kiteco-public/linux

Developing on Linux

Below are a set of instructions that get you from a vanilla Linux (Ubuntu) installation to building Kite.

You will have to generate SSH keys, setup GOPATH and clone the repository from the Get the codebase section below.

Setting up to develop on Linux

This document will go through how to setup a vanilla Linux (Ubuntu) installation to work with the Kite codebase. Once you complete this guide, you should be able to build both the Copilot and kited.exe, and run them together by running the ./run_kited.sh command in this directory.

Install Golang

Go to the official golang website and download the latest version of Golang supported by our codebase. You can also use devops/scripts/install-golang.sh

Get the codebase

Generate your ssh keys:

ssh-keygen -t rsa -b 4096 -C "your_email@example.com"

Add your PUBLIC KEY (e.g cat ~/.ssh/id_rsa.pub) it to your GitHub account under Settings > SSH Keys.

Setup your GOPATH and clone the repository

mkdir -p ~/go/src/github.com/kiteco
cd ~/go/src/github.com/kiteco
git clone git@github.com:kiteco/kiteco

Then, run

git lfs pull

to grab the necessary binary files.

Start building

Go to the linux directory.

For 1st time setup, you'll need to build the copilot first:

./build_electron.sh

To run Kite:

./run_kited.sh

If the visual copilot doesn't startup automatically, run:

./linux-unpacked/kite

If you get this error:

go build github.com/kiteco/kiteco/vendor/github.com/kiteco/tensorflow/tensorflow/go: invalid flag in #cgo LDFLAGS: -Wl,-rpath,/home/kite/go/src/github.com/kiteco/kiteco/vendor/github.com/kiteco/tensorflow/tensorflow/go/../../../../../../linux/tensorflow/lib,-z,undefs

Add export CGO_LDFLAGS_ALLOW=".*" to your .bashrc or .zshrc file.

VM Notes

If you're running Linux within a VM, and would like to mount a network drive to share code, the default directory is /mnt/hgfs/.