Skip to content

Commit

Permalink
feat(comprehend): impl connector
Browse files Browse the repository at this point in the history
  • Loading branch information
DenovVasil committed Sep 19, 2024
1 parent 43ab1c1 commit 6d9558f
Show file tree
Hide file tree
Showing 15 changed files with 1,470 additions and 123 deletions.

Large diffs are not rendered by default.

Large diffs are not rendered by default.

Original file line number Diff line number Diff line change
Expand Up @@ -9,7 +9,11 @@
import io.camunda.connector.api.annotation.OutboundConnector;
import io.camunda.connector.api.outbound.OutboundConnectorContext;
import io.camunda.connector.api.outbound.OutboundConnectorFunction;
import io.camunda.connector.comprehend.caller.SyncComprehendCaller;
import io.camunda.connector.comprehend.model.ComprehendRequest;
import io.camunda.connector.comprehend.model.ComprehendRequestData;
import io.camunda.connector.comprehend.model.ComprehendSyncRequestData;
import io.camunda.connector.comprehend.supplier.ComprehendClientSupplier;
import io.camunda.connector.generator.java.annotation.ElementTemplate;

@OutboundConnector(
Expand All @@ -32,8 +36,30 @@
icon = "icon.svg")
public class ComprehendConnectorFunction implements OutboundConnectorFunction {

private ComprehendClientSupplier clientSupplier;

private SyncComprehendCaller syncComprehendCaller;

public ComprehendConnectorFunction() {
clientSupplier = new ComprehendClientSupplier();
syncComprehendCaller = new SyncComprehendCaller();
}

public ComprehendConnectorFunction(
ComprehendClientSupplier clientSupplier, SyncComprehendCaller syncComprehendCaller) {
this.clientSupplier = clientSupplier;
this.syncComprehendCaller = syncComprehendCaller;
}

@Override
public Object execute(OutboundConnectorContext context) throws Exception {
var request = context.bindVariables(ComprehendRequest.class);
ComprehendRequestData requestData = request.getInput();
if (requestData instanceof ComprehendSyncRequestData) {

return syncComprehendCaller.call(clientSupplier.getSyncClient(request),(ComprehendSyncRequestData) requestData);
}
return null;
}

}
Original file line number Diff line number Diff line change
@@ -0,0 +1,108 @@
/*
* Copyright Camunda Services GmbH and/or licensed to Camunda Services GmbH
* under one or more contributor license agreements. Licensed under a proprietary license.
* See the License.txt file for more information. You may not use this file
* except in compliance with the proprietary license.
*/
package io.camunda.connector.comprehend.caller;

import com.amazonaws.services.comprehend.AmazonComprehendClient;
import com.amazonaws.services.comprehend.model.*;
import com.amazonaws.util.CollectionUtils;
import io.camunda.connector.comprehend.model.ComprehendAsyncRequestData;
import java.util.List;
import java.util.Map;
import java.util.Objects;
import org.apache.commons.lang3.StringUtils;

public class AsyncComprehendCaller
implements ComprehendCaller<StartDocumentClassificationJobResult, ComprehendAsyncRequestData> {

@Override
public StartDocumentClassificationJobResult call(
AmazonComprehendClient client, ComprehendAsyncRequestData asyncRequest) {
var docClassificationRequest = new StartDocumentClassificationJobRequest();

if (StringUtils.isNoneBlank(asyncRequest.clientRequestToken())) {
docClassificationRequest.withClientRequestToken(asyncRequest.clientRequestToken());
}

docClassificationRequest.withDataAccessRoleArn(asyncRequest.dataAccessRoleArn());

if (StringUtils.isNoneEmpty(asyncRequest.documentClassifierArn())) {
docClassificationRequest.withDocumentClassifierArn(asyncRequest.documentClassifierArn());
}

if (StringUtils.isNoneBlank(asyncRequest.flywheelArn())) {
docClassificationRequest.withFlywheelArn(asyncRequest.flywheelArn());
}

docClassificationRequest.withInputDataConfig(prepareInputConfig(asyncRequest));

if (StringUtils.isNoneEmpty(asyncRequest.jobName())) {
docClassificationRequest.withJobName(asyncRequest.jobName());
}

docClassificationRequest.withOutputDataConfig(prepareOutputDataConf(asyncRequest));

if (asyncRequest.tags() != null) {
docClassificationRequest.withTags(prepareTags(asyncRequest));
}

if (StringUtils.isNoneEmpty(asyncRequest.volumeKmsKeyId())) {
docClassificationRequest.withVolumeKmsKeyId(asyncRequest.volumeKmsKeyId());
}

docClassificationRequest.withVpcConfig(prepareVpcConfig(asyncRequest));

return client.startDocumentClassificationJob(docClassificationRequest);
}

private InputDataConfig prepareInputConfig(ComprehendAsyncRequestData request) {
var inputConfig =
new InputDataConfig()
.withS3Uri(request.inputS3Uri())
.withDocumentReaderConfig(prepareDocumentReaderConfig(request).orElse(null));

if (StringUtils.isNoneBlank(request.comprehendInputFormat().name())) {
inputConfig.withInputFormat(request.comprehendInputFormat().name());
}

return inputConfig;
}

private OutputDataConfig prepareOutputDataConf(ComprehendAsyncRequestData request) {
var outputConf = new OutputDataConfig().withS3Uri(request.outputS3Uri());

if (StringUtils.isNoneEmpty(request.outputKmsKeyId())) {
outputConf.withKmsKeyId(request.outputKmsKeyId());
}

return outputConf;
}

private List<Tag> prepareTags(ComprehendAsyncRequestData request) {
return request.tags().entrySet().stream().filter(Objects::nonNull).map(this::creatTag).toList();
}

private Tag creatTag(Map.Entry<String, String> entry) {
return new Tag().withKey(entry.getKey()).withValue(entry.getValue());
}

private VpcConfig prepareVpcConfig(ComprehendAsyncRequestData request) {
List<String> groupIds = request.securityGroupIds();
List<String> subnets = request.subnets();

if (CollectionUtils.isNullOrEmpty(groupIds) && CollectionUtils.isNullOrEmpty(subnets)) {
return null;
}

if (!CollectionUtils.isNullOrEmpty(groupIds) && !CollectionUtils.isNullOrEmpty(subnets)) {
return new VpcConfig().withSecurityGroupIds(groupIds).withSubnets(subnets);
} else {
String msg = "Or both VpcConfig fields SecurityGroupIds and Subnets or none";
LOGGER.warn(msg);
throw new IllegalArgumentException(msg);
}
}
}
Original file line number Diff line number Diff line change
@@ -0,0 +1,72 @@
/*
* Copyright Camunda Services GmbH and/or licensed to Camunda Services GmbH
* under one or more contributor license agreements. Licensed under a proprietary license.
* See the License.txt file for more information. You may not use this file
* except in compliance with the proprietary license.
*/
package io.camunda.connector.comprehend.caller;

import static io.camunda.connector.comprehend.model.ComprehendDocumentReadAction.TEXTRACT_ANALYZE_DOCUMENT;

import com.amazonaws.AmazonWebServiceResult;
import com.amazonaws.ResponseMetadata;
import com.amazonaws.services.comprehend.AmazonComprehendClient;
import com.amazonaws.services.comprehend.model.DocumentReadFeatureTypes;
import com.amazonaws.services.comprehend.model.DocumentReaderConfig;
import io.camunda.connector.comprehend.model.ComprehendDocumentReadMode;
import io.camunda.connector.comprehend.model.ComprehendRequestData;
import java.util.ArrayList;
import java.util.List;
import java.util.Optional;
import org.apache.commons.lang3.StringUtils;
import org.slf4j.Logger;
import org.slf4j.LoggerFactory;

public interface ComprehendCaller<
T extends AmazonWebServiceResult<ResponseMetadata>, R extends ComprehendRequestData> {

Logger LOGGER = LoggerFactory.getLogger(ComprehendCaller.class);

T call(AmazonComprehendClient client, R requestData);

default Optional<DocumentReaderConfig> prepareDocumentReaderConfig(
ComprehendRequestData requestData) {
if (StringUtils.isBlank(requestData.getDocumentReadAction().name())) {
return Optional.empty();
}

var documentReaderConfig =
new DocumentReaderConfig()
.withDocumentReadAction(requestData.getDocumentReadAction().name());

if (!requestData.getDocumentReadMode().equals(ComprehendDocumentReadMode.NO_DATA)) {
documentReaderConfig.withDocumentReadMode(requestData.getDocumentReadMode().name());
}

if (requestData.getDocumentReadAction().equals(TEXTRACT_ANALYZE_DOCUMENT)) {
List<String> features = prepareFeatures(requestData);
if (features.isEmpty()) { // todo check real example
LOGGER.warn("DocumentReadAction: TEXTRACT_ANALYZE_DOCUMENT but features not selected.");
throw new IllegalArgumentException(
"If you chose TEXTRACT_ANALYZE_DOCUMENT as the read action,"
+ " you must specify one feature types");
}
documentReaderConfig.withFeatureTypes(prepareFeatures(requestData));
}

return Optional.of(documentReaderConfig);


}

default List<String> prepareFeatures(ComprehendRequestData requestData) {
List<String> features = new ArrayList<>(2);
if (requestData.getFeatureTypeForms()) {
features.add(DocumentReadFeatureTypes.FORMS.name());
}
if (requestData.getFeatureTypeTables()) {
features.add(DocumentReadFeatureTypes.TABLES.name());
}
return features;
}
}
Original file line number Diff line number Diff line change
@@ -0,0 +1,28 @@
/*
* Copyright Camunda Services GmbH and/or licensed to Camunda Services GmbH
* under one or more contributor license agreements. Licensed under a proprietary license.
* See the License.txt file for more information. You may not use this file
* except in compliance with the proprietary license.
*/
package io.camunda.connector.comprehend.caller;

import com.amazonaws.services.comprehend.AmazonComprehendClient;
import com.amazonaws.services.comprehend.model.ClassifyDocumentRequest;
import com.amazonaws.services.comprehend.model.ClassifyDocumentResult;
import io.camunda.connector.comprehend.model.ComprehendSyncRequestData;

public class SyncComprehendCaller
implements ComprehendCaller<ClassifyDocumentResult, ComprehendSyncRequestData> {

@Override
public ClassifyDocumentResult call(
AmazonComprehendClient client, ComprehendSyncRequestData requestData) {
ClassifyDocumentRequest classifyDocumentRequest =
new ClassifyDocumentRequest()
.withText(requestData.text())
.withEndpointArn(requestData.endpointArn())
.withDocumentReaderConfig(prepareDocumentReaderConfig(requestData).orElse(null));

return client.classifyDocument(classifyDocumentRequest);
}
}
Loading

0 comments on commit 6d9558f

Please sign in to comment.