Gain insights into writing professional command-line programs in Go. Delve into structuring Go programs, writing idiomatic code, testing effectively, and impressing users and managers.

write professional command  - line programs in Go.png

data.tar.gz

Go run main.go

Run main.sh

Run go test

Run go test2

GoLive

GoLiveTesting

Multi-git-Custom

Pre+main-v0.2

Main_to_cmd

Run go test-Coverage

Ginkgo

Ginkgo-BDD

Ginkgo-e2e

Cobra

Cobra-NoInstall

Cobra-NoInstall-FullApp

Final_Cobra_Multi-git

Viper

Viper-etcd

Viper-Cobra

SelfUpdating

Make

Special build

Embedding_Data

Go run main.go-Simple

In this course you will learn why Go is an awesome language for writing command-line programs. You will learn how to structure your Go programs, how to write idiomatic Go code, how to test your programs and how to amaze your users, peers and managers. 

The focus is on command-line programs in order to keep the course down to earth, but most of the principles and best practices you will learn apply to any Go program.

Write Professional Command-line Programs in Go

## Adding the root command

Since multi-git is an existing application, we will not use the Cobra generator and just add the root command ourselves. Let's place it according to Cobra conventions in `cmd/root.go`:

```
package cmd

import (
	"fmt"
	"os"

	"github.com/spf13/cobra"
)

// rootCmd represents the base command when called without any sub-commands
var rootCmd = &cobra.Command{
	Use:   "multi-git",
	Short: "Runs git commands over multiple repos",
	Long: `Runs git commands over multiple repos.

Requires the following environment variables defined:	
MG_ROOT: root directory of target git repositories
MG_REPOS: list of repository names to operate on`,
	Args:  cobra.ExactArgs(1),
	Run: func(cmd *cobra.Command, args []string) { 
		
	},
}
```

The root command expects exactly one argument, which is the git command. If flags are necessary the user will surround them in quotes, so it is still a single string argument from the shell's point of view.

Let's also add a public Execute() function. This function will be called by the main() function.

```
func Execute() {
	err := rootCmd.Execute()
	if err != nil {
		fmt.Println(err)
		os.Exit(1)
	}
}
```

With the command skeleton in place it's time to implement it.


## Implementing the root command

Our current program already has the implementation isolated in the `Exec()` method of the `RepoManager` type from the `repo_manager` package. 

```
func (m *RepoManager) Exec(cmd string) (output map[string]string, err error) {
	output = map[string]string{}
	var components []string
	.
	.
	.
}
```

The `RepoManager` is initialized with the root directory and the names of all the repositories:

```
func NewRepoManager(baseDir string, 
                    repoNames []string, 
                    ignoreErrors bool) (repoManager *RepoManager, err error) {
	.
	.
	.
}
```

In addition, the `main()` function from cmd/mg/main.go parses the command-line flags and reads the environment variables in order to instantiate a `RepoManager` and invoke its `Exec()` method:

```
func main() {
	command := flag.String("command", "", "The git command")
	ignoreErros := flag.Bool(
		"ignore-errors",
		false,
		"Keep running after error if true")
	flag.Parse()

	// Get managed repos from environment variables
	root := os.Getenv("MG_ROOT")
	if root[len(root)-1] != '/' {
		root += "/"
	}

	repoNames := []string{}
	if len(os.Getenv("MG_REPOS")) > 0 {
		repoNames = strings.Split(os.Getenv("MG_REPOS"), ",")
	}

	repoManager, err := repo_manager.NewRepoManager(root, repoNames, *ignoreErros)
	if err != nil {
		log.Fatal(err)
	}

	output, err := repoManager.Exec(*command)
	if err != nil {
		fmt.Printf("command '%s' failed with error ", err)
	}

	for repo, out := range output {
		fmt.Printf("[%s]: git %s\n", path.Base(repo), *command)
		fmt.Println(out)
	}
}
```

This code needs to go into the root command. Very minimal changes are necessary. Instead of reading the flag `--command`, the root command will get the git command-line arguments as a slice of arguments. Here is the `Run()` method of the root command with the necessary adjustments:

```
Run: func(cmd *cobra.Command, args []string) {
	// Get managed repos from environment variables
	root := os.Getenv("MG_ROOT")
	if root[len(root)-1] != '/' {
		root += "/"
	}

	repoNames := []string{}
	if len(os.Getenv("MG_REPOS")) > 0 {
		repoNames = strings.Split(os.Getenv("MG_REPOS"), ",")
	}

	repoManager, err := repo_manager.NewRepoManager(root, repoNames, ignoreErrors)
	if err != nil {
		log.Fatal(err)
	}

	command := strings.Join(args, " ")
	output, err := repoManager.Exec(command)
	if err != nil {
		fmt.Printf("command '%s' failed with error ", err)
	}

	for repo, out := range output {
		fmt.Printf("[%s]: git %s\n", path.Base(repo), command)
		fmt.Println(out)
	}
},
```

## Providing the `--ignore-errors` Flag

The last piece of the puzzle is adding the `--ignore-errors` flag. The best place to do this is in an init() function:

```
func init() {
	rootCmd.Flags().BoolVar(
		&ignoreErrors,
		"ignore-errors",
		false,
		`will continue executing the command for all repos if ignore-errors
                 is true otherwise it will stop execution when an error occurs`)	
}
```

In the next lesson, we will refactor the main.go file to use our Cobra root command and finalize the refactoring.



# Adding the root command

Since multi-git is an existing application, we will not use the Cobra generator and just add the root command ourselves. Let's place it according to Cobra conventions in `cmd/root.go`:

```
package cmd

import (
	"fmt"
	"os"

	"github.com/spf13/cobra"
)

// rootCmd represents the base command when called without any sub-commands
var rootCmd = &cobra.Command{
	Use:   "multi-git",
	Short: "Runs git commands over multiple repos",
	Long: `Runs git commands over multiple repos.

Requires the following environment variables defined:	
MG_ROOT: root directory of target git repositories
MG_REPOS: list of repository names to operate on`,
	Args:  cobra.ExactArgs(1),
	Run: func(cmd *cobra.Command, args []string) { 
		
	},
}
```

The root command expects exactly one argument, which is the git command. If flags are necessary the user will surround them in quotes, so it is still a single string argument from the shell's point of view.

Let's also add a public Execute() function. This function will be called by the main() function.

```
func Execute() {
	err := rootCmd.Execute()
	if err != nil {
		fmt.Println(err)
		os.Exit(1)
	}
}
```

With the command skeleton in place it's time to implement it.


# Implementing the root command

Our current program already has the implementation isolated in the `Exec()` method of the `RepoManager` type from the `repo_manager` package. 

```
func (m *RepoManager) Exec(cmd string) (output map[string]string, err error) {
	output = map[string]string{}
	var components []string
	.
	.
	.
}
```

The `RepoManager` is initialized with the root directory and the names of all the repositories:

```
func NewRepoManager(baseDir string, 
                    repoNames []string, 
                    ignoreErrors bool) (repoManager *RepoManager, err error) {
	.
	.
	.
}
```

In addition, the `main()` function from cmd/mg/main.go parses the command-line flags and reads the environment variables in order to instantiate a `RepoManager` and invoke its `Exec()` method:

```
func main() {
	command := flag.String("command", "", "The git command")
	ignoreErros := flag.Bool(
		"ignore-errors",
		false,
		"Keep running after error if true")
	flag.Parse()

	// Get managed repos from environment variables
	root := os.Getenv("MG_ROOT")
	if root[len(root)-1] != '/' {
		root += "/"
	}

	repoNames := []string{}
	if len(os.Getenv("MG_REPOS")) > 0 {
		repoNames = strings.Split(os.Getenv("MG_REPOS"), ",")
	}

	repoManager, err := repo_manager.NewRepoManager(root, repoNames, *ignoreErros)
	if err != nil {
		log.Fatal(err)
	}

	output, err := repoManager.Exec(*command)
	if err != nil {
		fmt.Printf("command '%s' failed with error ", err)
	}

	for repo, out := range output {
		fmt.Printf("[%s]: git %s\n", path.Base(repo), *command)
		fmt.Println(out)
	}
}
```

This code needs to go into the root command. Very minimal changes are necessary. Instead of reading the flag `--command`, the root command will get the git command-line arguments as a slice of arguments. Here is the `Run()` method of the root command with the necessary adjustments:

```
Run: func(cmd *cobra.Command, args []string) {
	// Get managed repos from environment variables
	root := os.Getenv("MG_ROOT")
	if root[len(root)-1] != '/' {
		root += "/"
	}

	repoNames := []string{}
	if len(os.Getenv("MG_REPOS")) > 0 {
		repoNames = strings.Split(os.Getenv("MG_REPOS"), ",")
	}

	repoManager, err := repo_manager.NewRepoManager(root, repoNames, ignoreErrors)
	if err != nil {
		log.Fatal(err)
	}

	command := strings.Join(args, " ")
	output, err := repoManager.Exec(command)
	if err != nil {
		fmt.Printf("command '%s' failed with error ", err)
	}

	for repo, out := range output {
		fmt.Printf("[%s]: git %s\n", path.Base(repo), command)
		fmt.Println(out)
	}
},
```

# Providing the `--ignore-errors` Flag

The last piece of the puzzle is adding the `--ignore-errors` flag. The best place to do this is in an init() function:

```
func init() {
	rootCmd.Flags().BoolVar(
		&ignoreErrors,
		"ignore-errors",
		false,
		`will continue executing the command for all repos if ignore-errors
                 is true otherwise it will stop execution when an error occurs`)	
}
```

In the next lesson, we will refactor the main.go file to use our Cobra root command and finalize the refactoring.



In this lesson, we will do the heavy lifting. First, we'll create a Cobra root command and then port the logic in the current main.go of multi-git into the root command.

Refactoring Multi-git to use Cobra - The Root Command

Overview

Restructuring multi-git

Testing 1,2,3...

Adding Smart Command-Line Parsing

Adding Advanced Configuration

Adding Bells and Whistles

Summary

Refactoring Multi-git to use Cobra - The Root Command

Adding the root command