Hi all!
I need an gopher advice. What is more idiomatic and do you see any problems
with my solutions?
Problem: I have a repository(persistence layer) for TodoLists. TodoLists
usually belongs to a user. My system also have super-user/admin interface
which can access all TodoLists. I can imagine something like this. But I
think it is maybe not good, because mixing not user-scoped API and user
scoped API
type TodoListID int
type TodoList struct {}
type TodoListRepository interface {
Add(t *TodoList) error
Update(t *TodoList) error
Remove(id TodoListID) error
All() ([]*TodoList, error)
Find(id TodoListID) (* TodoList, error)
// user scoped API
TodoListsForUserID(id UserID) ([]*TodoList, error)
AddTodo(t *Todo, u UserID)
RemoveTodo(t *Todo, u UserID)
}
So I got this idea: Separate the interfaces
type TodoListRepository interface {
Add(t *TodoList) error
Update(t *TodoList) error
Remove(id TodoListID) error
All() ([]*TodoList, error)
Find(id TodoListID) (* TodoList, error)
}
type UserTodoListsRepository interface {
All() ([]*TodoList, error)
AddTodo(t *Todo)
RemoveTodo(t *Todo)
}
So I would define a struct wich implements UserTodoListsRepository
interface and one struct for not user-scoped TodoListsRepository. Do I miss
something? Bad or good design?
Thanks all in advance!
--
You received this message because you are subscribed to the Google Groups
"golang-nuts" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to golang-nuts+unsubscr...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.
