Grav comes with a built-in command-line interface (CLI) which can be found at
bin/grav. The CLI is extremely useful for running recurring tasks such as clearing the cache, making backups, and more.
Accessing the CLI is a simple process but you need to use a terminal. On a mac this is called
Terminal, on windows its called
cmd and on Linux its just a shell. UNIX style commands are not natively available in Windows cmd. Installing the msysgit package on a Windows machine adds Git and Git BASH, which is an alternative command prompt that makes UNIX commands available. If you are accessing your server remotely, you most likely will use SSH to remotely log in to your server. Check out this great tutorial for more information on SSH.
Although some operations can be performed manually, by relying on the CLI, these tasks could be automated via cron-jobs that run daily.
To get a list of all the commands available in Grav, you can run the command:
$ bin/grav list
To get help for a specific command, you can prepend help to the command:
$ bin/grav help install
To determine if your server is running
cgi-fcgi on the command line, type the following:
$ php -v PHP 5.5.17 (cgi-fcgi) (built: Sep 19 2014 09:49:55) Copyright (c) 1997-2014 The PHP Group Zend Engine v2.5.0, Copyright (c) 1998-2014 Zend Technologies with the ionCube PHP Loader v4.6.1, Copyright (c) 2002-2014, by ionCube Ltd.
If you see a reference to
(cgi-fcgi) you will need to prefix all
bin/grav commands with
php-cli. Alternatively you can setup an alias in your shell with something like:
alias php="php-cli" which will ensure the CLI verison of PHP runs from the command line.
Every time you want to start a new project with Grav, you need to start with a clean Grav instance. Through the CLI, this process is super easy and takes only few seconds.
Launch a terminal or console and navigate to the grav folder (for the sake of this document we will assume it resides under
$ cd ~/Projects/grav
Run the Grav CLI to create a new project, with the destination being the location where your project will reside in (usually the webroot of your Web server). Let us assume we are creating a portfolio and we want it at
$ bin/grav new-project ~/webroot/portfolio
This will create a new Grav instance and download all the dependencies required.
To install the dependencies Grav relies on (error plugin, problems plugin, antimatter theme), launch a terminal or console and navigate to the grav folder where you want to install the dependencies and run the CLI command.
$ cd ~/webroot/my-grav-project $ bin/grav install
You should now have the dependencies installed under:
You can clear the cache by deleting all the files and folders under
The equivalent CLI command is:
$ cd ~/webroot/my-grav-project $ bin/grav clear-cache
There are several options for this command. They include:
Backing up your project is nothing more than creating an archive of the ROOT of Grav. No Database, no complications.
Of course you can simplify this even more by just using the Grav CLI. Supposing we have our
~/workspace/portfolio project and that we want to create a backup of it, here's what we will do:
$ cd ~/workspace/portfolio $ bin/grav backup
A new backup
portfolio-20140812174352.zip file has been created at the
backup/ folder of the project. The long number after the name is just the date in the format of year month day hour minute second.
If you installed Grav via GitHub and have manually installed composer-based vendor packages, you can easily update with:
$ bin/grav composer