Refactoring ResourceValidationTracker

org.hl7.fhir.validation/src/main/java/org/hl7/fhir/r5/validation/instancevalidator/utils/ResourceValidationTracker.java

@@ -0,0 +1,44 @@
+package org.hl7.fhir.r5.validation.instancevalidator.utils;
+
+import org.hl7.fhir.r5.model.StructureDefinition;
+import org.hl7.fhir.utilities.validation.ValidationMessage;
+
+import java.util.ArrayList;
+import java.util.HashMap;
+import java.util.List;
+import java.util.Map;
+
+/**
+ * The validator keeps one of these classes for each resource it validates (e.g. when chasing references)
+ * <p>
+ * It keeps track of the state of resource validation - a given resrouce may be validated agaisnt multiple
+ * profiles during a single validation, and it may be valid against some, and invalid against others.
+ * <p>
+ * We don't want to keep doing the same validation, so we remember the outcomes for each combination
+ * of resource + profile
+ * <p>
+ * Additionally, profile validation may be circular - e.g. a Patient profile that applies the same profile
+ * to linked patients, and 2 patient resources that link to each other. So this class also tracks validation
+ * in process.
+ * <p>
+ * If the validator comes back to the same point, and validates the same instance against the same profile
+ * while it's already validating, it assumes it's valid - it cannot have new errors that wouldn't already
+ * be found in the first iteration - in other words, there's no errors
+ *
+ * @author graha
+ */
+public class ResourceValidationTracker {
+    private Map<String, List<ValidationMessage>> validations = new HashMap<>();
+
+    public void startValidating(StructureDefinition sd) {
+        validations.put(sd.getUrl(), new ArrayList<ValidationMessage>());
+    }
+
+    public List<ValidationMessage> getOutcomes(StructureDefinition sd) {
+        return validations.get(sd.getUrl());
+    }
+
+    public void storeOutcomes(StructureDefinition sd, List<ValidationMessage> errors) {
+        validations.put(sd.getUrl(), errors);
+    }
+}