mkprj - Project Architecture Generator

Shell & JavaScript library to build files and folders architectures, defined in a Yaml config file and parsed with Node.js.

Sample

#### Sample architecture ####

# Dependencies #
    # Frontend package manager
bower.json: 0
bower_components: 0
    # Server package manager
package.json: 0
node_modules: 0
    # Installation
!install.sh: |
    bower install -save jquery
    npm install -save socket.io

# Documentation #
README.md: @README.md
Sample.todo: @blank.todo

# Root files #
index.html: @index.html     # Client index
index.js: 1                 # Server index

# Application #
app:
    Sample.js: 1

    library:
        jQuery.js: ~bower_components/jquery/jquery.min.js

    model:
        Machin.js: 1
        Chouette.js: 1

    view:
        style:
            reset.css: @style/reset.css
            theme.css: @style/theme.css
        template:
            Machin.html: 1

# End of architecture

Installation

• Node.js is required to use this library. http://nodejs.org/

• After having cloned or copied mkprj project, open a terminal and change directory to its root.

  cd /path/to/mkprj

  Then install the JS dependencies this way :

  ./install.sh

  This will install 3 libraries :

  • fs-extra : A powerful extension to the native file system manager fs. https://github.com/jprichardson/node-fs-extra
  • yamljs : The best Yaml file parser for Node.js. https://github.com/jeremyfa/yaml.js
  • colors : A great library to easily add colors to your terminal. https://github.com/Marak/colors.js

Adding alias command mkprj (facultative)

To add the command mkprj to execute this library, go to your session directory :

cd /home/myname

Then open a hidden file called .bash_profile or .bashrc and add this line at the bottom :

alias mkprj='/path/to/mkprj/mkprj.sh'

Re-open your terminal and check if the alias is correctly initialized :

mkprj dir

You should get an echo of the absolute path of the mkprj library.

Creating a new project architecture

Create a folder for your new project and go inside with a terminal.

cd /home/prj
mkdir patate
cd patate

Then just execute the shell script :

mkprj
# or if not aliased :
/path/to/mkprj/mkprj.sh

This will initialize your project with a Yaml configuration file called architecture.yml interactively chosen. The prompt will pause to let you edit the configuration file, and then will parse it and generate your project architecture with files and folders.

Setting architecture.yml

Why Yaml ?

Yaml files are easy and pleasant to read, besides Json files looks rougher. The key precedes :, and deepness is based on indentation (spaces or tabs). Thus it is important to have a clean configuration file.

config: 123
root:
    folder:
        a:
            A: file
        b:
            B1: file
            B2: file
    test:
        file: true

This looks quite nice to read, and it is useful for concepting your application architecture.

In JavaScript, after having parsed this example, you would get :

var result = yaml.parse(fs.readFileSync('architecture.yml', 'utf8')) ;

console.log(typeof result.config) ; // number
console.log(result.root.folder.a.A) ; // "file"
console.log(typeof result.root.test.file) ; // boolean

Note : This Yaml parser ables to have no unique key root, both config and root are key roots. It is not the case to some other Yaml parsers...

Managing files and folders

• Optional values

  Values given defines the action or the content of each element. Here is an exhaustive list of values :

  • 0 : Skips the element, like a facultative file/folder but useful to write in the architecture.

  • 1 true : Generates an empty file if it does not exist.

  • 2 : Generates a folder if it does not exist.

  • false : Deletes the file/folder if it exists.

  • > : Generates a file with the following as inline content.

  file: >
      This is inline content.

  • | : Generates a file with the following as block content.

  file: |
      This is content
      with multiple lines.

  • @path/to/file : Copy the file from mkprj modules mkprj/MODULES/path/to/file if the destination does not exist.

  This is useful when having previously wrote default scripts/templates/config stored in global modules.

  Example :

  reset.css: @style/reset.css
  theme.css: @style/default.css
  index.html: @template/default.html
  loader.js: @script/loader.js

  • ~path/to/file : Copy the file from the project's root /home/prj/patate/path/to/file if the destination does not exist.

  This is useful for instance when you use Bower. Bower is a client-side package manager, but when you install a package, it downloads the full repository which can be heavy. Copying a script inside your own project allows you to clone the only required script and put it in your application.

  Example :

  app:
      lib:
          jquery.js: ~/bower_components/jquery/jquery.min.js

  Applying for those who don't want to use heavy client web application managers such as Yeoman.

• Optional key prefixes

  You can also execute a pre-operation by adding a prefix to your key.

  • !key: : Delete the file before creating it anew.

  This is useful if you wish to write directly the content of the file in the architecture config (thus it will "replace" that file). But it is also useful to reset files such as logs.

  !init.sh: |
      rm ._*
      rm .DS_STORE
      rm .somethinghidden
  
  !bugs.log: 1

Note

ReadMe documentation in progress !