I have a struct
type mapKey string
var key1 mapKey = \"someKey\"
var key2 mapKey = \"anotherKey\"
type SampleMap map[mapKey]string
Incomi
There is no shortcut to coerce maps or arrays from one type to another, as there is for individual types (e.g. mapKey("str") ).
Setting the keys isn't hard though, you can just have a for loop:
params := map[string]string{"someKey": "bar"}
// Copy to type SampleMap
for k, v := range params {
m[mapKey(k)] = v
}
There isn't much point to having two extra types though (for key and map) unless you enforce limits in some way by using accessors, not allowing direct access. This feels like code translated from another language?
In the absence of other details, I would do this:
// These are the recognised key types for params
const (
key1 = "someKey"
key2 = "anotherKey"
)
// Work with this sort of map till you come to convert values:
// When checking keys or using them, use the constants above.
params map[string]string
myVal := params[key1]
What's the rationale for using two types here, to control which keys are used?