NanoForge

Build and runner tool for nanoservices

Installation

We want to work on packaging for operating systems later but for now you can install using the following command:

wget -qO- https://raw.githubusercontent.com/nanoservicesforge/NanoForge/main/scripts/install.sh | sh

Usage

Before we can use NanoForge we need to package a nanoservice and build it using scratch as seen below:

FROM scratch

COPY ./your_package .

Declaring a nanoservice in your build

And this is enough to package your nanoservice in a Docker image. Now we move onto declaring our nanoservice in our Cargo.toml with the following (nan-one is a real toy example of a nanoservice on Docker Hub):

[nanoservices.nan-one]
dev_image = "maxwellflitton/nan-one"
prod_image = "maxwellflitton/nan-one"
entrypoint = "."

Preparing your build

Now we can prepare our build using the following command (you pwd should be the root of your project):

nanoforge prep

This command will create a .nanoservices_cache and pull the Docker image unpacking the files into the cache. The build tool will then scan all subdirectories looking for nanoservices in all the Cargo.toml files in the project. The relative path will then be calculated for each nanoservice and defined in the Cargo.toml files. For instance, our nan-one nanoservice will have the following path:

[dependencies.nan-one]
path = "../.nanoservices_cache/domain_services/nanoservices/maxwellflitton_nan-one/."

You can then use this in your build.