crux_cache

v0.2.0

crux_cache v0.2.0 Crux.Cache behaviour View Source

Behaviour all caches must implement. (Looking at custom ones you may want to write)

There are exceptions:

  • User cache:

    • Implement a me/1 function setting the own user id
    • A me/0 and me!/0 function getting the own user

  • Guild cache:

    • A bit more, you probably want to take a look at the code of the Crux.Cache.Guild module

      Custom caches should be put under a Crux.Cache.Provider. (Can be combined with default caches)

      Also worth a look:

  • Crux.Cache.None - A dummy Crux.Cache and Crux.Cache.Provider, not caching anything.

Link to this section Summary

Types

key()

Default caches are using Discord Snowflakes as identifiers

Callbacks

delete(id)

Deletes data from the cache by key

fetch(id)

Fetches data from the cache by key

fetch!(id)

Fetches data from the cache by key, raises if not found

insert(data)

Inserts data into the cache

start_link(args)

Used to start anything fitting under a supervision tree, like for example a GenServer, instructed with handling the cache

update(data)

Inserts data into the cache

Link to this section Types

Link to this type

key() View Source 
key() :: non_neg_integer()

Default caches are using Discord Snowflakes as identifiers.

Link to this section Callbacks

Link to this callback

delete(id) View Source 
delete(id :: key()) :: :ok

Deletes data from the cache by key.

Link to this callback

fetch(id) View Source 
fetch(id :: key()) :: {:ok, term()} | :error

Fetches data from the cache by key.

Link to this callback

fetch!(id) View Source 
fetch!(id :: identifier()) :: term() | no_return()

Fetches data from the cache by key, raises if not found.

Link to this callback

insert(data) View Source 
insert(data :: term()) :: term()

Inserts data into the cache.

Returns the atomified data allowing the operation to be chained.

For example something like that:

id =
  raw_data
  |> Cache.insert()
  |> Map.get(:id)
Link to this callback

start_link(args) View Source (optional) 
start_link(args :: term()) :: Supervisor.on_start()

Used to start anything fitting under a supervision tree, like for example a GenServer, instructed with handling the cache.

Optional, you maybe want to use external caching, e.g. Redis, not requiring anything like that.

Link to this callback

update(data) View Source 
update(data :: term()) :: term()

Inserts data into the cache.

Returns "updated" data including changes by merging.
For example from a message embed update to a full message object

content =
  partial_message # only contains `:id`, `:channel_id`, and `:embeds`
  |> Cache.update()
  |> Map.get(:content) # present if the message was cached previously