Version 2 of the Twitter API has been the primary way to access the developer platform since November, but, with around 10 years of blog tutorials about v1.1, there is a lot less information shared by the developer community about the new version. Let's start to change that!
Here's a great question I just answered in the Twitter tag on Stack Overflow:
"How do I find the number of followers for a Twitter user in API v2?"
You can see the full question (and my answer) embedded below. I thought I'd post here in order to expand on my response a little.
I am surprised to see that there is no way to simply get the number of followers for one Twitter account.
I found many answers (Follower count number in Twitter, How to obtain follower count in Twitter API 1.1?, How to get followers count from twitter…
The answer is quite simple, but not everyone is familiar with the new data formats yet. In v2, we have reduced the number of fields returned by default, and you can opt to select the information that is useful for your application, using the fields and expansions parameters.
The follower and following counts for a user are part of the public_metrics fields in the User object.
The API endpoint and parameters are in the format
https://api.twitter.com/2/users/[ID]?user.fields=public_metrics,[any other fields]
Below is an example of the output, using the twurl command line tool (I wrote a post about twurl here). In this case, twurl handles the authentication etc, but you could also use the Twitter API v2 Postman Collection, or the API Explorer from the API Tools available on the developer website, or a library of your choice.
$ twurl -j "/2/users/786491?user.fields=public_metrics,created_at"
{
"data": {
"id": "786491",
"username": "andypiper",
"created_at": "2007-02-21T15:14:48.000Z",
"name": "andypiper.xyz",
"public_metrics": {
"followers_count": 16570,
"following_count": 3247,
"tweet_count": 134651,
"listed_count": 826
}
}
}
In this call, we've requested user ID 786491 (which is me! give me a follow, if you feel like it). By default without any parameters, calls to the users lookup endpoint only return the id, name and username values. Here, we added the user.fields to also ask for the public_metrics - followers, following, number of Tweets I've posted, and the number of lists the account has been added to.
So the answer to the original question: to get the follower count, ask for the public_metrics field in the User object, and access data.public_metrics.followers_count in the response.
Let's take it one step further. Expansions and fields are powerful. Using an expansion, we can also ask for my current pinned Tweet and more details about the contents of that Tweet, in a single API call:
$ twurl -j "/2/users/786491?user.fields=public_metrics,created_at,pinned_tweet_id&expansions=pinned_tweet_id&tweet.fields=created_at,public_metrics,source,context_annotations,entities"
{
"data": {
"public_metrics": {
"followers_count": 16571,
"following_count": 3247,
"tweet_count": 134654,
"listed_count": 826
},
"pinned_tweet_id": "1482011997814374406",
"id": "786491",
"username": "andypiper",
"created_at": "2007-02-21T15:14:48.000Z",
"name": "andypiper.xyz"
},
"includes": {
"tweets": [
{
"public_metrics": {
"retweet_count": 31,
"reply_count": 5,
"like_count": 547,
"quote_count": 1
},
"entities": {
"urls": [
{
"start": 113,
"end": 136,
"url": "https://t.co/lCaoNcXcyL",
"expanded_url": "https://dev.to/andypiper/what-resources-help-you-to-learn-a-new-api-2cc9",
"display_url": "dev.to/andypiper/what…",
"images": [
{
"url": "https://pbs.twimg.com/news_img/1482011998338654212/0NiDLEms?format=jpg&name=orig",
"width": 1000,
"height": 500
},
{
"url": "https://pbs.twimg.com/news_img/1482011998338654212/0NiDLEms?format=jpg&name=150x150",
"width": 150,
"height": 150
}
],
"status": 200,
"title": "What resources help you to learn a new API?",
"description": "In my day job, when I'm not learning things by reading and sharing here on DEV, I'm a developer...",
"unwound_url": "https://dev.to/andypiper/what-resources-help-you-to-learn-a-new-api-2cc9"
}
],
"annotations": [
{
"start": 97,
"end": 103,
"probability": 0.9056,
"type": "Product",
"normalized_text": "Twitter"
}
]
},
"text": "I'd love to know more about how you learn to use new APIs, and how I can help you to learn about Twitter API v2. https://t.co/lCaoNcXcyL",
"id": "1482011997814374406",
"source": "Twitter Web App",
"context_annotations": [
{
"domain": {
"id": "46",
"name": "Brand Category",
"description": "Categories within Brand Verticals that narrow down the scope of Brands"
},
"entity": {
"id": "781974596752842752",
"name": "Services"
}
},
{
"domain": {
"id": "47",
"name": "Brand",
"description": "Brands and Companies"
},
"entity": {
"id": "10045225402",
"name": "Twitter"
}
}
],
"created_at": "2022-01-14T15:29:29.000Z"
}
]
}
}
🤯
OK, that's a lot more information! Let's break it down.
Here, I did the following:
- added the
user.fieldto requestpinned_tweet_id, which returned"pinned_tweet_id": "1482011997814374406" - also, I asked for this to be expanded in the response (
expansions=pinned_tweet_id). That gets expanded inside of theincludesarray in the response. - by default, the API only returns the
idandtextfields for a Tweet object, so I also specified that I wanted more information. I asked fortweet.fieldsand specifiedcreated_at,public_metrics,source,context_annotations, andentities-
created_at: creation time of the Tweet, the UTC time when I posted it. -
public_metrics: similar to metrics on the User object, we get data about the numbers of retweets, replies, likes and quotes of this Tweet. -
source: the name of the App that was used to post the Tweet (the Twitter website in this case) -
context_annotations: OK, this is super cool. We get back an array of information (domain/entity pairs) that cover what Twitter has detected this Tweet to be about. In this case I was posting about the Twitter API, so it pulled out that I'm talking about a brand and company (Twitter) in the Tweet. There are also domains and entities for things like movies, sports, products, award shows, and so on. You can even search for Tweets by entity value using the v2 API - very powerful. -
entities: this is also an improvement / additional set of data over v1.1. The entities identified in my original Tweet are broken out and expanded. In this case, there was a shortened URL to a blog post here on DEV. This is expanded to show the full URL, and what is displayed in the Tweet, as well as the URL to the image that is displayed in the website summary large image card (and the dimensions of that image), and the description from that card.
-
Read more about the new Twitter data formats and Tweet payloads on the Developer Blog. Talk to us via the Twitter Developer Community forums.
Share what you learn about Twitter API v2 here on DEV! Use the twitter tag. It will help others to learn, too.
Finally, if you found this interesting or helpful, I'd love to hear from you in the comments, or on Twitter. Happy coding! 💻
Top comments (5)
Two questions:
As the original author, unfortunately I'm no longer able to help here. You could try the Twitter developer forums instead. I do not know whether Twitter employees check DEV any more :-(
No worries. I figured it was a long shot since this post was 1.5 years ago AND the cost of Twitter Developer subscriptions is ridiculous! I'll keep looking around 😃
For anyone looking for your Twitter ID, check this site (worked as of July 20, 2023): commentpicker.com/twitter-id.php
Yes, generally Twitter does a lot to reduce the data available via the APIs. There are 3rd party tools that try handling that, like followays.com for example, though. Works as of May 2024.