mirror of
https://github.com/ash-project/ash_archival.git
synced 2024-09-21 05:52:56 +12:00
44 lines
1.2 KiB
Markdown
44 lines
1.2 KiB
Markdown
# Un-archiving
|
|
|
|
If you want to unarchive a resource that uses a base filter, you will need to define a separate resource that uses the same storage and has no base filter. The rest of this guide applies for folks who _aren't_ using a `base_filter`.
|
|
|
|
Un-archiving can be accomplished by creating a read action that is skipped, using `exclude_read_actions`. Then, you can create an update action that sets that attribute to `nil`. For example:
|
|
|
|
```elixir
|
|
archive do
|
|
...
|
|
exclude_read_actions :archived
|
|
end
|
|
|
|
actions do
|
|
read :archived do
|
|
filter expr(not is_nil(archived_at))
|
|
end
|
|
|
|
update :unarchive do
|
|
change set_attribute(:archived_at, nil)
|
|
# if an individual record is used to unarchive
|
|
# it must use the `archived` read action for its atomic upgrade
|
|
atomic_upgrade_with :archived
|
|
end
|
|
end
|
|
```
|
|
|
|
You could then do something like this:
|
|
|
|
```elixir
|
|
Resource
|
|
|> Ash.get!(id, action: :archived)
|
|
|> Ash.Changeset.for_update(:unarchive, %{})
|
|
|> Ash.update!()
|
|
```
|
|
|
|
More idiomatically, you would define a code interfaceon the domain, and call that:
|
|
|
|
```elixir
|
|
# to unarchive by `id`
|
|
Resource
|
|
|> Ash.Query.for_read(:archived, %{})
|
|
|> Ash.Query.filter(id == ^id)
|
|
|> Domain.unarchive!()
|
|
```
|