AWSTemplateFormatVersion: 2010-09-09
Parameters:
  TargetLanguageCode:
    Type: String
    Default: de
  InputBucketName:
    Type: String
  OutputBucketName:
    Type: String
  IAMRoleName:
    Type: String
    Default: TranslationLambdaExecRole
  LambdaFunctionName:
    Type: String
Resources:
  translateLambdaFunctionPolicy:
    Type: AWS::IAM::ManagedPolicy
    Properties:
      PolicyDocument:
        Version: '2012-10-17'
        Statement:
              - Effect: Allow
                Action:
                  - 'logs:CreateLogGroup'
                  - 'logs:CreateLogStream'
                  - 'logs:PutLogEvents'
                Resource:
                  - !Sub
                    - 'arn:aws:logs:${Region}:${AccountId}:log-group:/aws/lambda/${LogGroupName}:*'
                    - Region: !Ref "AWS::Region"
                      AccountId: !Ref "AWS::AccountId"
                      LogGroupName: !Ref LambdaFunctionName
              - Effect: Allow
                Action: 
                  - 's3:GetObject'
                  - 's3:PutObject'
                Resource:
                  - !Sub 
                    - 'arn:aws:s3:::${InputBucket}/*'
                    - InputBucket: !Ref InputBucketName
                  - !Sub 
                    - 'arn:aws:s3:::${OutputBucket}/*'
                    - OutputBucket: !Ref OutputBucketName
  IAMR1LC1M:
    Type: 'AWS::IAM::Role'
    Properties:
      RoleName: !Ref IAMRoleName
      AssumeRolePolicyDocument:
        Version: 2012-10-17
        Statement:
          - Effect: Allow
            Principal:
              Service:
                - lambda.amazonaws.com
            Action:
              - 'sts:AssumeRole'
      Path: /
      ManagedPolicyArns:
        - !Ref translateLambdaFunctionPolicy
        - arn:aws:iam::aws:policy/TranslateReadOnly
  LF4L95L:
    Type: 'AWS::Lambda::Function'
    Properties:
      FunctionName: !Ref LambdaFunctionName
      Handler: index.lambda_handler
      Runtime: python3.7
      Timeout: 900
      Environment:
        Variables:
          SourceLanguageCode: auto
          TargetBucketName: !Ref OutputBucketName
          TargetLanguageCode: !Ref TargetLanguageCode
      Role: !GetAtt 
        - IAMR1LC1M
        - Arn
      Code:
        ZipFile: |
            import boto3
            from botocore.exceptions import ClientError
            import os
            import sys
            
            def lambda_handler(event, context):
                
                #boto3 connections
                s3 = boto3.resource('s3')
                translate = boto3.client('translate')
            
                for record in event['Records']:
                    
                    #defining variables
                    SOURCE_BUCKET_NAME = record['s3']['bucket']['name']
                    TARGET_BUCKET_NAME = os.environ['TargetBucketName']
                    download_key = record['s3']['object']['key']
                    object_name = download_key.split("/")[-1]
                    object_input_path = download_key.split("/")[0]
                    upload_key = os.environ['TargetLanguageCode'] +"/" + object_input_path + "/" + object_name
                    read_path = '/tmp/' + object_name
                    write_path = '/tmp/translated-' + object_name
            
                    #creating and opening a new txt file which is where the translated text will be put
                    new_file = open(write_path, "w+")
                    
                    #downloading the file to be translated
                    s3.Bucket(SOURCE_BUCKET_NAME).download_file(download_key, read_path)
                    
                    with open(read_path, "r") as text_file:
                        #reading all the content in one step
                        all_content = text_file.read() 
                        #splitting the content in paragraphs and storing it in an array
                        content_array = all_content.split("\n")
            
                    #translating the text using Amazon Translate
                    for i in range(len(content_array)):
                        if content_array[i] != "":
                            if sys.getsizeof(content_array[i]) > 5000:
                                a = content_array[i].split()[:5]
                                print ("Error occurred - please break down the following paragraph into smaller chunks in order to avoid hitting Amazon Translate API limits.")
                                print ("Paragraph starting with\""+' '.join(a)+"...\"")
                            else:
            
                                try:
                                    result = translate.translate_text(Text=content_array[i],
                                                      SourceLanguageCode = os.environ['SourceLanguageCode'],
                                                      TargetLanguageCode = os.environ['TargetLanguageCode'])
                                except ClientError as e:
                                    print ("Unexpected error: %s" % e)
            
                            #writing the translated text to local document
                            new_file.write(result["TranslatedText"])
            
                            #===if you want to perform the translation operation by each sentence, uncomment this segment of the code===
                            #splitting the content by sentences (end of a sentence identified by full stop ".") and storing it in an array
                            #    sentence_array = content_array[i].split(".")
                            #    for j in range(len(sentence_array)):
                            #        if sentence_array[j] != "":
                            #            try:
                            #                result = translate.translate_text(Text=sentence_array[j],
                            #                                  SourceLanguageCode = os.environ['SourceLanguageCode'],
                            #                                  TargetLanguageCode = os.environ['TargetLanguageCode'])
                            #            except ClientError as e:
                            #                print ("Unexpected error: %s" % e)
                            #        new_file.write(result["TranslatedText"])
                            #        new_file.write(".") 
                            new_file.write("\n")
                        else:
                            new_file.write("\n")
            
                    new_file.close()
            
                    #writing the translated document to S3
                    s3.Bucket(TARGET_BUCKET_NAME).upload_file(write_path, upload_key)
                    
                    #removing the local files
                    os.remove(read_path)
                    os.remove(write_path)
  S3B3UNSQ:
    Type: 'AWS::S3::Bucket'
    Properties:
      BucketName: !Ref InputBucketName
      VersioningConfiguration:
        Status: Enabled
      NotificationConfiguration:
        LambdaConfigurations:
          - Event: 's3:ObjectCreated:Put'
            Function: !GetAtt 
              - LF4L95L
              - Arn
  BucketPermission:
    Type: 'AWS::Lambda::Permission'
    Properties:
      Action: 'lambda:InvokeFunction'
      FunctionName: !Ref LF4L95L
      Principal: s3.amazonaws.com
      SourceAccount: !Ref 'AWS::AccountId'
      SourceArn: !Sub 'arn:aws:s3:::${InputBucketName}'
  S3B5BC28:
    Type: 'AWS::S3::Bucket'
    Properties:
      BucketName: !Ref OutputBucketName
      VersioningConfiguration:
        Status: Enabled
Outputs:
  IAMRoleExecutionForLambdaARN:
    Description: The ARN of the IAM role that was created for the Lambda to assume
    Value: !GetAtt 
      - IAMR1LC1M
      - Arn
  LambdaFunctionArn:
    Description: The ARN of the Lambda function
    Value: !GetAtt 
      - LF4L95L
      - Arn