dfg - Dockerfile Generator

dfg is both a go library and an executable that produces valid Dockerfiles using various input channels.

Table of Contents

• Overview
• Installation
  • Installing as an Executable
  • Installing as a Library
• Getting Started
  • Using dfg as an Executable
  • Using dfg as a Library
• Examples
  • YAML File Example
  • Library Usage Example
• TODO

Overview

dfg is a Dockerfile generator that accepts input data from various sources, produces and redirects the generated Dockerfile to an output target such as a file or stdout.

It is especially useful for generating Dockerfile instructions conditionally since Dockerfile language has no control flow logic.

Installation

Installing as an Executable

• MacOS

curl -o dfg -L https://github.com/ozankasikci/dockerfile-generator/releases/download/v0.1.0/dfg_v0.1.0_darwin_amd64
chmod +x dfg && sudo mv dfg /usr/local/bin

• Linux

curl -o dfg -L https://github.com/ozankasikci/dockerfile-generator/releases/download/v0.1.0/dfg_v0.1.0_linux_amd64
chmod +x dfg && sudo mv dfg /usr/local/bin

• Windows

curl -o dfg.exe -L https://github.com/ozankasikci/dockerfile-generator/releases/download/v0.1.0/dfg_v0.1.0_windows_amd64.exe

Installing as a Library

go get -u github.com/ozankasikci/dockerfile-generator

Getting Started

Using dfg as an Executable

Available commands:

dfg generate --input path/to/yaml --out Dockerfile generates a file named Dockerfile

dfg generate --input path/to/yaml --target-field ".server.dockerfile" --out Dockerfile generates a file named Dockerfile reading the .server.dockerfile field of the YAML file.

dfg generate --help lists available flags

Using dfg as a Library

When using dfg as a go library, you need to pass a []dfg.Stage slice as data. This approach enables and encourages multi staged Dockerfiles. Dockerfile instructions will be generated in the same order as in the []dfg.Instruction slice.

Some Instructions accept a runForm field which specifies if the Instruction should be run in the shell form or the exec form. If the runForm is not specified, it will be chosen based on Dockerfile best practices.

For detailed usage example please see Library Usage Example

Examples

Single YAML File per Dockerfile Example

stages:
  final:
    - from:
        image: kstaken/apache2
    - run:
        runForm: shell
        params:
          - apt-get update &&
          - apt-get install -y
          - php5
          - libapache2-mod-php5 &&
          - apt-get clean &&
          - rm -rf /var/lib/apt/lists/*
    - cmd:
        params:
          - /usr/sbin/apache2
          - -D
          - FOREGROUND

Use dfg as binary:

dfg generate -i ./example-input-files/apache-php.yaml --stdout

Or as a library

data, err := dfg.NewDockerFileDataFromYamlFile("./example-input-files/apache-php.yaml")
tmpl := dfg.NewDockerfileTemplate(data)
err = tmpl.Render(output)

Output

FROM kstaken/apache2
RUN apt-get update && apt-get install -y php5 libapache2-mod-php5 && apt-get clean && rm -rf /var/lib/apt/lists/*
CMD ["/usr/sbin/apache2", "-D", "FOREGROUND"]

YAML File Example With Multiple Configurations (Allows using an existing YAML file)

someConfig:
  key: value
serverConfig:
  dockerfile:
    stages:
      final:
        - from:
            image: kstaken/apache2
        - run:
            runForm: shell
            params:
              - apt-get update &&
              - apt-get clean &&
              - rm -rf /var/lib/apt/lists/*

Use dfg as binary:

dfg generate -i ./example-input-files/test-input-with-target-key-6.yaml --target-field ".serverConfig.dockerfile" --stdout

Or as a library

data, err := dfg.NewDockerFileDataFromYamlField("./example-input-files/test-input-with-target-key-6.yaml", ".serverConfig.dockerfile")
tmpl := dfg.NewDockerfileTemplate(data)
err = tmpl.Render(output)

Output

FROM kstaken/apache2
RUN apt-get update && apt-get clean && rm -rf /var/lib/apt/lists/*

Library Usage Example

package main

import dfg "github.com/ozankasikci/dockerfile-generator"

func main() {
	data := &dfg.DockerfileData{
		Stages: []dfg.Stage{
			// Stage 1 - Builder Image
			// An instruction is just an interface, so you can pass custom structs as well
			[]dfg.Instruction{
				dfg.From{
					Image: "golang:1.7.3", As: "builder",
				},
				dfg.User{
					User: "ozan",
				},
				dfg.Workdir{
					Dir: "/go/src/github.com/ozankasikci/dockerfile-generator/",
				},
				dfg.RunCommand{
					Params: []string{"go", "get", "-d", "-v", "golang.org/x/net/html"},
				},
				dfg.CopyCommand{
					Sources: []string{"app.go"}, Destination: ".",
				},
				dfg.RunCommand{
					Params: []string{"CGO_ENABLED=0", "GOOS=linux", "go", "build", "-a", "-installsuffix", "cgo", "-o", "app", "."},
				},
			},
			// Stage 2 - Final Image
			[]dfg.Instruction{
				dfg.From{
					Image: "alpine:latest", As: "final",
				},
				dfg.RunCommand{
					Params: []string{"apk", "--no-cache", "add", "ca-certificates"},
				},
				dfg.User{
					User: "root", Group: "admin",
				},
				dfg.Workdir{
					Dir: "/root/",
				},
				dfg.CopyCommand{
					From: "builder", Sources: []string{"/go/src/github.com/ozankasikci/dockerfile-generator/app"}, Destination: ".",
				},
				dfg.Cmd{
					Params: []string{"./app"},
				},
			},
		},
	}
	tmpl := dfg.NewDockerfileTemplate(data)
	
	// write to a file
	file, err := os.Create("Dockerfile")
	err = tmpl.Render(file)
	
	// or write to stdout
	err = tmpl.Render(os.Stdout)
}

Output

FROM golang:1.7.3 as builder
USER ozan
WORKDIR /go/src/github.com/ozankasikci/dockerfile-generator/
RUN go get -d -v golang.org/x/net/html
COPY app.go .
RUN CGO_ENABLED=0 GOOS=linux go build -a -installsuffix cgo -o app .

FROM alpine:latest as final
RUN apk --no-cache add ca-certificates
USER root:admin
WORKDIR /root/
COPY --from=builder /go/src/github.com/ozankasikci/dockerfile-generator/app .
CMD ["./app"]

TODO

• Add reading Dockerfile data from an existing yaml file support
• Implement json file input channel
• Implement stdin input channel
• Implement toml file input channel