A simple Go program to consume Twitter's filtered stream API.

This program depends on two environment variables to work (you can copy .env.sample as .env as well):

• TWITTER_API_KEY
• TWITTER_API_SECRET

There are also two optional environment variables:

• TWITTER_APP_NAME which is the name of your Twitter application (used as user-agent in the API requests)
• TWITTER_ACCESS_TOKEN which is highly recommended since Twitter discourage you to use the endpoint that grants this token (read further on to know how to get this token using this client)

You can either run the commands with go run main.go or calling the binary generated with go build main.go.

$ go run main.go token

This command displays the token to be used for authentication. It's highly recommended to save this value as an environment variable called TWITTER_ACCESS_TOKEN (or save it in your .env).

$ go run main.go stream

This command starts to listen to the stream of tweets. Start this program in an instance of the terminal and then move to another instance to create your rules – and wait to see the magic! Once a tweet matches the rules you create in the next step, it will be saved in the data/ directory.

$ go run main.go rule ls

This command lists the existing rules for the filtered stream.

$ go run main.go rule new --query YOUR-SEARCH-TERM-HERE

This command creates a new rule using the text passed with the --query option.

$ go run main.go rule rm

This command deletes all existing rules.

I'm still learning Go, so any feedback, issue, comment or pull request is welcome!

I know we have libraries to handle the requests to Twitter API, but this bare implementation was a way I've found to study net/http, ok?