
});

So the expected result is something with like 20 videos
But i got {videos: Array(0), playlists: Array(0), streams: Array(0)}

With most of my research there is no error, but there is some with request containing a certain word like:
l'enfant sauvage
Hero Of War
Bombtrack
Incense
Thunderstruck

Versions:

• Package: 7.0.8 (npm -v youtube-scraper)
• Node: 15.2.1

Explicit Content Filter

Adult Content Filter
I looked on all sites and docs but was unable to find a filter for Explicit Content

You should add this feature

Language

Is your feature request related to a problem? Please describe.
Apparently the json I'm getting now is in Ukrainian and I don't know if I can change it to English

Describe the solution you'd like
Changing the language according to preferences

Additional context

Add proxy support

Simply to bypass limits or blocks.

Proxied requests should be optional.
Options should be passed as arguments while searching, Something like:

search('OK Go', { limit : 5, proxy : { ... } })

Country filter

How to apply country filter for search query so that it will returns videos of that particular country

TypeError: Cannot read property 'simpleText' of undefined

Describe the bug
Hi there! I got this error on my debugger TypeError: Cannot read property 'simpleText' of undefined from searching some channel.
Channel is hiding the information about subscribers

I think you need add try catch for this function. Thank you

const convertSubs = (channel: any): number => {
    const count = channel.subscriberCountText.simpleText.split(' ').shift();

    // If there's no K, M or B at the end.
    if (!isNaN(+count)) return +count;

    const char = count.slice(-1);
    let slicedCount = Number(count.slice(0, -1));

    switch (char.toLowerCase()) {
        case 'k':
            slicedCount *= 1000;
            break;
        case 'k':
            slicedCount *= 1e6;
            break;
        case 'b':
            slicedCount *= 1e9;
            break;
    }

    return ~~slicedCount;
};

To Reproduce
Steps to reproduce the behavior:

1. Search query is UCadOPMAkX21lbuA1yj0JfCQ
2. See error

Expected behavior
It should return the channel's information with subscriberCount is zero or just '-'

Screenshots

Versions:

• Package: 2.1.14
• Node: 10.21.0

Additional context
Add any other context about the problem here.

proxy

hi
thanks for this source code
how use proxy if blocket my ip?

Video descriptions are always blank

Video description snippets are always blank. It appears the render data now uses detailedMetadataSnippets instead of descriptionSnippet

Versions:

• Package: 2.1.13

Video count unavailable for channels

Related to #66. It seems when YouTube added handles they broke some other datasets, so video counts will be 0 until YouTube makes them available again.

Custom Filter

Please make sure you have searched for your question in the closed issues before opening an issue

Hi, you mention in the readme that there are custom filter options, but don't really explain how to use it. Are you able to give a bit more info? And would I be able to use it to limit the searches to come from specific channels?

Google ads are being scraped

Describe the bug
Google ads are sometimes scraped causing an incorrect search result. This is tricky to reproduce because the response is not consistent. A sample is required to properly filter out the results.

Expected behavior
Google ads should not be scraped.

Screenshots

Versions:

• Package: 0.1.7
• Node: v13.8.0

Additional context
Add any other context about the problem here.

Search by page

How to perform paging, load search by page？

TypeError: Cannot read property 'itemSectionRenderer' of undefined

Describe the bug
Running v2.0.5 but getting this error
Failed to extract video data. Please report this issue on GitHub so it can be fixed.

To Reproduce
Steps to reproduce the behavior:
yarn add scrape-youtube
include using promises:

    let output = await youtube.search(req.query.q, {
      page: req.query.page || 1
    })

yarn start

Results: Failed to extract video data. Please report this issue on GitHub so it can be fixed.

TypeError: Cannot read property 'contents' of undefined

Describe the bug
About 1/4 times my search return Failed to extract video data. Please report this issue on GitHub so it can be fixed..

To Reproduce
The function I'm running this in, and the search terms are below:

const search = async function (searchTerm, link, viewers, channel) {
    return youtube.search(searchTerm, { type: 'live' }).then(results => {
        const unembeddable = [
            'w_Ma8oQLmSM' //ABC
        ];
        let verified = results.streams.filter(result => result.channel.link == link);
        let popular = verified.filter(stream => stream.watching > viewers);
        popular.map(item => (item, (item.gridChannel = channel)));
        let embeddable = popular.filter(item => unembeddable.indexOf(item.id));
        return embeddable;
    });
};

let sky = search('Sky News live', 'https://www.youtube.com/user/skynews', 100, 'Sky News');
let cnn = search('CNN live', 'https://www.youtube.com/user/CNN', 100, 'CNN');
let euronews = search('Euronews live', 'https://www.youtube.com/user/Euronews', 100, 'Euronews');
let abcUS = search('ABC News live', 'https://www.youtube.com/user/ABCNews', 100, 'ABC News USA');
let cnaSingapore = search('CNA', 'https://www.youtube.com/user/channelnewsasia', 100, 'CNA');
let abcAUS = search('ABC News', 'https://www.youtube.com/channel/UCVgO39Bk5sMo66-6o6Spn6Q', 100, 'ABC News AUS');
let foxnews = search('Fox News live', 'https://www.youtube.com/user/FoxNewsChannel', 100, 'Fox News Channel');
let dw = search('DW News live', 'https://www.youtube.com/channel/UCknLrEdhRCp1aegoMqRaCZg', 100, 'DW');
let msnbc = search('msnbc live', 'https://www.youtube.com/user/msnbcleanforward', 100, 'MSNBC');
let aje = search('Al Jazeera live', 'https://www.youtube.com/user/AlJazeeraEnglish', 100, 'Al Jazeera English');
let france24 = search('France 24 live', 'https://www.youtube.com/user/france24english', 100, 'France24');
let nbcnews = search('NBC News live', 'https://www.youtube.com/user/NBCNews', 100, 'NBC News');
let cbsnews = search('CBS News live', 'https://www.youtube.com/user/CBSNewsOnline', 100, 'CBS News');
let pbsnews = search('PBS News', 'https://www.youtube.com/user/PBSNewsHour', 100, 'PBS Newshour');

Expected behavior
The data is returned without the error 😄

Versions:

• Package: 2.0.7
• Node: 12.18.3

Additional context
Happy to help reproduce this or log out better data to help 👍

URL Support

Is your feature request related to a problem? Please describe.

I'm always frustrated when i add the url in query

Describe the solution you'd like
Add URL support for query, so someone can search the video from url.

Discriminate between "verified artist" and "verified channel"

Verified artists (for music) on youtube have a musical note next to the channel name. This needs to be extracted and included in the search results.

Search erroring out

Describe the bug
Every time I want to search something, it just says: 'Failed to extract video data. The request may have been blocked'

To Reproduce
Steps to reproduce the behavior:
Just do any search with any string

Versions:

• Package: 1.1.0 or 2.0.1
• Node: lts

youtube.search is not a function

Describe the bug
When trying to use search it returns with youtube.search is not a function.

Here is the code below

import youtube from 'scrape-youtube';
const getYouTubeDetails = await youtube.search(`${useSong.artist} - ${useSong.title}`);
console.log(getYouTubeDetails);

Returns with error:

UnhandledPromiseRejectionWarning: TypeError: youtube.search is not a function
    at Timeout._onTimeout (file:///C:/Users/ctouc/Documents/GitHub/billboard-scraper/youtube.js:25:53)

search is not a function

In lastest (2.1.10) version "search" function not working, when i install earlier version (2.1.9) it work

Search Pagination

Hi, It would be nice if we have a pagination option with this plugin.

CORS Error

Describe the bug
I always get a CORS error when trying to scrape youtube from a frontend project

To Reproduce
Steps to reproduce the behavior:

1. Use scrape-youtube in a react project (any frontend project should work)
2. try to scrape the results for a search term
3. See error in console:
   

Versions:

• Package: Newest
• Node: v14.3.0

Maybe try adding https://cors-anywhere.herokuapp.com?
I already tried that but I somehow still get the CORS errors.

the link string on response objects conflicts with the String.link() function

Im using node v14.18.0 and when i try call the link string from a response object it returns the String.link() function instead
A solution to this could be replacing all "links" with "url"

Get the ID of a youtube channel

Hello. I'm writing here because I don't know exactly where to do it.

I'm trying to find a way to get the ID of a YouTube channel, through your "npm".

I know that your "npm" is used to obtain the information of the videos, streams, and playlists of a youtube channel, but I don't know if it is possible to get this type of information

TypeError: The "listener" argument must be of type Function. Received type object

Getting error TypeError: The "listener" argument must be of type Function. Received type object

Any idea why?

Code is in react with a contentEditable div checking for keyboard esacpe command this way:

console.log(term) // correct value typed inside the contentEditable field

const checkReturns = (event) => {
        if(event.key === 'Enter'){

            const options = {
                requestOptions: {
                    headers: {
                        Cookie: 'PREF=f2=8000000'
                    }
                }
            };

            youtube.search(term, options).then(results => {   // this throws the error at .search
                console.log(results.videos)
            });
        }
    }

The react return is as follow:

return (
        <div id="namebox" className="areatext list-group">
            <div
                ref={ref}
                id="first_line"
                contentEditable="true"
                style={{ minHeight: '26px' }}
                onInput={(e) => setTerm(e.currentTarget.textContent)}
                onKeyPress={(event) => checkReturns(event)}>
            </div>
        </div>
    )

Unconfirmed CPU spikes

Received report of extremely high CPU spikes from a single search. This will need to be confirmed before a fix is pushed to ensure the issue is resolved.

on the aws free tier ec2 it spikes up to >80%
and about the same on my 6700k

Learn semantic versioning please

Is your feature request related to a problem? Please describe.
I'm always frustrated when developers increment minor version, but change how API works

Describe the solution you'd like
The change of 8 days ago should be an 2.0.0 because it breaks previous programs using previous version

Additional context
https://semver.org