Merge branch 'main' into hotfix/null-point

# Conflicts: # search-commons/src/main/java/no/unit/nva/search/resource/SimplifiedMutator.java
BIBSYSDEV · Jan 9, 2025 · d740a0c · d740a0c
2 parents ebe785b + ff7c7c5
commit d740a0c
Show file tree

Hide file tree

Showing 150 changed files with 7,538 additions and 7,077 deletions.
diff --git a/.git-blame-ignore-revs b/.git-blame-ignore-revs
@@ -1,2 +1,5 @@
 # style: Reformat project with Spotless
 2dfd69cbc977fa45922485f1d252d81fc0da7447
+
+# style: Reformat project
+0825cca42191e030ef7c46cc9d86c8d004370d85
diff --git a/batch-index/src/main/java/no/unit/nva/indexingclient/AggregationsValidator.java b/batch-index/src/main/java/no/unit/nva/indexingclient/AggregationsValidator.java
@@ -1,7 +1,6 @@
 package no.unit.nva.indexingclient;
 
 import com.fasterxml.jackson.databind.JsonNode;
-
 import java.util.ArrayList;
 import java.util.List;
 

diff --git a/batch-index/src/main/java/no/unit/nva/indexingclient/BatchIndexer.java b/batch-index/src/main/java/no/unit/nva/indexingclient/BatchIndexer.java
@@ -1,106 +1,102 @@
 package no.unit.nva.indexingclient;
 
+import java.util.Arrays;
+import java.util.List;
+import java.util.stream.Collectors;
+import java.util.stream.Stream;
 import no.unit.nva.identifiers.SortableIdentifier;
 import no.unit.nva.indexingclient.models.IndexDocument;
 import no.unit.nva.s3.ListingResult;
 import no.unit.nva.s3.S3Driver;
-
 import nva.commons.core.paths.UnixPath;
-
 import org.opensearch.action.bulk.BulkItemResponse;
 import org.opensearch.action.bulk.BulkItemResponse.Failure;
 import org.opensearch.action.bulk.BulkResponse;
 import org.slf4j.Logger;
 import org.slf4j.LoggerFactory;
-
 import software.amazon.awssdk.services.s3.S3Client;
 
-import java.util.Arrays;
-import java.util.List;
-import java.util.stream.Collectors;
-import java.util.stream.Stream;
-
 public class BatchIndexer implements IndexingResult<SortableIdentifier> {
 
-    private static final Logger logger = LoggerFactory.getLogger(BatchIndexer.class);
-    private final ImportDataRequestEvent importDataRequest;
-    private final S3Driver s3Driver;
-    private final IndexingClient openSearchRestClient;
-    private final int numberOfFilesPerEvent;
-    private IndexingResultRecord<SortableIdentifier> processingResult;
-
-    public BatchIndexer(
-            ImportDataRequestEvent importDataRequestEvent,
-            S3Client s3Client,
-            IndexingClient openSearchRestClient,
-            int numberOfFilesPerEvent) {
-        this.importDataRequest = importDataRequestEvent;
-        this.openSearchRestClient = openSearchRestClient;
-        this.s3Driver = new S3Driver(s3Client, importDataRequestEvent.getBucket());
-        this.numberOfFilesPerEvent = numberOfFilesPerEvent;
-    }
-
-    public IndexingResult<SortableIdentifier> processRequest() {
-
-        ListingResult listFilesResult = fetchNextPageOfFilenames();
-        List<IndexDocument> contents =
-                fileContents(listFilesResult.getFiles()).collect(Collectors.toList());
-        List<SortableIdentifier> failedResults = indexFileContents(contents);
-        this.processingResult =
-                new IndexingResultRecord<>(
-                        failedResults,
-                        listFilesResult.getListingStartingPoint(),
-                        listFilesResult.isTruncated());
-
-        return this;
-    }
-
-    private Stream<IndexDocument> fileContents(List<UnixPath> files) {
-        return files.stream().map(s3Driver::getFile).map(IndexDocument::fromJsonString);
-    }
-
-    @Override
-    public List<SortableIdentifier> failedResults() {
-        return this.processingResult.failedResults();
-    }
-
-    @Override
-    public String nextStartMarker() {
-        return processingResult.nextStartMarker();
-    }
-
-    @Override
-    public boolean truncated() {
-        return this.processingResult.truncated();
-    }
-
-    private ListingResult fetchNextPageOfFilenames() {
-        return s3Driver.listFiles(
-                UnixPath.of(importDataRequest.getS3Path()),
-                importDataRequest.getStartMarker(),
-                numberOfFilesPerEvent);
-    }
-
-    private List<SortableIdentifier> indexFileContents(List<IndexDocument> contents) {
-
-        Stream<BulkResponse> result = openSearchRestClient.batchInsert(contents.stream());
-        List<SortableIdentifier> failures = collectFailures(result).collect(Collectors.toList());
-        failures.forEach(this::logFailure);
-        return failures;
-    }
-
-    private <T> void logFailure(T failureMessage) {
-        logger.warn("Failed to index resource:{}", failureMessage);
-    }
-
-    private Stream<SortableIdentifier> collectFailures(Stream<BulkResponse> indexActions) {
-        return indexActions
-                .filter(BulkResponse::hasFailures)
-                .map(BulkResponse::getItems)
-                .flatMap(Arrays::stream)
-                .filter(BulkItemResponse::isFailed)
-                .map(BulkItemResponse::getFailure)
-                .map(Failure::getId)
-                .map(SortableIdentifier::new);
-    }
+  private static final Logger logger = LoggerFactory.getLogger(BatchIndexer.class);
+  private final ImportDataRequestEvent importDataRequest;
+  private final S3Driver s3Driver;
+  private final IndexingClient openSearchRestClient;
+  private final int numberOfFilesPerEvent;
+  private IndexingResultRecord<SortableIdentifier> processingResult;
+
+  public BatchIndexer(
+      ImportDataRequestEvent importDataRequestEvent,
+      S3Client s3Client,
+      IndexingClient openSearchRestClient,
+      int numberOfFilesPerEvent) {
+    this.importDataRequest = importDataRequestEvent;
+    this.openSearchRestClient = openSearchRestClient;
+    this.s3Driver = new S3Driver(s3Client, importDataRequestEvent.getBucket());
+    this.numberOfFilesPerEvent = numberOfFilesPerEvent;
+  }
+
+  public IndexingResult<SortableIdentifier> processRequest() {
+
+    ListingResult listFilesResult = fetchNextPageOfFilenames();
+    List<IndexDocument> contents =
+        fileContents(listFilesResult.getFiles()).collect(Collectors.toList());
+    List<SortableIdentifier> failedResults = indexFileContents(contents);
+    this.processingResult =
+        new IndexingResultRecord<>(
+            failedResults,
+            listFilesResult.getListingStartingPoint(),
+            listFilesResult.isTruncated());
+
+    return this;
+  }
+
+  private Stream<IndexDocument> fileContents(List<UnixPath> files) {
+    return files.stream().map(s3Driver::getFile).map(IndexDocument::fromJsonString);
+  }
+
+  @Override
+  public List<SortableIdentifier> failedResults() {
+    return this.processingResult.failedResults();
+  }
+
+  @Override
+  public String nextStartMarker() {
+    return processingResult.nextStartMarker();
+  }
+
+  @Override
+  public boolean truncated() {
+    return this.processingResult.truncated();
+  }
+
+  private ListingResult fetchNextPageOfFilenames() {
+    return s3Driver.listFiles(
+        UnixPath.of(importDataRequest.getS3Path()),
+        importDataRequest.getStartMarker(),
+        numberOfFilesPerEvent);
+  }
+
+  private List<SortableIdentifier> indexFileContents(List<IndexDocument> contents) {
+
+    Stream<BulkResponse> result = openSearchRestClient.batchInsert(contents.stream());
+    List<SortableIdentifier> failures = collectFailures(result).collect(Collectors.toList());
+    failures.forEach(this::logFailure);
+    return failures;
+  }
+
+  private <T> void logFailure(T failureMessage) {
+    logger.warn("Failed to index resource:{}", failureMessage);
+  }
+
+  private Stream<SortableIdentifier> collectFailures(Stream<BulkResponse> indexActions) {
+    return indexActions
+        .filter(BulkResponse::hasFailures)
+        .map(BulkResponse::getItems)
+        .flatMap(Arrays::stream)
+        .filter(BulkItemResponse::isFailed)
+        .map(BulkItemResponse::getFailure)
+        .map(Failure::getId)
+        .map(SortableIdentifier::new);
+  }
 }
diff --git a/batch-index/src/main/java/no/unit/nva/indexingclient/Constants.java b/batch-index/src/main/java/no/unit/nva/indexingclient/Constants.java
@@ -4,9 +4,7 @@
 
 import com.typesafe.config.Config;
 import com.typesafe.config.ConfigFactory;
-
 import nva.commons.core.JacocoGenerated;
-
 import software.amazon.awssdk.http.urlconnection.UrlConnectionHttpClient;
 import software.amazon.awssdk.regions.Region;
 import software.amazon.awssdk.services.eventbridge.EventBridgeClient;

diff --git a/batch-index/src/main/java/no/unit/nva/indexingclient/EmitEventUtils.java b/batch-index/src/main/java/no/unit/nva/indexingclient/EmitEventUtils.java
@@ -4,43 +4,40 @@
 import static no.unit.nva.indexingclient.Constants.MANDATORY_UNUSED_SUBTOPIC;
 
 import com.amazonaws.services.lambda.runtime.Context;
-
+import java.time.Instant;
 import org.slf4j.Logger;
 import org.slf4j.LoggerFactory;
-
 import software.amazon.awssdk.services.eventbridge.EventBridgeClient;
 import software.amazon.awssdk.services.eventbridge.model.PutEventsRequest;
 import software.amazon.awssdk.services.eventbridge.model.PutEventsRequestEntry;
 
-import java.time.Instant;
-
 public final class EmitEventUtils {
 
-    private static final Logger logger = LoggerFactory.getLogger(EmitEventUtils.class);
-
-    private EmitEventUtils() {}
-
-    public static void emitEvent(
-            EventBridgeClient eventBridgeClient,
-            ImportDataRequestEvent importDataRequest,
-            Context context) {
-        var putEventRequestEntry = eventEntry(importDataRequest, context);
-        logger.debug("BusName:" + BATCH_INDEX_EVENT_BUS_NAME);
-        logger.debug("Event:" + putEventRequestEntry.toString());
-        var putEventRequest = PutEventsRequest.builder().entries(putEventRequestEntry).build();
-        var response = eventBridgeClient.putEvents(putEventRequest);
-        logger.debug(response.toString());
-    }
-
-    private static PutEventsRequestEntry eventEntry(
-            ImportDataRequestEvent importDataRequest, Context context) {
-        return PutEventsRequestEntry.builder()
-                .eventBusName(BATCH_INDEX_EVENT_BUS_NAME)
-                .detailType(MANDATORY_UNUSED_SUBTOPIC)
-                .source(EventBasedBatchIndexer.class.getName())
-                .time(Instant.now())
-                .detail(importDataRequest.toJsonString())
-                .resources(context.getInvokedFunctionArn())
-                .build();
-    }
+  private static final Logger logger = LoggerFactory.getLogger(EmitEventUtils.class);
+
+  private EmitEventUtils() {}
+
+  public static void emitEvent(
+      EventBridgeClient eventBridgeClient,
+      ImportDataRequestEvent importDataRequest,
+      Context context) {
+    var putEventRequestEntry = eventEntry(importDataRequest, context);
+    logger.debug("BusName:" + BATCH_INDEX_EVENT_BUS_NAME);
+    logger.debug("Event:" + putEventRequestEntry.toString());
+    var putEventRequest = PutEventsRequest.builder().entries(putEventRequestEntry).build();
+    var response = eventBridgeClient.putEvents(putEventRequest);
+    logger.debug(response.toString());
+  }
+
+  private static PutEventsRequestEntry eventEntry(
+      ImportDataRequestEvent importDataRequest, Context context) {
+    return PutEventsRequestEntry.builder()
+        .eventBusName(BATCH_INDEX_EVENT_BUS_NAME)
+        .detailType(MANDATORY_UNUSED_SUBTOPIC)
+        .source(EventBasedBatchIndexer.class.getName())
+        .time(Instant.now())
+        .detail(importDataRequest.toJsonString())
+        .resources(context.getInvokedFunctionArn())
+        .build();
+  }
 }