gleam/erlang/atom
Types
Atom is a special string-like data-type that is most commonly used for interfacing with code written in other BEAM languages such as Erlang and Elixir. It is preferable to define your own custom types to use instead of atoms where possible.
Atoms are not used much in typical Gleam code!
Creating atoms
We can create atoms with the the create
function, though we must be careful when doing so as atoms are never
garbage collected. If we create too many atoms (for example, if we convert
user input into atoms) we may hit the max limit of atoms and cause the
virtual machine to crash.
pub type Atom
Values
pub fn cast_from_dynamic(a: dynamic.Dynamic) -> Atom
pub fn create(a: String) -> Atom
Creates an atom from a string, inserting a new value into the virtual machine’s atom table if an atom does not already exist for the given string.
We must be careful when using this function as there is a limit to the number of atom that can fit in the virtual machine’s atom table. Never convert user input into atoms as filling the atom table will cause the virtual machine to crash!
pub fn decoder() -> decode.Decoder(Atom)
A dynamic decoder for atoms.
You almost certainly should not use this to work with externally defined functions. They return known types, so you should define the external functions with the correct types, defining wrapper functions in Erlang if the external types cannot be mapped directly onto Gleam types.
pub fn get(a: String) -> Result(Atom, Nil)
Finds an existing atom for the given string.
If no atom is found in the virtual machine’s atom table for the string then an error is returned.
pub fn to_dynamic(a: Atom) -> dynamic.Dynamic
Convert an atom to a dynamic value, throwing away the type information.
This may be useful for testing decoders.