Fixed an NPE when parsing malformed subjects (#1291)

Author: bashir2

pipelines/batch/src/main/java/com/google/fhir/analytics/FetchResources.java

@@ -1,5 +1,5 @@
 /*
- * Copyright 2020-2024 Google LLC
+ * Copyright 2020-2025 Google LLC
  *
  * Licensed under the Apache License, Version 2.0 (the "License");
  * you may not use this file except in compliance with the License.
@@ -68,20 +68,16 @@ static String getSubjectPatientIdOrNull(Resource resource) {
     Property subject = resource.getNamedProperty("subject");
     if (subject != null) {
       List<Base> values = subject.getValues();
-      if (values.size() == 1) {
+      if (values.size() == 1 && values.get(0) instanceof Reference) {
         Reference reference = (Reference) values.get(0);
         // TODO: Find a more generic way to check if this is a reference to a Patient. With the
         // current OpenMRS setup, reference.getType() is null so we cannot rely on that.
         String refStr = reference.getReference();
-        Matcher matcher = PATIENT_REFERENCE.matcher(refStr);
-        if (matcher.matches()) {
-          patientId = matcher.group(1);
-        }
-        if (patientId == null) {
-          log.warn(
-              String.format(
-                  "Ignoring subject of %s with id %s because it is not a Patient reference: %s",
-                  resource.getResourceType(), resource.getId(), refStr));
+        if (refStr != null) {
+          Matcher matcher = PATIENT_REFERENCE.matcher(refStr);
+          if (matcher.matches()) {
+            patientId = matcher.group(1);
+          }
         }
       }
       if (values.size() > 1) {
@@ -91,6 +87,9 @@ static String getSubjectPatientIdOrNull(Resource resource) {
                 resource.getResourceType(), resource.getId()));
       }
     }
+    if (patientId == null) {
+      log.warn("Ignoring subject of {} with id {}", resource.getResourceType(), resource.getId());
+    }
     return patientId;
   }
 

pipelines/batch/src/test/java/com/google/fhir/analytics/FetchResourcesTest.java

@@ -1,5 +1,5 @@
 /*
- * Copyright 2020-2024 Google LLC
+ * Copyright 2020-2025 Google LLC
  *
  * Licensed under the Apache License, Version 2.0 (the "License");
  * you may not use this file except in compliance with the License.
@@ -39,6 +39,7 @@
 import org.apache.beam.sdk.transforms.Create;
 import org.apache.beam.sdk.values.KV;
 import org.apache.beam.sdk.values.PCollection;
+import org.hamcrest.core.IsNull;
 import org.hl7.fhir.r4.model.Bundle;
 import org.hl7.fhir.r4.model.Observation;
 import org.junit.Before;
@@ -86,6 +87,15 @@ public void testGetPatientId() throws IOException {
     assertThat(patientId, equalTo(expectedId));
   }
 
+  @Test
+  public void testGetPatientIdBadSubject() throws IOException {
+    URL url = Resources.getResource("observation_bad_subject.json");
+    String obsStr = Resources.toString(url, StandardCharsets.UTF_8);
+    Observation observation = parser.parseResource(Observation.class, obsStr);
+    String patientId = FetchResources.getSubjectPatientIdOrNull(observation);
+    assertThat(patientId, IsNull.nullValue());
+  }
+
   @Test
   public void testPatientIdsFromBundle() {
     List<SearchSegmentDescriptor> searchSegments =

pipelines/batch/src/test/resources/observation_bad_subject.json

@@ -0,0 +1,42 @@
+{
+  "resourceType": "Observation",
+  "id": "0adb6cfb-0ae3-42df-9d39-fc68ec4334cd",
+  "meta": {
+    "tag": [
+      {
+        "system": "http://hl7.org/fhir/v3/ObservationValue",
+        "code": "SUBSETTED",
+        "display": "Resource encoded in summary mode"
+      }
+    ]
+  },
+  "status": "final",
+  "code": {
+    "coding": [
+      {
+        "code": "162169AAAAAAAAAAAAAAAAAAAAAAAAAAAAAA",
+        "display": "Text of encounter note"
+      }
+    ]
+  },
+  "subject": {
+    "identifier": {
+      "type": {
+        "coding": [ {
+          "system": "http://terminology.hl7.org/CodeSystem/v2-0203",
+          "code": "NI",
+          "display": "National unique individual identifier"
+        } ]
+      },
+      "system": "http://some.system.com",
+      "value": "somevalue"
+    }
+  },
+  "context": {
+    "reference": "Encounter/2df7151b-fdde-4d1b-8dfa-aeaae4bb0430"
+  },
+  "effectiveDateTime": "2020-09-22T13:54:41-04:00",
+  "issued": "2020-09-22T13:54:41.000-04:00",
+  "valueString": "Adding a test diagnosis"
+}
+