Tembo CLI

Tembo CLI allows users to experience Tembo locally, as well as, manage and deploy to Tembo Cloud. It abstracts away complexities of configuring, managing, and running Postgres in a local environment.

Local Testing

Clone this repo and run:

cargo install --path .

If the install path is in your shell path, you can then run tembo help and other tembo commands.

Commands

tembo init

The tembo init command initializes your environment with following files:

• tembo.toml configuration file
• migrations directory for sql migrations
• ~/.tembo/context file with various contexts user can connect to

For more information: tembo init --help

tembo context list/set

tembo context works like kubectl context. User can set context for local docker environment or tembo cloud (dev/qa/prod) with org_id. When they run any of the other commands it will run in the context selected. Default context will be local.

tembo apply

Validates Tembo.toml (same as tembo validate) and applies the changes to the context selected.

• applies changes and runs migration for all dbs
  • local docker: wraps docker build/run + sqlx migration
  • tembo-cloud: calls the api in appropriate environment

tembo delete

• local docker: runs docker stop & rm command
• tembo-cloud: calls delete tembo api endpoint

Generating Rust Client from API

OpenAPI Generator tool is used to generate Rust Client.

Install OpenAPI Generator if not already by following steps here

Control plane API client

Go to temboclient directory in your terminal.

Delete the contents of the directory first and then run following command to re-generate the rust client code for the API.

openapi-generator generate -i https://api.tembo.io/api-docs/openapi.json  -g rust -o . --additional-properties=packageName=temboclient

• Go to temboclient/src/lib.rs & add followng line at the top to disable clippy for the generated code

#![allow(clippy::all)]

• Create /temboclient/src/models/impls.rs file & add following code to it:

TODO: Find a better way to do this.

use std::str::FromStr;

use super::{Cpu, Storage, StackType, Memory, Environment};

impl FromStr for Cpu {
	type Err = ();

	fn from_str(input: &str) -> core::result::Result<Cpu, Self::Err> {
			match input {
					"1"  => Ok(Cpu::Variant1),
					"2"  => Ok(Cpu::Variant2),
					"4"  => Ok(Cpu::Variant4),
					"8" => Ok(Cpu::Variant8),
					"16" => Ok(Cpu::Variant16),
					"32" => Ok(Cpu::Variant32),
					_      => Err(()),
			}
	}
}

impl FromStr for Memory {
	type Err = ();

	fn from_str(input: &str) -> core::result::Result<Memory, Self::Err> {
			match input {
					"1Gi"  => Ok(Memory::Variant1Gi),
					"2Gi"  => Ok(Memory::Variant2Gi),
					"4Gi"  => Ok(Memory::Variant4Gi),
					"8Gi" => Ok(Memory::Variant8Gi),
					"16Gi" => Ok(Memory::Variant16Gi),
					"32Gi" => Ok(Memory::Variant32Gi),
					_      => Err(()),
			}
	}
}

impl FromStr for Environment {
	type Err = ();

	fn from_str(input: &str) -> core::result::Result<Environment, Self::Err> {
			match input {
					"dev"  => Ok(Environment::Dev),
					"test"  => Ok(Environment::Test),
					"prod"  => Ok(Environment::Prod),
					_      => Err(()),
			}
	}
}

impl FromStr for Storage {
	type Err = ();

	fn from_str(input: &str) -> core::result::Result<Storage, Self::Err> {
			match input {
					"10Gi" => Ok(Storage::Variant10Gi),
					"50Gi" => Ok(Storage::Variant50Gi),
					"100Gi" => Ok(Storage::Variant100Gi),
					"200Gi" => Ok(Storage::Variant200Gi),
					"300Gi" => Ok(Storage::Variant300Gi),
					"400Gi" => Ok(Storage::Variant400Gi),
					"500Gi" => Ok(Self::Variant500Gi),
					_ => Err(()),
			}
	}
}

impl ToString for StackType {
	fn to_string(&self) -> String {
			match self {
					Self::Standard => String::from("Standard"),
					Self::MessageQueue => String::from("MessageQueue"),
					Self::MachineLearning => String::from("MachineLearning"),
					Self::Olap => String::from("OLAP"),
					Self::Oltp => String::from("OLTP"),
					Self::VectorDb => String::from("VectorDB"),
					Self::DataWarehouse => String::from("DataWarehouse"),
			}
	}
}

• Add following line towards the end of /temboclient/src/models/mod.rs

pub mod impls;

Contributing

Before you start working on something, it's best to check if there is an existing plan first. Join our Slack community and ask there.

Semver

Tembo CLI is following Semantic Versioning 2.0.