Utilizzare CreateJob con un SDK AWS o una CLI

    /// <summary>
    /// Create an AWS Glue job.
    /// </summary>
    /// <param name="jobName">The name of the job.</param>
    /// <param name="roleName">The name of the IAM role to be assumed by
    /// the job.</param>
    /// <param name="description">A description of the job.</param>
    /// <param name="scriptUrl">The URL to the script.</param>
    /// <returns>A Boolean value indicating the success of the action.</returns>
    public async Task<bool> CreateJobAsync(string dbName, string tableName, string bucketUrl, string jobName, string roleName, string description, string scriptUrl)
    {
        var command = new JobCommand
        {
            PythonVersion = "3",
            Name = "glueetl",
            ScriptLocation = scriptUrl,
        };

        var arguments = new Dictionary<string, string>
        {
            { "--input_database", dbName },
            { "--input_table", tableName },
            { "--output_bucket_url", bucketUrl }
        };

        var request = new CreateJobRequest
        {
            Command = command,
            DefaultArguments = arguments,
            Description = description,
            GlueVersion = "3.0",
            Name = jobName,
            NumberOfWorkers = 10,
            Role = roleName,
            WorkerType = "G.1X"
        };

        var response = await _amazonGlue.CreateJobAsync(request);
        return response.HttpStatusCode == HttpStatusCode.OK;
    }

        Aws::Client::ClientConfiguration clientConfig;
        // Optional: Set to the AWS Region in which the bucket was created (overrides config file).
        // clientConfig.region = "us-east-1";

    Aws::Glue::GlueClient client(clientConfig);

        Aws::Glue::Model::CreateJobRequest request;
        request.SetName(JOB_NAME);
        request.SetRole(roleArn);
        request.SetGlueVersion(GLUE_VERSION);

        Aws::Glue::Model::JobCommand command;
        command.SetName(JOB_COMMAND_NAME);
        command.SetPythonVersion(JOB_PYTHON_VERSION);
        command.SetScriptLocation(
                Aws::String("s3://") + bucketName + "/" + PYTHON_SCRIPT);
        request.SetCommand(command);

        Aws::Glue::Model::CreateJobOutcome outcome = client.CreateJob(request);

        if (outcome.IsSuccess()) {
            std::cout << "Successfully created the job." << std::endl;
        }
        else {
            std::cerr << "Error creating the job. " << outcome.GetError().GetMessage()
                      << std::endl;
            deleteAssets(CRAWLER_NAME, CRAWLER_DATABASE_NAME, "", bucketName,
                         clientConfig);
            return false;
        }

Come creare un processo di trasformazione dei dati

L’esempio create-job seguente crea un processo di streaming che esegue uno script archiviato in S3.

aws glue create-job \
    --name my-testing-job \
    --role AWSGlueServiceRoleDefault \
    --command '{ \
        "Name": "gluestreaming", \
        "ScriptLocation": "s3://amzn-s3-demo-bucket/folder/" \
    }' \
    --region us-east-1 \
    --output json \
    --default-arguments '{ \
        "--job-language":"scala", \
        "--class":"GlueApp" \
    }' \
    --profile my-profile \
    --endpoint https://glue.us-east-1.amazonaws.com

Contenuto di test_script.scala.

import com.amazonaws.services.glue.ChoiceOption
import com.amazonaws.services.glue.GlueContext
import com.amazonaws.services.glue.MappingSpec
import com.amazonaws.services.glue.ResolveSpec
import com.amazonaws.services.glue.errors.CallSite
import com.amazonaws.services.glue.util.GlueArgParser
import com.amazonaws.services.glue.util.Job
import com.amazonaws.services.glue.util.JsonOptions
import org.apache.spark.SparkContext
import scala.collection.JavaConverters._

object GlueApp {
    def main(sysArgs: Array[String]) {
        val spark: SparkContext = new SparkContext()
        val glueContext: GlueContext = new GlueContext(spark)
        // @params: [JOB_NAME]
        val args = GlueArgParser.getResolvedOptions(sysArgs, Seq("JOB_NAME").toArray)
        Job.init(args("JOB_NAME"), glueContext, args.asJava)
        // @type: DataSource
        // @args: [database = "tempdb", table_name = "s3-source", transformation_ctx = "datasource0"]
        // @return: datasource0
        // @inputs: []
        val datasource0 = glueContext.getCatalogSource(database = "tempdb", tableName = "s3-source", redshiftTmpDir = "", transformationContext = "datasource0").getDynamicFrame()
        // @type: ApplyMapping
        // @args: [mapping = [("sensorid", "int", "sensorid", "int"), ("currenttemperature", "int", "currenttemperature", "int"), ("status", "string", "status", "string")], transformation_ctx = "applymapping1"]
        // @return: applymapping1
        // @inputs: [frame = datasource0]
        val applymapping1 = datasource0.applyMapping(mappings = Seq(("sensorid", "int", "sensorid", "int"), ("currenttemperature", "int", "currenttemperature", "int"), ("status", "string", "status", "string")), caseSensitive = false, transformationContext = "applymapping1")
        // @type: SelectFields
        // @args: [paths = ["sensorid", "currenttemperature", "status"], transformation_ctx = "selectfields2"]
        // @return: selectfields2
        // @inputs: [frame = applymapping1]
        val selectfields2 = applymapping1.selectFields(paths = Seq("sensorid", "currenttemperature", "status"), transformationContext = "selectfields2")
        // @type: ResolveChoice
        // @args: [choice = "MATCH_CATALOG", database = "tempdb", table_name = "my-s3-sink", transformation_ctx = "resolvechoice3"]
        // @return: resolvechoice3
        // @inputs: [frame = selectfields2]
        val resolvechoice3 = selectfields2.resolveChoice(choiceOption = Some(ChoiceOption("MATCH_CATALOG")), database = Some("tempdb"), tableName = Some("my-s3-sink"), transformationContext = "resolvechoice3")
        // @type: DataSink
        // @args: [database = "tempdb", table_name = "my-s3-sink", transformation_ctx = "datasink4"]
        // @return: datasink4
        // @inputs: [frame = resolvechoice3]
        val datasink4 = glueContext.getCatalogSink(database = "tempdb", tableName = "my-s3-sink", redshiftTmpDir = "", transformationContext = "datasink4").writeDynamicFrame(resolvechoice3)
        Job.commit()
    }
}

Output:

{
    "Name": "my-testing-job"
}

Per ulteriori informazioni, consulta Creazione di processi in AWS Glue nella Guida per gli sviluppatori di AWS Glue.

    /**
     * Creates a new AWS Glue job.
     *
     * @param glueClient     the AWS Glue client to use for the operation
     * @param jobName        the name of the job to create
     * @param iam            the IAM role to associate with the job
     * @param scriptLocation the location of the script to be used by the job
     * @throws GlueException if there is an error creating the job
     */
    public static void createJob(GlueClient glueClient, String jobName, String iam, String scriptLocation) {
        try {
            JobCommand command = JobCommand.builder()
                .pythonVersion("3")
                .name("glueetl")
                .scriptLocation(scriptLocation)
                .build();

            CreateJobRequest jobRequest = CreateJobRequest.builder()
                .description("A Job created by using the AWS SDK for Java V2")
                .glueVersion("2.0")
                .workerType(WorkerType.G_1_X)
                .numberOfWorkers(10)
                .name(jobName)
                .role(iam)
                .command(command)
                .build();

            glueClient.createJob(jobRequest);
            System.out.println(jobName + " was successfully created.");

        } catch (GlueException e) {
            throw e;
        }
    }

const createJob = (name, role, scriptBucketName, scriptKey) => {
  const client = new GlueClient({});

  const command = new CreateJobCommand({
    Name: name,
    Role: role,
    Command: {
      Name: "glueetl",
      PythonVersion: "3",
      ScriptLocation: `s3://${scriptBucketName}/${scriptKey}`,
    },
    GlueVersion: "3.0",
  });

  return client.send(command);
};

        $role = $iamService->getRole("AWSGlueServiceRole-DocExample");

        $jobName = 'test-job-' . $uniqid;

        $scriptLocation = "s3://$bucketName/run_job.py";
        $job = $glueService->createJob($jobName, $role['Role']['Arn'], $scriptLocation);

    public function createJob($jobName, $role, $scriptLocation, $pythonVersion = '3', $glueVersion = '3.0'): Result
    {
        return $this->glueClient->createJob([
            'Name' => $jobName,
            'Role' => $role,
            'Command' => [
                'Name' => 'glueetl',
                'ScriptLocation' => $scriptLocation,
                'PythonVersion' => $pythonVersion,
            ],
            'GlueVersion' => $glueVersion,
        ]);
    }

Esempio 1: questo esempio crea un nuovo processo in AWS Glue. Il valore del nome del comando è sempre glueetl. AWS Glue supporta l’esecuzione di script di processo scritti in Python o Scala. In questo esempio, lo script di processo (MyTestGlueJob.py) è scritto in Python. I parametri Python vengono specificati nella variabile $DefArgs e quindi passati al comando PowerShell nel parametro DefaultArguments, che accetta una tabella hash. I parametri della variabile $JobParams provengono dall’API CreateJob, documentata nell’argomento relativo all’API Jobs (https://docs.aws.amazon.com/glue/latest/dg/aws-glue-api-jobs-job.html) nella documentazione di riferimento dell’API AWS Glue.

$Command = New-Object Amazon.Glue.Model.JobCommand
$Command.Name = 'glueetl'
$Command.ScriptLocation = 's3://amzn-s3-demo-source-bucket/admin/MyTestGlueJob.py'
$Command

$Source = "source_test_table"
$Target = "target_test_table"
$Connections = $Source, $Target

$DefArgs = @{
     '--TempDir' = 's3://amzn-s3-demo-bucket/admin'
     '--job-bookmark-option' = 'job-bookmark-disable'
     '--job-language' = 'python'
     }
$DefArgs

$ExecutionProp = New-Object Amazon.Glue.Model.ExecutionProperty
$ExecutionProp.MaxConcurrentRuns = 1
$ExecutionProp

$JobParams = @{
    "AllocatedCapacity"    = "5"
    "Command"              = $Command
    "Connections_Connection" = $Connections
    "DefaultArguments"  = $DefArgs
    "Description"       = "This is a test"
    "ExecutionProperty" = $ExecutionProp
    "MaxRetries"        = "1"
    "Name"              = "MyOregonTestGlueJob"
    "Role"              = "Amazon-GlueServiceRoleForSSM"
    "Timeout"           = "20"
     }

New-GlueJob @JobParams

Esempio 1: questo esempio crea un nuovo processo in AWS Glue. Il valore del nome del comando è sempre glueetl. AWS Glue supporta l’esecuzione di script di processo scritti in Python o Scala. In questo esempio, lo script di processo (MyTestGlueJob.py) è scritto in Python. I parametri Python vengono specificati nella variabile $DefArgs e quindi passati al comando PowerShell nel parametro DefaultArguments, che accetta una tabella hash. I parametri della variabile $JobParams provengono dall’API CreateJob, documentata nell’argomento relativo all’API Jobs (https://docs.aws.amazon.com/glue/latest/dg/aws-glue-api-jobs-job.html) nella documentazione di riferimento dell’API AWS Glue.

$Command = New-Object Amazon.Glue.Model.JobCommand
$Command.Name = 'glueetl'
$Command.ScriptLocation = 's3://amzn-s3-demo-source-bucket/admin/MyTestGlueJob.py'
$Command

$Source = "source_test_table"
$Target = "target_test_table"
$Connections = $Source, $Target

$DefArgs = @{
     '--TempDir' = 's3://amzn-s3-demo-bucket/admin'
     '--job-bookmark-option' = 'job-bookmark-disable'
     '--job-language' = 'python'
     }
$DefArgs

$ExecutionProp = New-Object Amazon.Glue.Model.ExecutionProperty
$ExecutionProp.MaxConcurrentRuns = 1
$ExecutionProp

$JobParams = @{
    "AllocatedCapacity"    = "5"
    "Command"              = $Command
    "Connections_Connection" = $Connections
    "DefaultArguments"  = $DefArgs
    "Description"       = "This is a test"
    "ExecutionProperty" = $ExecutionProp
    "MaxRetries"        = "1"
    "Name"              = "MyOregonTestGlueJob"
    "Role"              = "Amazon-GlueServiceRoleForSSM"
    "Timeout"           = "20"
     }

New-GlueJob @JobParams

class GlueWrapper:
    """Encapsulates AWS Glue actions."""

    def __init__(self, glue_client):
        """
        :param glue_client: A Boto3 Glue client.
        """
        self.glue_client = glue_client


    def create_job(self, name, description, role_arn, script_location):
        """
        Creates a job definition for an extract, transform, and load (ETL) job that can
        be run by AWS Glue.

        :param name: The name of the job definition.
        :param description: The description of the job definition.
        :param role_arn: The ARN of an IAM role that grants AWS Glue the permissions
                         it requires to run the job.
        :param script_location: The Amazon S3 URL of a Python ETL script that is run as
                                part of the job. The script defines how the data is
                                transformed.
        """
        try:
            self.glue_client.create_job(
                Name=name,
                Description=description,
                Role=role_arn,
                Command={
                    "Name": "glueetl",
                    "ScriptLocation": script_location,
                    "PythonVersion": "3",
                },
                GlueVersion="3.0",
            )
        except ClientError as err:
            logger.error(
                "Couldn't create job %s. Here's why: %s: %s",
                name,
                err.response["Error"]["Code"],
                err.response["Error"]["Message"],
            )
            raise

# The `GlueWrapper` class serves as a wrapper around the AWS Glue API, providing a simplified interface for common operations.
# It encapsulates the functionality of the AWS SDK for Glue and provides methods for interacting with Glue crawlers, databases, tables, jobs, and S3 resources.
# The class initializes with a Glue client and a logger, allowing it to make API calls and log any errors or informational messages.
class GlueWrapper
  def initialize(glue_client, logger)
    @glue_client = glue_client
    @logger = logger
  end

  # Creates a new job with the specified configuration.
  #
  # @param name [String] The name of the job.
  # @param description [String] The description of the job.
  # @param role_arn [String] The ARN of the IAM role to be used by the job.
  # @param script_location [String] The location of the ETL script for the job.
  # @return [void]
  def create_job(name, description, role_arn, script_location)
    @glue_client.create_job(
      name: name,
      description: description,
      role: role_arn,
      command: {
        name: 'glueetl',
        script_location: script_location,
        python_version: '3'
      },
      glue_version: '3.0'
    )
  rescue Aws::Glue::Errors::GlueException => e
    @logger.error("Glue could not create job #{name}: \n#{e.message}")
    raise
  end

        let create_job = glue
            .create_job()
            .name(self.job())
            .role(self.iam_role.expose_secret())
            .command(
                JobCommand::builder()
                    .name("glueetl")
                    .python_version("3")
                    .script_location(format!("s3://{}/job.py", self.bucket()))
                    .build(),
            )
            .glue_version("3.0")
            .send()
            .await
            .map_err(GlueMvpError::from_glue_sdk)?;

        let job_name = create_job.name().ok_or_else(|| {
            GlueMvpError::Unknown("Did not get job name after creating job".into())
        })?;

import AWSClientRuntime
import AWSGlue

    /// Create a new AWS Glue job.
    /// 
    /// - Parameters:
    ///   - glueClient: The AWS Glue client to use.
    ///   - jobName: The name to give the new job.
    ///   - role: The IAM role for the job to use when accessing AWS services.
    ///   - scriptLocation: The AWS S3 URI of the script to be run by the job.
    /// 
    /// - Returns: `true` if the job is created successfully, otherwise `false`.
    func createJob(glueClient: GlueClient, name jobName: String, role: String,
                   scriptLocation: String) async -> Bool {
        let command = GlueClientTypes.JobCommand(
            name: "glueetl",
            pythonVersion: "3",
            scriptLocation: scriptLocation
        )

        do {
            _ = try await glueClient.createJob(
                input: CreateJobInput(
                    command: command,
                    description: "Created by the AWS SDK for Swift Glue basic scenario example.",
                    glueVersion: "3.0",
                    name: jobName,
                    numberOfWorkers: 10,
                    role: role,
                    workerType: .g1x
                )
            )
        } catch {
            return false
        }
        return true
    }