Extract Data from Any Document like Resume, Biil, or any Form with AWS Textract Synchronous Snippet --- Tekraze.com

extract-data-textract.java

//Imports for packages used
import com.amazonaws.auth.AWSStaticCredentialsProvider;
import com.amazonaws.auth.BasicAWSCredentials;
import com.amazonaws.client.builder.AwsClientBuilder.EndpointConfiguration;
import com.amazonaws.services.textract.AmazonTextract;
import com.amazonaws.services.textract.AmazonTextractClientBuilder;
import com.amazonaws.services.textract.model.AnalyzeDocumentRequest;
import com.amazonaws.services.textract.model.AnalyzeDocumentResult;
import com.amazonaws.services.textract.model.Document;
import com.amazonaws.services.textract.model.S3Object;

//

/******************************** Step 1 
Set credentials and create a credential object for authorization */

String access_key_id = "YOUR_APP_ID";
String access_secret_key = "YOUR_APP_KEY";
BasicAWSCredentials awsCreds = new BasicAWSCredentials(access_key_id, access_secret_key);        
AWSStaticCredentialsProvider awsStaticCredentialsProvider = new AWSStaticCredentialsProvider(awsCreds);
        
/**************************** Step 2
Create Endpoint URL for Service you want to use
here we using textract at us-east-2
*/
String region = "us-east-2";        
String regionUrl = "https://textract.us-east-2.amazonaws.com";        
EndpointConfiguration endpoint = new EndpointConfiguration(regionUrl, region);

/*************************************************** Step 3 
Intialize The Textract Client
*/      
AmazonTextract textractClient = AmazonTextractClientBuilder.standard()
  .withCredentials(awsStaticCredentialsProvider)
  .withEndpointConfiguration(endpoint)
  .build();

/******************************************** Step 4
Create a Document from Either S3 Object or from a local file
*/

/********** Using S3 Object*************/
/*** S3 object like PDF, PNG, JPG ***/
// Initialize Document

// Intialize S3 Object
S3Object s3Object = new S3Object()
  .withBucket(bucketName) // Name of bucket where object is stored
  .withName(objectName); // Name of Object to get

// Intialize document
Document doc = new Document()
  .withS3Object(s3Object);

/************* Using Local Document converted as Image  *****************/
// Read as byte[] from the input file
byte[] fileAsBytes = file.getBytes();

// Intialize document and pass byte[] from file
Document document = new Document()        
  .withBytes(ByteBuffer.wrap(fileAsBytes));

/********************** Step 5*************
* Initialize AnalyzeDocument Request
*/        
AnalyzeDocumentRequest request = new AnalyzeDocumentRequest()
  .withFeatureTypes("TABLES","FORMS") // Features to get or anyone from table to form 
  .withDocument(document); // the document we created in step 4

/************************Step 6*********************
* Initialize AnlyzeDocument Result
*/
AnalyzeDocumentResult result = textractClient
  .analyzeDocument(request);

/**************You can now use this result ******************/

/****************** Step 7***************************
*Getting Data from Blocks
Blocks are the Nested data having data of a Page, a line, or a word and their relationships.
***/
// List of Blocks        
List<com.amazonaws.services.textract.model.Block> blocks = result.getBlocks();

// Get Specific data by iterating blocks List     
blocks.forEach(
  block -> { // Single block
    System.out.println(block.getBlockType()); // Type like LINE, WORD
    System.out.println(block.getText()); // Text from the block from file
    System.out.println(block.getConfidence()); // Confidence of data extracted is good
    System.out.println(block.getEntityTypes()); // Entity Types like Key or Value or something else
    System.out.println(block.getRelationships()); // Relationship between blocks like A key may link to a value
  }        
);