Querying an index

To search an index with the console

1. Sign in to the AWS Management Console and open the Amazon Kendra console at http://console.aws.amazon.com/kendra/.

2. On the navigation pane, choose Indexes.

3. Choose your index.

4. In the navigation menu, choose the option to search your index.

5. Enter a query in the text box and then press enter.

6. Amazon Kendra returns the results of the search.

To search an index with Python or Java

• The following example searches an index. Change the value of query to your search query and index_id or indexId to the index identifier of the index that you want to search.

  You can also get the query ID for the search as part of the response elements when you call the Query API.

  Python

  import boto3
  import pprint
  
  kendra = boto3.client("kendra")
  
  # Provide the index ID
  index_id = "index-id"
  # Provide the query text
  query = "query text"
  
  response = kendra.query(
          QueryText = query,
          IndexId = index_id)
  
  print("\nSearch results for query: " + query + "\n")        
  
  for query_result in response["ResultItems"]:
  
      print("-------------------")
      print("Type: " + str(query_result["Type"]))
          
      if query_result["Type"]=="ANSWER" or query_result["Type"]=="QUESTION_ANSWER":
          answer_text = query_result["DocumentExcerpt"]["Text"]
          print(answer_text)
  
      if query_result["Type"]=="DOCUMENT":
          if "DocumentTitle" in query_result:
              document_title = query_result["DocumentTitle"]["Text"]
              print("Title: " + document_title)
          document_text = query_result["DocumentExcerpt"]["Text"]
          print(document_text)
  
      print("------------------\n\n")  
      

  Java

  package com.amazonaws.kendra;
  
  import software.amazon.awssdk.services.kendra.KendraClient;
  import software.amazon.awssdk.services.kendra.model.QueryRequest;
  import software.amazon.awssdk.services.kendra.model.QueryResponse;
  import software.amazon.awssdk.services.kendra.model.QueryResultItem;
  
  public class SearchIndexExample {
      public static void main(String[] args) {
          KendraClient kendra = KendraClient.builder().build();
  
          String query = "query text";
          String indexId = "index-id";
  
          QueryRequest queryRequest = QueryRequest
              .builder()
              .queryText(query)
              .indexId(indexId)
              .build();
  
          QueryResponse queryResponse = kendra.query(queryRequest);
  
          System.out.println(String.format("\nSearch results for query: %s", query));
          for(QueryResultItem item: queryResponse.resultItems()) {
              System.out.println("----------------------");
              System.out.println(String.format("Type: %s", item.type()));
  
              switch(item.type()) {
                  case QUESTION_ANSWER:
                  case ANSWER:
                      String answerText = item.documentExcerpt().text();
                      System.out.println(answerText);
                      break;
                  case DOCUMENT:
                      String documentTitle = item.documentTitle().text();
                      System.out.println(String.format("Title: %s", documentTitle));
                      String documentExcerpt = item.documentExcerpt().text();
                      System.out.println(String.format("Excerpt: %s", documentExcerpt));
                      break;
                  default:
                      System.out.println(String.format("Unknown query result type: %s", item.type()));
  
              }
  
              System.out.println("-----------------------\n");
          }
      }
  }

To search an index using Postman

1. Create a new collection in Postman and set the request type to POST.

2. Enter the endpoint URL. For example, https://kendra.<region>.amazonaws.com.

3. Select the Authorization tab and enter the following information.

   • Type—Select AWS signature.

   • AccessKey—Enter the access key generated when you create an IAM user.

   • SecretKey—Enter the secret key generated when you create an IAM user.

   • AWS Region—Enter the region of you index. For example, us-west-2.

   • Service Name—Enter kendra. This is case sensitive, so must be lower case.

     Warning

     If you enter the incorrect service name or don't use lowercase, an error is thrown once you select Send to send the request: "Credential should be scoped to the correct service 'kendra'."

     You must also check that you entered the correct access key and secret key.

4. Select the Headers tab and enter the following key and value information.

   • Key: X-Amz-Target

     Value: com.amazonaws.kendra.AWSKendraFrontendService.Query

   • Key: Content-Encoding

     Value: amz-1.0

5. Select the Body tab and do the following.

   • Choose the raw JSON type for the body of the request.

   • Enter a JSON that includes your index ID and query text.

     {
         "IndexId": "index-id",
         "QueryText": "enter a query here"
     }

     Warning

     If your JSON doesn't use the correct indendation, an error is thrown: "SerializationException". Check the indendation in your JSON.

6. Select Send (near the top right).

To search an index in Spanish in the console

1. Sign in to the AWS Management Console and open the Amazon Kendra console at http://console.aws.amazon.com/kendra/.

2. In the navigation menu, choose Indexes and choose your index.

3. In the navigation menu, choose the option to search your index.

4. In the search settings, select the Languages dropdown and choose Spanish.

5. Enter a query into the text box and then press enter.

6. Amazon Kendra returns the results of the search in Spanish.

To search an index in Spanish using the CLI, Python or Java

• The following example searches an index in Spanish. Change the value searchString to your search query and the value indexID to the identifier of the index that you want to search. The language code for Spanish is es. You can replace this with your own language code.

  CLI

  {
    "EqualsTo":{      
      "Key": "_language_code",
      "Value": {
      "StringValue": "es"
      }
    }
  }

  Python

  import boto3
  import pprint
  
  kendra = boto3.client("kendra")
  
  # Provide the index ID
  index_id = "index-id"
  # Provide the query text
  query = "search-string"
  
  # Includes the index ID, query text, and language attribute filter
  response = kendra.query(
          QueryText = query,
          IndexId = index_id,
          AttributeFilter = {
              "EqualsTo": {      
                  "Key": "_language_code",
                  "Value": {
                      "StringValue": "es"
                      }
                  }
              })
  
  print ("\nSearch results|Resultados de la búsqueda: " + query + "\n")        
  
  for query_result in response["ResultItems"]:
  
      print("-------------------")
      print("Type: " + str(query_result["Type"]))
          
      if query_result["Type"]=="ANSWER" or query_result["Type"]=="QUESTION_ANSWER":
          answer_text = query_result["DocumentExcerpt"]["Text"]
          print(answer_text)
  
      if query_result["Type"]=="DOCUMENT":
          if "DocumentTitle" in query_result:
              document_title = query_result["DocumentTitle"]["Text"]
              print("Title: " + document_title)
          document_text = query_result["DocumentExcerpt"]["Text"]
          print(document_text)
  
      print("------------------\n\n")
      
                                  

  Java

  package com.amazonaws.kendra;
  
  import software.amazon.awssdk.services.kendra.KendraClient;
  import software.amazon.awssdk.services.kendra.model.QueryRequest;
  import software.amazon.awssdk.services.kendra.model.QueryResponse;
  import software.amazon.awssdk.services.kendra.model.QueryResultItem;
  
  public class SearchIndexExample {
      public static void main(String[] args) {
          KendraClient kendra = KendraClient.builder().build();
  
          String query = "searchString";
          String indexId = "indexID";
  
          QueryRequest queryRequest = QueryRequest.builder()
              .queryText(query)
              .indexId(indexId)
              .attributeFilter(
                   AttributeFilter.builder()
                       .withEqualsTo(
                           DocumentAttribute.builder()
                               .withKey("_language_code")
                               .withValue("es")
                               .build())
                       .build())
              .build();
  
          QueryResponse queryResponse = kendra.query(queryRequest);
  
          System.out.println(String.format("\nSearch results|
                                            Resultados de la búsqueda: %s", query));
          for(QueryResultItem item: queryResponse.resultItems()) {
              System.out.println("----------------------");
              System.out.println(String.format("Type: %s", item.type()));
  
              switch(item.type()) {
                  case QUESTION_ANSWER:
                  case ANSWER:
                      String answerText = item.documentExcerpt().text();
                      System.out.println(answerText);
                      break;
                  case DOCUMENT:
                      String documentTitle = item.documentTitle().text();
                      System.out.println(String.format("Title: %s", documentTitle));
                      String documentExcerpt = item.documentExcerpt().text();
                      System.out.println(String.format("Excerpt: %s", documentExcerpt));
                      break;
                  default:
                      System.out.println(String.format("Unknown query result type: %s", item.type()));
  
              }
  
              System.out.println("-----------------------\n");
          }
      }
  }