uc-cdis · mfshao · Sep 18, 2025 · Jul 25, 2025 · Aug 5, 2025 · Aug 12, 2025
diff --git a/devHelper/scripts/commands.sh b/devHelper/scripts/commands.sh
@@ -80,6 +80,14 @@ curl -iv -X PUT "${ESHOST}/${indexName}" \
                 "days_to_follow_up": { "type": "integer" },
                 "follow_up_label": { "type": "keyword", "fields": { "analyzed": {"type": "text", "analyzer": "ngram_analyzer", "search_analyzer": "search_analyzer", "term_vector": "with_positions_offsets"} } }
               }
+            },
+            "treatments_array_under_subjects": {
+              "type": "nested",
+              "properties": {
+                "dose_amount": { "type": "integer" },
+                "test_article_name_array_under_treatments": { "type": "keyword", "fields": { "analyzed": {"type": "text", "analyzer": "ngram_analyzer", "search_analyzer": "search_analyzer", "term_vector": "with_positions_offsets"} } },
+                "submitter_id": { "type": "keyword", "fields": { "analyzed": {"type": "text", "analyzer": "ngram_analyzer", "search_analyzer": "search_analyzer", "term_vector": "with_positions_offsets"} } }
+              }
             }
           }
         },

diff --git a/genData/genData.js b/genData/genData.js
@@ -153,6 +153,7 @@ async function run() {
   });
   const { body: count } = await client.count({ index: esIndex });
   console.log(count);
+  console.log(arrayFields);
 
   if (configIndex) {
     const data = [

diff --git a/genData/types.js b/genData/types.js
@@ -1,4 +1,4 @@
-function fakerType(key, value, arrayFields) {
+function fakerType(key, value, arrayFields, nestedFieldKeys = []) {
   let fieldType;
   const properties = {};
   const required = [];
@@ -10,7 +10,7 @@ function fakerType(key, value, arrayFields) {
         fieldType = {
           type: 'array', items: { type: 'boolean', properties, required }, minItems: 0, maxItems: 10,
         };
-        arrayFields.push(key);
+        arrayFields.push(`${nestedFieldKeys.join('.')}.${key}`);
       } else {
         fieldType = { type: 'boolean' };
       }
@@ -26,7 +26,7 @@ function fakerType(key, value, arrayFields) {
           minItems: 0,
           maxItems: 10,
         };
-        arrayFields.push(key);
+        arrayFields.push(`${nestedFieldKeys.join('.')}.${key}`);
       } else {
         fieldType = { type: 'string', faker: 'name.findName' };
       }
@@ -37,7 +37,7 @@ function fakerType(key, value, arrayFields) {
         fieldType = {
           type: 'array', items: { type: 'number', properties, required }, minItems: 0, maxItems: 10,
         };
-        arrayFields.push(key);
+        arrayFields.push(`${nestedFieldKeys.join('.')}.${key}`);
       } else {
         fieldType = { type: 'number' };
       }
@@ -48,21 +48,27 @@ function fakerType(key, value, arrayFields) {
         fieldType = {
           type: 'array', items: { type: 'integer', properties, required }, minItems: 0, maxItems: 10,
         };
-        arrayFields.push(key);
+        arrayFields.push(`${nestedFieldKeys.join('.')}.${key}`);
       } else {
         fieldType = { type: 'integer' };
       }
       break;
     case 'nested':
+      nestedFieldKeys.push(key);
       Object.entries(value.properties).forEach(([k, v]) => {
-        properties[k] = fakerType(k, v);
+        properties[k] = fakerType(k, v, arrayFields, [...nestedFieldKeys]);
         required.push(k);
       });
       fieldType = {
         type: 'array', items: { type: 'object', properties, required }, minItems: 0, maxItems: 10,
       };
       if (key.includes('array')) {
-        arrayFields.push(key);
+        // since we already pushed the current key into this array
+        if (nestedFieldKeys.length) {
+          arrayFields.push(nestedFieldKeys.join('.'));
+        } else {
+          arrayFields.push(key);
+        }
       }
       break;
     default:

diff --git a/genData/valueBank.json b/genData/valueBank.json
@@ -1,72 +1,156 @@
 {
-    "gender": ["male", "female", "unknown"],
-    "ethnicity": ["American Indian", "Pacific Islander", "Black", "Multi-racial", "White", "Haspanic" ],
-    "race": ["white", "black", "hispanic", "asian", "mixed", "not reported" ],
-    "vital_status": ["Alive", "Dead", "no data" ],
-    "file_type": ["mRNA Array", "Unaligned Reads", "Lipdomic MS", "Protionic MS", "1Gs Ribosomes", "Unknown" ],
-    "file_format": ["BEM", "BAM", "BED", "CSV", "FASTQ", "RAW", "TAR", "TSV", "TXT", "IDAT" ],
-    "auth_resource_path": ["/programs/jnkns/projects/jenkins", "/programs/DEV/projects/test", "/programs/external/projects/test"],
-    "sensitive": [ "true", "false" ],
-    "study": ["study_1", "study_2", "study_3"],
-    "file_id": ["file_id_1", "file_id_2", "file_id_3"],
-    "subject_id": ["subject_id_1", "subject_id_2", "subject_id_3"],
-    "project": ["jnkns-jenkins", "DEV-test", "external-test" ],
+    "gender": [
+        "male",
+        "female",
+        "unknown"
+    ],
+    "ethnicity": [
+        "American Indian",
+        "Pacific Islander",
+        "Black",
+        "Multi-racial",
+        "White",
+        "Haspanic"
+    ],
+    "race": [
+        "white",
+        "black",
+        "hispanic",
+        "asian",
+        "mixed",
+        "not reported"
+    ],
+    "vital_status": [
+        "Alive",
+        "Dead",
+        "no data"
+    ],
+    "file_type": [
+        "mRNA Array",
+        "Unaligned Reads",
+        "Lipdomic MS",
+        "Protionic MS",
+        "1Gs Ribosomes",
+        "Unknown"
+    ],
+    "file_format": [
+        "BEM",
+        "BAM",
+        "BED",
+        "CSV",
+        "FASTQ",
+        "RAW",
+        "TAR",
+        "TSV",
+        "TXT",
+        "IDAT"
+    ],
+    "auth_resource_path": [
+        "/programs/jnkns/projects/jenkins",
+        "/programs/DEV/projects/test",
+        "/programs/external/projects/test"
+    ],
+    "sensitive": [
+        "true",
+        "false"
+    ],
+    "study": [
+        "study_1",
+        "study_2",
+        "study_3"
+    ],
+    "file_id": [
+        "file_id_1",
+        "file_id_2",
+        "file_id_3"
+    ],
+    "subject_id": [
+        "subject_id_1",
+        "subject_id_2",
+        "subject_id_3"
+    ],
+    "project": [
+        "jnkns-jenkins",
+        "DEV-test",
+        "external-test"
+    ],
     "visits": [
         {
             "days_to_visit": 1,
             "visit_label": "vst_lbl_1",
             "follow_ups": {
                 "days_to_follow_up": 1,
                 "follow_up_label": "flup_lbl_1"
-            }
+            },
+            "treatments_array_under_subjects": [
+                {
+                    "submitter_id": "treatment_1",
+                    "dose_amount": 1,
+                    "test_article_name_array_under_treatments": ["article_1a", "article_1b"]
+                }
+            ]
         },
         {
             "days_to_visit": 2,
             "visit_label": "vst_lbl_2",
             "follow_ups": {
                 "days_to_follow_up": 2,
                 "follow_up_label": "flup_lbl_2"
-            }
+            },
+            "treatments_array_under_subjects": [
+                {
+                    "submitter_id": "treatment_2",
+                    "dose_amount": 2,
+                    "test_article_name_array_under_treatments": ["article_2a", "article_2b"]
+                }
+            ]
         },
         {
             "days_to_visit": 3,
             "visit_label": "vst_lbl_3",
             "follow_ups": {
                 "days_to_follow_up": 3,
                 "follow_up_label": "flup_lbl_3"
-            }
+            },
+            "treatments_array_under_subjects": [
+                {
+                    "submitter_id": "treatment_3",
+                    "dose_amount": 3,
+                    "test_article_name_array_under_treatments": ["article_3a", "article_3b"]
+                }
+            ]
         }
     ],
     "some_nested_array_field": [
         [
             {
-              "some_integer_inside_nested": 1,
-              "some_string_inside_nested": "first"
+                "some_integer_inside_nested": 1,
+                "some_string_inside_nested": "first"
             },
             {
                 "some_integer_inside_nested": 2,
                 "some_string_inside_nested": "second"
-              }
+            }
         ],
         [
             {
-              "some_integer_inside_nested": 3,
-              "some_string_inside_nested": "third"
+                "some_integer_inside_nested": 3,
+                "some_string_inside_nested": "third"
             },
             {
                 "some_integer_inside_nested": 4,
                 "some_string_inside_nested": "forth"
-              }
+            }
         ],
         [
             {
-              "some_integer_inside_nested": 5,
-              "some_string_inside_nested": "fifth"
+                "some_integer_inside_nested": 5,
+                "some_string_inside_nested": "fifth"
             },
             {
                 "some_integer_inside_nested": 6,
                 "some_string_inside_nested": "sixth"
-              }
+            }
         ]
     ]
 }
diff --git a/src/server/__mocks__/mockDataFromES.js b/src/server/__mocks__/mockDataFromES.js
@@ -325,6 +325,28 @@ const mockESMapping = () => {
                   },
                 },
               },
+              treatments_array_under_subjects: {
+                type: 'nested',
+                properties: {
+                  dose_amount: { type: 'integer' },
+                  test_article_name_array_under_treatments: {
+                    type: 'keyword',
+                    fields: {
+                      analyzed: {
+                        type: 'text', analyzer: 'ngram_analyzer', search_analyzer: 'search_analyzer', term_vector: 'with_positions_offsets',
+                      },
+                    },
+                  },
+                  submitter_id: {
+                    type: 'keyword',
+                    fields: {
+                      analyzed: {
+                        type: 'text', analyzer: 'ngram_analyzer', search_analyzer: 'search_analyzer', term_vector: 'with_positions_offsets',
+                      },
+                    },
+                  },
+                },
+              },
             },
           },
           gender: {
@@ -393,7 +415,7 @@ const mockArrayConfig = () => {
           _id: 'gen3-dev-subject',
           _score: 1.0,
           _source: {
-            array: ['some_array_integer_field', 'some_array_string_field'],
+            array: ['some_array_integer_field', 'some_array_string_field', 'visits.treatments_array_under_subjects', 'visits.treatments_array_under_subjects.test_article_name_array_under_treatments'],
           },
         },
       ],

diff --git a/src/server/__tests__/schema.test.js b/src/server/__tests__/schema.test.js
@@ -65,13 +65,19 @@ describe('Schema', () => {
       _matched:[MatchedItem]
     }
     type visits {
-      days_to_visit:Int,
-      visit_label:String,
-      follow_ups:follow_ups,
+      days_to_visit: Int,
+      visit_label: String,
+      follow_ups: follow_ups,
+      treatments_array_under_subjects: [treatments_array_under_subjects],
     }
     type follow_ups {
-      days_to_follow_up:Int,
-      follow_up_label:String,
+      days_to_follow_up: Int,
+      follow_up_label: String,
+    }
+    type treatments_array_under_subjects {
+      dose_amount: Int,
+      test_article_name_array_under_treatments: [String],
+      submitter_id: String,
     }
     type File {
       gen3_resource_path: String,

diff --git a/src/server/es/index.js b/src/server/es/index.js
@@ -508,7 +508,7 @@ class ES {
         let field = f;
         if (f.endsWith(config.analyzedTextFieldSuffix)) {
           // remove ".analyzed" suffix from field name
-          field = f.substr(0, f.length - config.analyzedTextFieldSuffix.length);
+          field = f.substring(0, f.length - config.analyzedTextFieldSuffix.length);
         }
         return {
           field,

diff --git a/src/server/schema.js b/src/server/schema.js
@@ -18,12 +18,16 @@ const esgqlTypeMapping = {
 
 const histogramTypePrefix = 'RegularAccess';
 
-const getGQLType = (esInstance, esIndex, field, esFieldType) => {
+const getGQLType = (esInstance, esIndex, field, esFieldType, nestedFieldKeys = []) => {
   const gqlType = esgqlTypeMapping[esFieldType];
   if (!gqlType) {
     throw new Error(`Invalid type ${esFieldType} for field ${field} in index ${esIndex}`);
   }
-  const isArrayField = esInstance.isArrayField(esIndex, field);
+  let fieldForArrayCheck = field;
+  if (nestedFieldKeys.length) {
+    fieldForArrayCheck = `${nestedFieldKeys.join('.')}.${field}`;
+  }
+  const isArrayField = esInstance.isArrayField(esIndex, fieldForArrayCheck);
   if (isArrayField && esFieldType !== 'nested') {
     return `[${gqlType}]`;
   }
@@ -68,11 +72,11 @@ const getQuerySchemaForType = (esType) => {
     ): [${esTypeObjName}]`;
 };
 
-const getFieldGQLTypeMapForProperties = (esInstance, esIndex, properties) => {
+const getFieldGQLTypeMapForProperties = (esInstance, esIndex, properties, nestedFieldKeys = []) => {
   const result = Object.keys(properties).map((field) => {
     const esFieldType = (properties[field].esType)
       ? properties[field].esType : properties[field].type;
-    const gqlType = getGQLType(esInstance, esIndex, field, esFieldType);
+    const gqlType = getGQLType(esInstance, esIndex, field, esFieldType, nestedFieldKeys);
 
     return {
       field, type: gqlType, esType: esFieldType, properties: properties[field].properties,
@@ -109,7 +113,7 @@ const getTypeSchemaForOneIndex = (esInstance, esIndex, esType) => {
     const esFieldType = fieldESTypeMap[fieldKey].type;
     if (esFieldType === 'nested' && !existingFields.has(fieldKey)) {
       const { properties } = fieldESTypeMap[fieldKey];
-      queueTypes.push({ type: `${fieldKey}`, properties });
+      queueTypes.push({ type: `${fieldKey}`, properties, nestedFieldKeys: [fieldKey] });
       existingFields.add(fieldKey);
     }
   });
@@ -123,10 +127,10 @@ const getTypeSchemaForOneIndex = (esInstance, esIndex, esType) => {
 
   while (queueTypes.length > 0) {
     const t = queueTypes.shift();
-    const gqlTypes = getFieldGQLTypeMapForProperties(esInstance, esIndex, t.properties);
+    const gqlTypes = getFieldGQLTypeMapForProperties(esInstance, esIndex, t.properties, t.nestedFieldKeys);
     gqlTypes.forEach((entry) => {
       if (entry.esType === 'nested' && !existingFields.has(entry.field)) {
-        queueTypes.push({ type: `${entry.field}`, properties: entry.properties });
+        queueTypes.push({ type: `${entry.field}`, properties: entry.properties, nestedFieldKeys: [...t.nestedFieldKeys, entry.field] });
         existingFields.add(entry.field);
         fieldToArgs[entry.field] = getArgsByField(entry.field, entry.properties);
       }