scrape-twitter

scrape-twitter

🐦 Access Twitter data without an API key

This module provides command line interfaces to scrape: profiles, timelines, connections, likes, search and conversations.

It also exposes both streams and a promise returning function to help accessing Twitter in your own applications.

Real-time firehoses can be created using the companion module monitor-head-stream.

Features

• Get Twitter data without being required to configure an API key.
• Twitter can't constrain access as easily as they can to an API or an individual API key. Any constraints introduced would apply to their public site. A scraper can be fixed; you are no longer beholden to Twitter.
• Grab timelines, whole conversations, profiles, connections, likes, etc.
• Automatically pages to fetch all tweets.
• Provides metadata on how tweet replies are linked together. e.g. isReplyToId

Example

Get profile

$ scrape-twitter profile sebinsua
# ...

Get timeline

$ scrape-twitter timeline nouswaves
# ...

Get likes

This command requires a valid login. It will check for the following environment variables: TWITTER_USERNAME, TWITTER_PASSWORD, TWITTER_KDT. But can also pick these up from a dotenv file at the path ~/.scrape-twitter. The first time you login you will be asked to store the TWITTER_KDT - this is used by Twitter to recognise your device.

$ scrape-twitter likes sebinsua
# ...

Get connections

This command also requires a valid login.

$ scrape-twitter connections sebinsua --type=following
# ...

Get conversation

$ scrape-twitter conversation ctbeiser 691766715835924484
# ...

Search

$ scrape-twitter search --query "from:afoolswisdom motivation" --type latest
# ...

Get list

$ scrape-twitter list nouswaves list
# ...

JSON interface plays nicely with CLI tools like jq, coreutils/gshuf and terminal-notifier

For example, a MOTD-like script might contain:

scrape-twitter search --query="from:afoolswisdom knowledge" | jq -r '.[].text' | gshuf -n 1 | terminal-notifier -title "Knowledge (MOTD)"

Install

With yarn:

yarn global add scrape-twitter

With npm:

npm install -g scrape-twitter

API

new TimelineStream(username: string, { retweets: boolean, replies: boolean, count: ?number })

Create a ReadableStream<Tweet> for the timeline of a username.

new LikeStream(username: string, { count: ?number, env: process.env })

Create a ReadableStream<Tweet> for the likes of a username.

new ConnectionStream(username: string, type: 'following' | 'followers', process.env)

Create a ReadableStream<UserConnection> for the connections of a username.

new ConversationStream(username: string, id: string, { count: ?number })

Create a ReadableStream<Tweet> for the conversation that belongs to a username and tweet id.

new ThreadedConversationStream(id: string)

Create a ReadableStream<Tweet> for the thread that belongs to a tweet id.

new TweetStream(query: string, type: 'top' | 'latest', { count: ?number })

Create a ReadableStream<Tweet> for the tweets that match a query and type.

new ListStream(username: string, list: string, { count: ?number })

Create a ReadableStream<Tweet> for the username's list.

getUserProfile(username: string)

Get a Promise<UserProfile> for a particular username.

Changelog

0.8.4 (2017-02-02)

Bug Fixes

• username: treat _ as string #2 (1652c26)

<a name="0.8.3"></a>