README.md

# Mnesiac

[![Build Status](https://travis-ci.org/beardedeagle/mnesiac.svg?branch=master)](https://travis-ci.org/beardedeagle/mnesiac) [![codecov](https://codecov.io/gh/beardedeagle/mnesiac/branch/master/graph/badge.svg)](https://codecov.io/gh/beardedeagle/mnesiac) [![Hex.pm](http://img.shields.io/hexpm/v/mnesiac.svg?style=flat)](https://hex.pm/packages/mnesiac) [![Hex.pm downloads](https://img.shields.io/hexpm/dt/mnesiac.svg?style=flat)](https://hex.pm/packages/mnesiac)

Mnesia autoclustering made easy!

Docs can be found at [https://hexdocs.pm/mnesiac](https://hexdocs.pm/mnesiac).

## Installation

Simply add `mnesiac` to your list of dependencies in `mix.exs`:

```elixir
def deps do
  [
    {:mnesiac, "~> 0.3"}
  ]
end
```

Edit your app's config.exs to add the list of mnesia stores:

```elixir
config :mnesiac,
  stores: [Mnesiac.ExampleStore, ...],
  schema_type: :disc_copies, # defaults to :ram_copies
  table_load_timeout: 600_000 # milliseconds, default is 600_000
```

Then add `mnesiac` to your supervision tree:

- With `libcluster` using the `Cluster.Strategy.Epmd` strategy:

```elixir
  ...

    topology = Application.get_env(:libcluster, :topologies)
    hosts = topology[:myapp][:config][:hosts]

    children = [
      {Cluster.Supervisor, [topology, [name: MyApp.ClusterSupervisor]]},
      {Mnesiac.Supervisor, [hosts, [name: MyApp.MnesiacSupervisor]]},
      ...
    ]

  ...
```

- Without `libcluster`:

```elixir
  ...

    children = [
      {
        Mnesiac.Supervisor,
        [
          [:"test01@127.0.0.1", :"test02@127.0.0.1"],
          [name: MyApp.MnesiacSupervisor]
        ]
      },
      ...
    ]

  ...
```

## Usage

### Table creation

Create a table store, `use Mnesiac.Store`, and add it to your app's config.exs. 

All stores *MUST* implement its own `store_options/0`, which returns a keyword list of table options.

There are three optional callbacks which can be implemented:

- `init_store/0`, which allows users to implement custom table initialisation logic.
- `copy_store/0`, which allows users to implement a custom call to copy a store.
- `resolve_conflict/1`, which allows a user to implement logic when table data is found on both the remote node and local node when connecting to a cluster. This currently has no default implementation.

```elixir
defmodule MyApp.ExampleStore do
  @moduledoc false
  require Record

  use Mnesiac.Store

  Record.defrecord(
    :example,
    __MODULE__,
    id: nil,
    topic_id: nil,
    event: nil
  )

  @type example ::
          record(
            :example,
            id: String.t(),
            topic_id: String.t(),
            event: String.t()
          )

  @impl true
  def store_options,
    do: [
      attributes: example() |> example() |> Keyword.keys(),
      index: [:topic_id],
      ram_copies: [node()]
    ]
end
```

### Clustering

If you are using `libcluster` or another clustering library just ensure that clustering library starts earlier than `mnesiac`. That's all, you don't need to do anything else.

If you are not using `libcluster` or similar clustering library then:

- When a node joins to an erlang/elixir cluster, run the `Mnesiac.init_mnesia()` function on the *new node*. This will initialize and copy table contents from the other online nodes.

## Development

Ensure you have the proper language versions installed. To do this, an `asdf` tools file is provided. Run the following:

```shell
git clone https://github.com/beardedeagle/mnesiac.git
git checkout -b MyFeature
asdf install
mix local.hex --force
mix local.rebar --force
mix deps.get --force
mix deps.compile --force
mix compile --force
mix check
```

**_NOTICE:_** You can find the `asdf` tool [here][1].

## Testing

Before you run any tests, ensure that you have cleaned up mnesia:

```shell
mix purge.db
```

Test results and coverage reports are generated by running the following:

```shell
mix coveralls.html --trace --slowest 10 --no-start
```

## Notice

This library was built standing on the shoulders of giants. A big thanks goes out to Mustafa Turan. The original library this was forked from can be found here: <https://github.com/mustafaturan/mnesiam>.

Happy coding!

[1]: https://github.com/asdf-vm/asdf