# `Licet.Schema`
[🔗](https://github.com/netoum/licet/blob/v0.1.0/lib/licet/schema.ex#L1)

Parses schemas written in SpiceDB's schema language.

A running instance parses its schema itself, from the `:schema` option or
`Licet.write_schema/3`. Call `parse/1` directly to check a schema file, for
example in a test or before a deploy.

Licet supports definitions, relations, permissions, the operators `+`, `&`,
`-`, and `->`, the arrow functions `.any()` and `.all()`, `nil`, wildcards,
userset subjects such as `team#member`, and expiration. Caveats and `self`
are rejected. [Schema language](zed.md) is the full reference.

# `t`

```elixir
@type t() :: %{
  definitions: map(),
  expiration: boolean(),
  source: String.t(),
  hash: String.t()
}
```

# `parse`

```elixir
@spec parse(String.t()) :: {:ok, t()} | {:error, Licet.Error.t()}
```

Parses and validates schema text.

On success, returns a compiled schema. Its fields other than `:source` are
internal. On failure, returns an `:invalid_schema` error whose `details`
hold the `:line` and `:column` of the problem.

## Examples

    iex> {:ok, schema} =
    ...>   Licet.Schema.parse("""
    ...>   definition user {}
    ...>
    ...>   definition document {
    ...>     relation owner: user
    ...>     permission edit = owner
    ...>   }
    ...>   """)
    iex> schema.definitions |> Map.keys() |> Enum.sort()
    ["document", "user"]

    iex> {:error, error} = Licet.Schema.parse("definition document {\n  permission edit = owner\n}")
    iex> error.reason
    :invalid_schema
    iex> error.details
    %{line: 2, column: 21}

# `to_zed`

```elixir
@spec to_zed(t()) :: String.t()
```

Prints a compiled schema as text.

The result parses to the same definitions. Comments and formatting from the
original text aren't preserved; definitions are sorted by name.

## Examples

    iex> {:ok, schema} = Licet.Schema.parse("definition user {}")
    iex> Licet.Schema.to_zed(schema)
    "definition user {\n}"

---

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