marcel/familienarchiv
1
0
Fork 0
Code Issues 119 Pull Requests 2 Actions Packages Projects Releases Wiki Activity

perf(ocr): resolve person names in single batch query in getTrainingInfo

Browse Source 
Replace the per-run getById loop with a single getAllById call on distinct
person IDs, eliminating the N+1 query when training history contains multiple
sender model runs.

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
This commit is contained in:
Marcel
2026-04-17 21:21:12 +02:00
committed by marcel
parent b23118268b
commit 3ee4424556
2 changed files with 40 additions and 9 deletions
Download Patch File Download Diff File Expand all files Collapse all files

17
backend/src/main/java/org/raddatz/familienarchiv/service/OcrTrainingService.java
View File

@@ -202,16 +202,15 @@ public class OcrTrainingService {
List<OcrTrainingRun> recentRuns = trainingRunRepository.findTop20ByOrderByCreatedAtDesc();
OcrTrainingRun lastRun = recentRuns.isEmpty() ? null : recentRuns.get(0);
List<UUID> distinctPersonIds = recentRuns.stream()
.map(OcrTrainingRun::getPersonId)
.filter(Objects::nonNull)
.distinct()
.collect(java.util.stream.Collectors.toList());
Map<String, String> personNames = new HashMap<>();
for (OcrTrainingRun run : recentRuns) {
if (run.getPersonId() != null && !personNames.containsKey(run.getPersonId().toString())) {
try {
personNames.put(run.getPersonId().toString(),
personService.getById(run.getPersonId()).getDisplayName());
} catch (Exception e) {
log.debug("Could not resolve display name for person {}: {}", run.getPersonId(), e.getMessage());
}
}
if (!distinctPersonIds.isEmpty()) {
personService.getAllById(distinctPersonIds)
.forEach(p -> personNames.put(p.getId().toString(), p.getDisplayName()));
}
return new TrainingInfoResponse(

32
backend/src/test/java/org/raddatz/familienarchiv/service/OcrTrainingServiceTest.java
View File

@@ -4,6 +4,7 @@ import org.junit.jupiter.api.BeforeEach;
import org.junit.jupiter.api.Test;
import org.raddatz.familienarchiv.exception.DomainException;
import org.raddatz.familienarchiv.model.OcrTrainingRun;
import org.raddatz.familienarchiv.model.Person;
import org.raddatz.familienarchiv.model.TrainingStatus;
import org.raddatz.familienarchiv.model.TranscriptionBlock;
import org.raddatz.familienarchiv.repository.OcrTrainingRunRepository;
@@ -232,6 +233,37 @@ class OcrTrainingServiceTest {
run.getStatus() == TrainingStatus.FAILED && run.getErrorMessage() != null));
}
// ─── getTrainingInfo: batch person name resolution ────────────────────────
@Test
void getTrainingInfo_resolves_person_names_in_single_batch_call() {
UUID personA = UUID.randomUUID();
UUID personB = UUID.randomUUID();
List<OcrTrainingRun> runs = List.of(
OcrTrainingRun.builder().id(UUID.randomUUID()).status(TrainingStatus.DONE)
.personId(personA).blockCount(5).documentCount(1).modelName("sender_a").build(),
OcrTrainingRun.builder().id(UUID.randomUUID()).status(TrainingStatus.DONE)
.personId(personB).blockCount(5).documentCount(1).modelName("sender_b").build(),
OcrTrainingRun.builder().id(UUID.randomUUID()).status(TrainingStatus.DONE)
.personId(personA).blockCount(5).documentCount(1).modelName("sender_a").build()
);
when(runRepository.findTop20ByOrderByCreatedAtDesc()).thenReturn(runs);
when(exportService.queryEligibleBlocks()).thenReturn(List.of());
Person pa = Person.builder().id(personA).firstName("Anna").lastName("Müller").build();
Person pb = Person.builder().id(personB).firstName("Karl").lastName("Bauer").build();
when(personService.getAllById(argThat(ids -> ids.size() == 2 && ids.containsAll(List.of(personA, personB)))))
.thenReturn(List.of(pa, pb));
when(healthClient.isHealthy()).thenReturn(true);
OcrTrainingService.TrainingInfoResponse info = service.getTrainingInfo();
verify(personService, never()).getById(any());
verify(personService, times(1)).getAllById(any());
assertThat(info.personNames()).containsKey(personA.toString());
assertThat(info.personNames()).containsKey(personB.toString());
}
// ─── Orphan recovery ──────────────────────────────────────────────────────
@Test