View Source Ash.Seed (ash v3.4.36)
Helpers for seeding data, useful for quickly creating lots of data either for database seeding or testing.
Important: this bypasses resource actions, and goes straight to the data layer. No action changes or validations are run. The only thing that it does at the moment is ensure that default values for attributes are set, it does not validate that required attributes are set (although the data layer may do that for you, e.g with ash_postgres).
Summary
Functions
Returns :__keep_nil__, allowing to ensure a default value is not used when you want the value to be nil.
Seed using a record (instance of a resource) as input.
Performs a direct call to the data layer of a resource with the provided input.
Performs an upsert operation on the data layer of a resource with the provided input and identities.
The usage is the same as seed!/1, but it will update the record if it already exists.
Usage is the same as seed!/2, but it will update the record if it already exists based on the identities.
Functions
Returns :__keep_nil__, allowing to ensure a default value is not used when you want the value to be nil.
Seed using a record (instance of a resource) as input.
If the passed in struct was retrieved from the data layer already (i.e already seeded),
then it is returned and nothing is done. Otherwise, the attributes and relationships are
used as input to seed/2, after having any %Ash.NotLoaded{} values stripped out.
Any nil values will be overwritten with their default values. To avoid this, either use seed/2
in which providing the key will have it not set the default values.
If you want to force nil to be accepted and prevent the default value from being set, use the
keep_nil/0 function provided here, which returns :__keep_nil__. Alternatively, use
seed!(Post, %{text: nil}).
See seed!/2 for more information.
Performs a direct call to the data layer of a resource with the provided input.
If a list is provided as input, then you will get back that many results.
To set a tenant, use the tenant option.
Performs an upsert operation on the data layer of a resource with the provided input and identities.
The usage is the same as seed!/1, but it will update the record if it already exists.
Ash.Seed.upsert!(%User{email: 'test@gmail.com', name: 'Test'}, identity: :email)Usage is the same as seed!/2, but it will update the record if it already exists based on the identities.