[ALS-6173] small hpds optimization

hms-dbmi · Mar 27, 2024 · 6df032b · 6df032b
1 parent 35797d0
commit 6df032b
Show file tree

Hide file tree

Showing 3 changed files with 13 additions and 13 deletions.
diff --git a/data/src/main/java/edu/harvard/hms/dbmi/avillach/hpds/data/phenotype/PhenoCube.java b/data/src/main/java/edu/harvard/hms/dbmi/avillach/hpds/data/phenotype/PhenoCube.java
@@ -217,14 +217,15 @@ public void setLoadingMap(List<KeyAndValue<V>> newMap) {
 		this.loadingMap= newMap ;
 	}
 
-	public List<KeyAndValue<V>> getValuesForKeys(Set<Integer> patientIds) {
+	public List<KeyAndValue<V>> getValuesForKeys(List<Integer> sortedPatientIds) {
 		List<KeyAndValue<V>> values = new ArrayList<>();
 		int x = 0;
-		for(Integer id : patientIds) {
+		for(Integer id : sortedPatientIds) {
 			while(x < sortedByKey.length && sortedByKey[x].key<id) {
 				x++;
 			}
-			while(x < sortedByKey.length && sortedByKey[x].key==id) {
+			if (x >= sortedByKey.length) { return values; }
+			while(sortedByKey[x].key==id) {
 				values.add(sortedByKey[x]);
 				x++;
 			}

diff --git a/...essing/src/main/java/edu/harvard/hms/dbmi/avillach/hpds/processing/TimelineProcessor.java b/...essing/src/main/java/edu/harvard/hms/dbmi/avillach/hpds/processing/TimelineProcessor.java
@@ -2,13 +2,8 @@
 
 import java.io.FileNotFoundException;
 import java.io.IOException;
-import java.util.ArrayList;
-import java.util.Collections;
-import java.util.HashMap;
-import java.util.LinkedHashMap;
-import java.util.List;
+import java.util.*;
 import java.util.Map.Entry;
-import java.util.Set;
 import java.util.stream.Collectors;
 
 import edu.harvard.hms.dbmi.avillach.hpds.data.phenotype.KeyAndValue;
@@ -46,7 +41,8 @@ public void runQuery(Query query, AsyncResult asyncResult) throws NotEnoughMemor
 		query.setRequiredFields(new ArrayList<>());
 
 		// list patients involved
-		Set<Integer> patientIds = abstractProcessor.getPatientSubsetForQuery(query);
+		List<Integer> patientIds = new ArrayList<>(abstractProcessor.getPatientSubsetForQuery(query));
+		patientIds.sort(Integer::compareTo);
 
 		// get start time for the timeline
 		long startTime = Long.MAX_VALUE;

diff --git a/...in/java/edu/harvard/hms/dbmi/avillach/hpds/processing/timeseries/TimeseriesProcessor.java b/...in/java/edu/harvard/hms/dbmi/avillach/hpds/processing/timeseries/TimeseriesProcessor.java
@@ -98,11 +98,14 @@ private void exportTimeData(Query query, AsyncResult result, TreeSet<Integer> id
 		pathList.addAll(query.getRequiredFields());
 		pathList.addAll(query.getCategoryFilters().keySet());
 		pathList.addAll(query.getNumericFilters().keySet());
-
-		addDataForConcepts(pathList, exportedConceptPaths, idList, result);
+
+		ArrayList<Integer> ids = new ArrayList<>(idList);
+		ids.sort(Integer::compareTo);
+
+		addDataForConcepts(pathList, exportedConceptPaths, ids, result);
 	}
 
-	private void addDataForConcepts(Collection<String> pathList, Set<String> exportedConceptPaths, TreeSet<Integer> idList, AsyncResult result) throws IOException {
+	private void addDataForConcepts(Collection<String> pathList, Set<String> exportedConceptPaths, List<Integer> idList, AsyncResult result) throws IOException {
 		for (String conceptPath : pathList) {
 			//skip concepts we may already have encountered
 			if(exportedConceptPaths.contains(conceptPath)) {