Supersense is a simple and customizable event board. The main goal of supersense is given an open source alternative to the social walls or multiple source event trackers. At the moment, supersense implements only two sources: Twitter & Github, but the architecture was development in a scalable and extendable way.
Supersense was created with a micro-service paradigm in mind, is a stateless service easy to horizontal scale (but also can have persistence and to be an stateful service). The project has two principal services called: Core and Observer, each one is distributed using docker images and published in the DockerHub with the tags minskylab/supersense and minskylab/supersense-observer respectively.
Core
Supersense Core was written in Go and uses GraphQL as API layer to communicate with the observer or any other custom API consumer, the main goal of core is provide a powerful and realtime API to broadcast all the events coming from the configurase sources (Twitter, Github, etc...).
Observer
Supersense Observer was build with Typescript and React, and all the source are bundled* using snowpack a interesting modern technology that use a modern feature called ESM (ES Modules) to distribute files in a unbundled mode, hence we have very fast development flow.
* Snowpack offers two ways to use it, with its unbundled proposal using ESM and another where you can use the classical bundlers (like Webpack, Parcel).
Write a .env config file like:
# .env file
SS_GRAPHQL_PLAYGROUND=true
SS_TWITTER_CONSUMER_KEY=<YOUR_TWITTER_CONSUMER_KEY>
SS_TWITTER_CONSUMER_SECRET=<YOUR_TWITTER_CONSUMER_SECRET_KEY>
SS_TWITTER_ACCESS_TOKEN=<YOUR_TWITTER_ACCESS_TOKEN_KEY>
SS_TWITTER_ACCESS_SECRET=<YOUR_TWITTER_ACCESS_SECRET_KEY>
SS_TWITTER_QUERY="#peru"
SS_GITHUB_TOKEN=<GITHUB_TOKEN>
SS_GITHUB_REPOS=minskylab/supersense,minskylab/figport,minskylab/baseSupersense core is delivered using docker hub. To run only, execute:
$ docker run -p 8080:8080 --env-file .env minskylab/supersense// TODO
// TODO
Currently supersense only accepts one way to config the service parameters: the environment variables. Below you can see all of this variables.
| Environment Variable Name | Default Value | Description |
|---|---|---|
| SS_PORT | 4000 | |
| SS_TWITTER_CONSUMER_KEY | - | |
| SS_TWITTER_CONSUMER_SECRET | - | |
| SS_TWITTER_ACCESS_TOKEN | - | |
| SS_TWITTER_ACCESS_SECRET | - | |
| SS_TWITTER_QUERY | - | |
| SS_GITHUB_TOKEN | - | |
| SS_GITHUB_REPOS | - | |
| SS_DUMMY_PERIOD | - | |
| SS_DUMMY_MESSAGE | - | |
| SS_SPOKESMAN | false | |
| SS_SPOKESMAN_NAME | - | |
| SS_SPOKESMAN_USERNAME | - | |
| SS_SPOKESMAN_EMAIL | - | |
| SS_PERSISTENCE | false | |
| SS_PERSISTENCE_BOLTDB_FILEPATH | - | |
| SS_PERSISTENCE_REDIS_ADDRESS | - | |
| SS_PERSISTENCE_REDIS_PASSWORD | - | |
| SS_GRAPHQL_PLAYGROUND | false |
// TODO
- More easy observer configuration
- Twitter Images
- Github project summary
- Github open issues (extends the event demux)
- Deployment Instructions
Before your deploy in your k8s cluster you need declare a secret with your configuration. You need to define a .env file with your configuration and use the following command to create a kubernetes generic secret.
$ kubectl create secret generic <your-secret-name> --from-env-file=.envThen, now add the minsky charts repository and install supersense setting your secretRef.
$ helm repo add minsky https://charts.minsky.cc
$ helm repo update # optional
$ helm install minsky/supersense --set config.secretRef=<your-secret-name>