It's a new implementation of cache based on ActiveSupport::Cache.
The implementation abstracts the cache in Adapter class on a
private method called cached_object, this method is intended
to be used on Adapters inside serializable_hash method in order
to cache each instance of the object that will be returned by
the serializer.
Some of its features are:
- A different syntax. (no longer need the cache_key method).
- An options argument that have the same arguments of ActiveSupport::Cache::Store, plus a key option that will be the prefix of the object cache on a pattern "#{key}-#{object.id}".
- It cache the objects individually and not the whole Serializer return, re-using it in different requests (as a show and a index method for example.)
When the requests asked for a nested attribute in the `include` and it's
missing or empty, don't break because the type of the object can't be
determined.
If the request is for a collection and none of the elements has the
attribute, it will not be added to the `linked` key, similar to what
happens with simple includes.
When linked resource had has_many links, all of them would use the
association from the first resource, causing all of the items to build
`links` with the same associations.
This fixes it by iterating over the serializers, not just the
attributes array.
If type is `author` but the association is called `writer`, the linked
resource key should be called `authors`, e.g
{
...
linked: {
authors: [{
...
}]
}
...
}
Currently, doing `include: author.bio` would work correctly, but not for
has_many associations such as `include: author.roles`. This fixes it.
The problem was basically that we were not handling arrays for has_many linked,
as happens for ArraySerializers.
The options passed to the render are partitioned into adapter options
and serializer options. 'include' and 'root' are sent to the adapter,
not sure what options would go directly to serializer, but leaving this
in until I understand that better.
