# `Parent.Supervisor`
[🔗](https://github.com/sasa1977/parent/blob/0.13.0/lib/parent/supervisor.ex#L1)

Supervisor of child processes.

This module works similarly to callbackless supervisors started with `Supervisor.start_link/2`.
To start a supervisor and some children you can do the following:

    Parent.Supervisor.start_link([
      child_spec1,
      child_spec2,
      # ...
    ])

To install a parent supervisor in the supervision tree you can provide child specification in the
shape of `{Parent.Supervisor, {children, parent_options}}`.

To build a dedicate supervisor module you can do:

    defmodule MySupervisor do
      use Parent.Supervisor

      def start_link({children, options}),
        do: Parent.Supervisor.start_link(children, options)

      # ...
    end

And now, you can install this supervisor in the supervision tree by passing
`{MySupervisor, {child_specs, parent_options}}` as the child specification to the parent.

You can interact with the running supervisor using functions from the `Parent.Client` module.
Refer to the `Parent` module for detailed explanation of child specifications, parent options,
and behaviour of parent processes.

In case you need more flexibility, take a look at `Parent.GenServer`.

# `child_spec`

```elixir
@spec child_spec({[Parent.start_spec()], Parent.GenServer.options()}) ::
  Parent.child_spec()
```

Returns a specification to start this module under a supervisor.
See `Supervisor`.

# `start_link`

```elixir
@spec start_link([Parent.start_spec()], Parent.GenServer.options()) ::
  GenServer.on_start()
```

Starts the parent process.

This function returns only after all the children have been started. If a child fails to start,
the parent process will terminate all successfully started children, and then itself.

---

*Consult [api-reference.md](api-reference.md) for complete listing*