Tiny, and simple Elasticsearch microservice to abstract searching objects!
Tsubasa was built to be a simple abstraction to not use the official SDKs to search objects within an Elasticsearch index, so this is just a simple way to retrieve data from it.
⚠️ Tsubasa is alpha software, it is NOT production ready! Beware~ >!<
Sweet, you want to use Tsubasa for your own use cases! You can install the Tsubasa server:
- using the Noelware Helm Charts;
- using the official Tsubasa Docker Image;
- locally under the main repository you see
This is the minimum system requirements to bootstrap Tsubasa. You don't need to worry about this if you're running on the Helm Chart since that handles it for you.
- 2GB or higher of system RAM
- 2 CPU Cores or higher
- Go 1.17 or higher
- An instance of Elasticsearch running. This supports single-node and multi-node instances and multiple authentication methods.
You can install Tsubasa on your Kubernetes cluster with a single command! You will need to index the Noelware Helm Charts under the "noel" user (which is me! :D)
⚠️ You are required to be using Kubernetes >=1.22 and Helm 3!
$ helm repo add noel https://charts.noelware.org/~/noelYou should be able to search the tsubasa repository when using the helm search command.
Now, you should be able to just run tsubasa on a single command:
$ helm install <my-release> noel/tsubasa...and the server should be running now. The helm chart assumes you installed an Elasticsearch cluster installed.
You can use the official Docker images on ghcr.io or on Docker Hub!
~ ; ... coming soon >o< ... ; ~
~ ; ... coming soon >o< ... ; ~
Tsubasa is configured using a TOML file which must be in the following locations:
/app/noel/tsubasa/config.tomlif using the Docker Image$ROOT/config.tomlif running locally or wanting to contribute to TsubasaTSUBASA_CONFIG_FILEenvironment variable, which will override both clauses above.
You can find an example in the documentation!
~ ; ... coming soon >o< ... ; ~
Tsubasa is released under the Apache 2.0 License by Noel!