#9 Clean code, add validator outline

internal/ddosml/ml.go

@@ -6,8 +6,8 @@ import (
 	"net/http"
 )
 
-type RequestRepository interface {
-	CreateRequest(ctx context.Context, req *Request) error
+type Validator interface {
+	ValidateRequest(ctx context.Context, req *http.Request) (bool, error)
 }
 
 type Request struct {
@@ -18,33 +18,27 @@ type Request struct {
 // DDOS protection.
 type DDOS struct {
 	// isEnabled is used in order to enable or disable the ddosml
-	isEnabled  bool
-	repository RequestRepository
+	isEnabled bool
+	validator Validator
 }
 
 // NewDDOS creates new ddos ML analyzer used to
 // analyze requests in order to find out if given
 // request is ddos attack or not.
-func NewDDOS(isEnabled bool, repository RequestRepository) *DDOS {
+func NewDDOS(isEnabled bool, validator Validator) *DDOS {
 	return &DDOS{
-		isEnabled:  isEnabled,
-		repository: repository,
+		isEnabled: isEnabled,
+		validator: validator,
 	}
 }
 
 // IsRequestSuspicious checks if given request is suspicious (and then saves it in the database in order to be
 // used in future evaluations of this validator)
 func (d *DDOS) IsRequestSuspicious(ctx context.Context, req *http.Request) (bool, error) {
-	ok, err := d.validateRequest(ctx, req)
+	ok, err := d.validator.ValidateRequest(ctx, req)
 	if err != nil {
-		return ok, fmt.Errorf("validate request: %w", err)
+		return ok, fmt.Errorf("validator validate request: %w", err)
 	}
 
 	return ok, nil
 }
-
-// validateRequest is used to validate if given request is ddos or not. The validation is based on the ML
-// model, which decides based on normal user decisions from the UI.
-func (d *DDOS) validateRequest(ctx context.Context, req *http.Request) (bool, error) {
-
-}

internal/ddosml/validator.go

@@ -1,8 +1,19 @@
 package ddosml
 
-// Validator is a core of the ddosml, it validates
+import "context"
+
+type RequestRepository interface {
+	CreateRequest(ctx context.Context, req *Request) error
+}
+
+// MLBasedModelValidator is a core of the ddosml, it validates
 // each request against trained model and from time to time,
 // it's retraining the model against new data, and also it
 // clears database.
-type Validator struct {
+type MLBasedModelValidator struct {
+	repository RequestRepository
+}
+
+func NewMlModelValidator(repository RequestRepository) *MLBasedModelValidator {
+	return &MLBasedModelValidator{repository: repository}
 }