My First Erlang Application

This is my first experience with Erlang. I've tried to build an event-sourced application with mysql as eventstore.

What did I do:

I took a project from Brian Hunter (https://github.com/bryanhunter/cqrs-with-erlang/tree/ndc-oslo) which was a part of this talk https://vimeo.com/97318824. It looked like a good base to make changes on so I...

• Swapped ETS to MySQL as eventstore to make my events persistent. (This was actually a lot work)

• Stored all events as JSON in MySQL. This is actually an ongoing expiriment. I found out that the "Record" data type didn't work well for me when retrieving records from the eventstore (My n00b erlang skills can be the problem as well!). The other reason is that I may want to create different read applications which will deal with incoming events. To publish those events and make them directly usable for example a php application it sounded like a good idea to store everything in JSON. I'm not convinced yet so this might be change in the future.

• Created a simple projection

• Swapped context. It's not about bankaccounts anymore. It's a about a blog and its drafts.

Feedback

What I need now is feedback, feedback to upgrade my erlang skills.

• One of the things I noticed was that if I want to add a new event I have to change code on many places. Is this normal?
• Maybe someone can point me to great resources where I can learn from.
• Can I make better use of pattern matching?
• Is my code (not) functional enough?
• etc all kinds of tips & tricks are welcome!

Getting started

Setup Mysql:

CREATE TABLE `events` (
  `id` int(10) unsigned NOT NULL AUTO_INCREMENT,
  `uuid` varchar(36) NOT NULL,
  `event` mediumtext NOT NULL,
  `event_name` varchar(250) DEFAULT NULL,
  `processed_date` datetime NOT NULL,
  PRIMARY KEY (`id`)
) ENGINE=InnoDB AUTO_INCREMENT=114 DEFAULT CHARSET=latin1; 

CREATE TABLE `refinements_per_blogitem` (
  `id` int(10) unsigned NOT NULL AUTO_INCREMENT,
  `uuid` varchar(36) NOT NULL,
  `count` int(11) NOT NULL,
  PRIMARY KEY (`id`)
) ENGINE=InnoDB AUTO_INCREMENT=7 DEFAULT CHARSET=latin1;

The project:

git clone https://github.com/weemen/erlang-eventsourcing.git
rebar get-deps
rebar compile
rebar shell
application:start(emysql).
emysql:add_pool(erlang_es_blog, [{size,1},
                     {user,"root"},
                     {password,"root"},
                     {database,"erlang_es"},
                     {encoding,utf8}]).
application:start(blog).

The commands you can execute, note uuids are different in your application!

blog:make_new_draft().
blog:refine_title_of_draft("21d84c10-30b0-4bf6-9c3d-b308c8c15afc","some title 27").
blog:refine_content_of_draft("21d84c10-30b0-4bf6-9c3d-b308c8c15afc","this is my content").
blog:publish_draft("21d84c10-30b0-4bf6-9c3d-b308c8c15afc").
blog:hide_draft("21d84c10-30b0-4bf6-9c3d-b308c8c15afc").
blog:renew_draft("21d84c10-30b0-4bf6-9c3d-b308c8c15afc").

Things still todo

• Create more projections
• Separate the projections from this application (by publishing events over something like rabbitmq)
• Write tests
• Simplefy code and remove the odd stuff.
• There is no real playhead functionality so events might still collide with each other.
• Mysql should be started out of the box
• Create decent log message instead of io:fwrite.
• Create something like a db migration file