This repository contains a prometheus "exporter" that will collect analytics about tweets for a user timeline
Analytics gathered belong to organic metrics from twitter
You need to have a twitter developper account in order to use this service
And then you will be able to configured with below flags :
Usage of ./twitter-analytics-exporter:
--consumer-key string key used to represents your Twitter developer app
--consumer-secret string secret used to represents your Twitter developer app
--user-access-token string user-specific token credentials used to authenticate OAuth 1.0a
--user-id string user-specific id from where to retrieve timeline
--user-secret-token string user-specific secret credentials used to authenticate OAuth 1.0a
pflag: help requested
You can also use environment variable by prefixing them by TAE_
followed by uppercase value and -
replace by _
. Ex: TAE_USER_ID="1477951133402513413"
consumer
information are credentials related to you twitter app
user
credentials are related to twitter user
This application uses Oauth1.0a without 3-legged Oauth meaning there is no "in-app" way of exchanging user credentials (for now ?)
Best way to deploy this app is using docker image
ghcr.io/laupse/twitter-analytics-exporter
This application is an exporter so it needs to be backed by a :
prometheus
server (or a grafana-agent with long-term storage) in order to scrape metrics, store them and make them available by queryinggrafana
in order to visualize
- Using OAuth1.0a comes with limitation,
- Only tweets younger than 30 days are gathered because of the organic metrics requested
- No easy way to configure scraping user that are not the user behind the app
Overcome this will needs developping a full Oauth 2.0 app