Erlang Automake is designed to remove the burden of writing and maintaining pure Erlang "Emakefile" manually. It also offers several other features whose purpose is to ease the creation of different build environment for your Erlang code.

1 Contents

• Introduction
• Running automake
• Eautomake file
• Running Erlang make from Emacs

2 Introduction

Erlang Automake is designed to remove the burden of writing and maintaining pure Erlang "Emakefile" manually. It also offers several other features whose purpose is to ease the creation of different build environment for your Erlang code.

3 Running automake

Simply type erl -s automake in your project directory and the Emakefile project file will be create automatically.

4 Advanced features

Erlang Automake can also create fresh build environment to help in your multi-package Erlang development.

One of the major use is the following:

• I am working on a project involving code from several repositories (For example, ejabberd, additional customer ejabberd plugins and Eunit and Quickcheck tests modules.
  
• I want to easily set up the environment for such a development. I do not want to spend time setting paths, preparing shell scripts to make sure everything is
  
• I want to be able to rebuild all changed code and reload it easily from the command-line.
  
• I want to be able to rebuild and reload a file directly from Emacs with Erlang mode.
  

This type of environment can be set up in 3 minutes with Erlang Automake. It does not implies that you have to stop writing "Makefile"s. For example, ejabberd still has its own build chain to compile Erlang and C files. However, once your C files are stable you can simply setup new build environment easily with Erlang Automake.

Here is how:

• Create an "Eautomake" file containing the path to all your Erlang applications needed for this project. Applications does not have to have the same structure. No Makefile or Emakefile is required as long as your code is pure Erlang.
• Setup your environment variables (for example ejabberd needs the path to its C library (EJABBERD_SO_PATH), path to its message files, etc). If a project contain C code, you thus have to build it separately in your path.
• type "erl -s automake" from the command-line in your project environment. Erlang automake will build an Emakefile allowing to rebuild all the Erlang module of your applications at once.
• type "erl -make" from the command-line in your project environment to rebuild your beam files from Erlang sources. As a default all beam files will be placed and updated in an ebin subdirectory.
• type "erl -pa ebin" to start Erlang, pointing to the default path containing all beam files. Your environment is now set up.
• if you have an Erlang shell already opened and running your project, you can reload the new code with the command: automake:reload().