twcrawl - Twitter Network Crawler

Starts at any given Twitter accounts (for example realDonaldTrump and elonmusk) and follows their relationships to download the profiles of all Twitter users into a local database for later batch processing (like analyzing the social sentiment).

Note: This crawler has been designed to prioritize crawling of accounts with the most followers. Depending on your use case you might need to first tweak the parameters a bit.

Setup

All required dependencies are defined in the requirements.txt file. Run pip install -r requirements.txt to install all of them if needed. Then copy the config.example.json into a config.json and fill in your Twitter API credentials.

Usage

Simply run src/main.py --users users.txt. users.txt should be a list of twitter handles to use as entry points of the crawling process, one per each line of the text file.

This will launch an endless running process, which crawls as many users as possible (and as fast as allowed by the Twitter API limits). You can pause the process by simply killing it and continue the crawling process by starting it again by executing src/main.py again (no more need for the -i parameter).

The crawled database will be stored into data/twitter.sqlite3 (or anywhere else if you override the default values in your config file).