add documentation

README.md

@@ -1,19 +1,62 @@
 # alog
-alog (Append-only Log) is an easy way to start using the Lambda/Kappa architecture in your Elixir/Phoenix Apps while still using PostgreSQL (with Ecto).
 
-This module provides some helper functions to make it easy to insert and retrieve the data you need.
+alog is an [Hex package](https://hexdocs.pm/alog) (Elixir) which
+provides some helper functions to make it easy to insert and retrieve the data you need in your Elixir/Phoenix applications.
+
+Instead of using CRUD concept to manage data, alog makes sure to keep the data immutable in the database.
+This means the data is not updated nor deleted but instead a new version representing the new state of the data is inserted in the database.
+
+## Why
+
+- See all history of the data
+- Debugging is easier
+- Analytics is built-in
 
 ## Usage
 
-  At the top of the schema you wish to use append only functions for, `use` this module:
+### Install alog
 
-  ``` elixir
-  use Alog
+  Add alog to your package dependencies in the `mix.exs` file:
+
+  ```elixir
+  defp deps do
+    [{:alog, "~> 0.1"}]
+  end
   ```
 
-  The append only functions will then be available to call as part of your schema.
+  Update your dependencies with `mix deps.get`
+
+### Define your schemas
+
+  For alog to work the schemas need to have defined:
+  - `inserted_at` field. You can use the [timestamps/1](https://hexdocs.pm/ecto/Ecto.Schema.html#timestamps/1) functions.
+
+  Since Ecto v3.0 the type of `inserted_at` is by default `naive_datetime`. We recommand to change it to `naive_datetime_usec` to save the timestamps with a precision of microseconds:
+
+    - Update your `Repo` configuration
+
+    ```elixir
+    config :my_app, MyApp.Repo,
+    migration_timestamps: [type: :naive_datetime_usec]
+    ```
+
+    - Define the timestamps option in your schemas
 
-  ## Example
+    ```elixir
+    @timestamps_opts [type: :naive_datetime_usec]
+    ```
+
+    see the `Calendar type` secction of http://blog.plataformatec.com.br/2018/10/a-sneak-peek-at-ecto-3-0-breaking-changes/
+
+  - `entry_id` field with the type `string`.
+
+  alog will save a new version of an item each time its state is changed. To keep track of all the states we need to a unique id which represent an item(and its older versions). alog use [Ecto.UUID](https://hexdocs.pm/ecto/Ecto.UUID.html) to generate the `entry_id`
+
+  - `deleted` field with the type `boolean` default to `false`.
+
+  alog will not remove items from the database. Instead it represent a deletion with the `deleted` field.
+
+  Once your schemas are defined, at the top of the one you wish to use append only functions for, `use` this module:
 
   ``` elixir
   defmodule MyApp.User do
@@ -22,8 +65,10 @@ This module provides some helper functions to make it easy to insert and retriev
 
     import Ecto.Changeset
 
+    @timestamps_opts [type: :naive_datetime_usec]
     schema "users" do
       ...
+      timestamps()
     end
 
     def changeset(user, attrs) do
@@ -32,5 +77,22 @@ This module provides some helper functions to make it easy to insert and retriev
   end
   ```
 
+  The append only functions will then be available to call as part of your schema:
+
+    - `Users.get`
+    - `Users.get_by`
+    - `Users.insert`
+    - `Users.update`
+    - `Users.get_history`
+    - `Users.delete`
+    - `Users.preload`
+
+
   Alog expects your `Repo` to belong to the same base module as the schema.
-  So if your schema is `MyApp.User`, or `MyApp.Accounts.User`, your Repo should be `MyApp.Repo`.
+  So if your schema is `MyApp.User`, or `MyApp.Accounts.User`, your Repo should be `MyApp.Repo`.
+
+## Useful links
+
+  - A step by step guide on how to create append only applications [phoenix-ecto-append-only-log-example](https://github.com/dwyl/phoenix-ecto-append-only-log-example)
+
+  - [Lambda architecture](https://en.wikipedia.org/wiki/Lambda_architecture)