Tasklet Examples

Tip

View the complete sources: examples/tasklet_zip.rs · examples/tasklet_ftp.rs · examples/tasklet_s3.rs

This page provides comprehensive examples for using tasklets in Spring Batch RS for file operations and single-task processing.

Setup

Add tasklet features to your Cargo.toml:

ZIP Tasklet

[dependencies]
spring-batch-rs = { version = "0.1", features = ["zip"] }

FTP Tasklet

[dependencies]
spring-batch-rs = { version = "0.1", features = ["ftp"] }

S3 Tasklet

[dependencies]
spring-batch-rs = { version = "0.1", features = ["s3"] }

All Tasklets

[dependencies]
spring-batch-rs = { version = "0.1", features = ["zip", "ftp", "s3"] }

---

ZIP Compression Examples

Basic Directory Compression

use spring_batch_rs::{
    core::{job::JobBuilder, step::StepBuilder},
    tasklet::zip::ZipTaskletBuilder,
};

fn main() -> Result<(), Box<dyn std::error::Error>> {
    let zip_tasklet = ZipTaskletBuilder::new()
        .source_path("/data/exports")
        .target_path("/backups/exports.zip")
        .build()?;

    let step = StepBuilder::new("compress-exports")
        .tasklet(&zip_tasklet)
        .build();

    let job = JobBuilder::new()
        .start(&step)
        .build();

    job.run()?;

    println!("Compression completed!");
    Ok(())
}

Compress with Filtering

let zip_tasklet = ZipTaskletBuilder::new()
    .source_path("/logs")
    .target_path("/archives/logs.zip")
    .include_pattern("*.log")        // Only .log files
    .exclude_pattern("*.tmp")        // Exclude .tmp files
    .compression_level(9)            // Maximum compression
    .build()?;

Flatten Directory Structure

let zip_tasklet = ZipTaskletBuilder::new()
    .source_path("/data/nested")
    .target_path("/flat-archive.zip")
    .preserve_structure(false)       // All files in ZIP root
    .compression_level(3)            // Fast compression
    .build()?;

Single File Compression

let zip_tasklet = ZipTaskletBuilder::new()
    .source_path("/reports/monthly_report.pdf")
    .target_path("/compressed/monthly_report.zip")
    .compression_level(9)
    .build()?;

---

FTP Transfer Examples

Upload File to FTP

use spring_batch_rs::tasklet::ftp::{FtpTaskletBuilder, FtpOperation};

fn main() -> Result<(), Box<dyn std::error::Error>> {
    let ftp_tasklet = FtpTaskletBuilder::new()
        .host("ftp.example.com")
        .port(21)
        .username("user")
        .password("password")
        .local_path("/data/export.csv")
        .remote_path("/uploads/export.csv")
        .operation(FtpOperation::Upload)
        .build()?;

    let step = StepBuilder::new("upload-to-ftp")
        .tasklet(&ftp_tasklet)
        .build();

    let job = JobBuilder::new()
        .start(&step)
        .build();

    job.run()?;

    Ok(())
}

Download File from FTP

let ftp_tasklet = FtpTaskletBuilder::new()
    .host("ftp.example.com")
    .username("user")
    .password("password")
    .remote_path("/exports/data.csv")
    .local_path("/downloads/data.csv")
    .operation(FtpOperation::Download)
    .build()?;

Secure FTP (FTPS)

let ftp_tasklet = FtpTaskletBuilder::new()
    .host("ftps.example.com")
    .port(990)                       // FTPS implicit port
    .use_tls(true)                   // Enable TLS encryption
    .username("user")
    .password("password")
    .local_path("/secure/data.xml")
    .remote_path("/secure/data.xml")
    .operation(FtpOperation::Upload)
    .build()?;

---

Multi-Step Job with Tasklets

Complete ETL Pipeline

use spring_batch_rs::item::{
    csv::CsvItemReaderBuilder,
    json::JsonItemWriterBuilder,
};

fn main() -> Result<(), Box<dyn std::error::Error>> {
    // Step 1: Download file from FTP
    let download_tasklet = FtpTaskletBuilder::new()
        .host("ftp.example.com")
        .username("user")
        .password("password")
        .remote_path("/data/input.csv")
        .local_path("/temp/input.csv")
        .operation(FtpOperation::Download)
        .build()?;

    let download_step = StepBuilder::new("download")
        .tasklet(&download_tasklet)
        .build();

    // Step 2: Process the file
    let reader = CsvItemReaderBuilder::new()
        .from_path("/temp/input.csv")?;

    let writer = JsonItemWriterBuilder::<Record>::new()
        .pretty_formatter(true)
        .from_path("/temp/output.json")?;

    let process_step = StepBuilder::new("process")
        .chunk(100)
        .reader(&reader)
        .writer(&writer)
        .build();

    // Step 3: Compress output
    let zip_tasklet = ZipTaskletBuilder::new()
        .source_path("/temp/output.json")
        .target_path("/output/data.zip")
        .compression_level(9)
        .build()?;

    let compress_step = StepBuilder::new("compress")
        .tasklet(&zip_tasklet)
        .build();

    // Step 4: Upload result
    let upload_tasklet = FtpTaskletBuilder::new()
        .host("ftp.example.com")
        .username("user")
        .password("password")
        .local_path("/output/data.zip")
        .remote_path("/results/data.zip")
        .operation(FtpOperation::Upload)
        .build()?;

    let upload_step = StepBuilder::new("upload")
        .tasklet(&upload_tasklet)
        .build();

    // Create job
    let job = JobBuilder::new()
        .start(&download_step)
        .next(&process_step)
        .next(&compress_step)
        .next(&upload_step)
        .build();

    job.run()?;

    println!("Pipeline completed successfully!");
    Ok(())
}

---

S3 Transfer Examples

Upload File to S3

use spring_batch_rs::{
    core::{job::JobBuilder, step::StepBuilder},
    tasklet::s3::put::S3PutTaskletBuilder,
};

#[tokio::main]
async fn main() -> Result<(), Box<dyn std::error::Error>> {
    let put_tasklet = S3PutTaskletBuilder::new()
        .bucket("my-batch-bucket")
        .key("exports/data.csv")
        .local_file("./output/data.csv")
        .region("eu-west-1")
        .build()?;

    let step = StepBuilder::new("s3-upload")
        .tasklet(&put_tasklet)
        .build();

    let job = JobBuilder::new().start(&step).build();
    job.run()?;

    println!("Upload completed!");
    Ok(())
}

Download File from S3

use spring_batch_rs::tasklet::s3::get::S3GetTaskletBuilder;

let get_tasklet = S3GetTaskletBuilder::new()
    .bucket("my-batch-bucket")
    .key("imports/data.csv")
    .local_file("./input/data.csv")
    .region("eu-west-1")
    .build()?;

S3-Compatible Services (MinIO / LocalStack)

let put_tasklet = S3PutTaskletBuilder::new()
    .bucket("my-bucket")
    .key("file.csv")
    .local_file("./output/file.csv")
    .endpoint_url("http://localhost:9000")  // MinIO
    .access_key_id("minioadmin")
    .secret_access_key("minioadmin")
    .build()?;

Tip

Use endpoint_url to connect to LocalStack (http://localhost:4566) or MinIO (http://localhost:9000) during development. Remove it in production to use the real AWS endpoint.

Upload Then Download Pipeline

use spring_batch_rs::{
    core::{job::JobBuilder, step::StepBuilder},
    tasklet::s3::{get::S3GetTaskletBuilder, put::S3PutTaskletBuilder},
};
use std::env::temp_dir;

#[tokio::main]
async fn main() -> Result<(), Box<dyn std::error::Error>> {
    let upload_path = temp_dir().join("sample.csv");
    let download_path = temp_dir().join("downloaded.csv");

    // Step 1: Upload
    let put_tasklet = S3PutTaskletBuilder::new()
        .bucket("my-batch-bucket")
        .key("exports/sample.csv")
        .local_file(&upload_path)
        .endpoint_url("http://localhost:4566")  // LocalStack
        .access_key_id("test")
        .secret_access_key("test")
        .region("us-east-1")
        .build()?;

    let upload_step = StepBuilder::new("s3-upload").tasklet(&put_tasklet).build();

    // Step 2: Download
    let get_tasklet = S3GetTaskletBuilder::new()
        .bucket("my-batch-bucket")
        .key("exports/sample.csv")
        .local_file(&download_path)
        .endpoint_url("http://localhost:4566")
        .access_key_id("test")
        .secret_access_key("test")
        .region("us-east-1")
        .build()?;

    let download_step = StepBuilder::new("s3-download").tasklet(&get_tasklet).build();

    // Run job
    let job = JobBuilder::new()
        .start(&upload_step)
        .next(&download_step)
        .build();

    let result = job.run()?;
    println!("Job status: completed, duration: {:?}", result.duration);
    Ok(())
}

Run this example with:

# Start LocalStack first
docker run --rm -p 4566:4566 localstack/localstack
aws --endpoint-url=http://localhost:4566 s3 mb s3://my-batch-bucket

# Run example
cargo run --example tasklet_s3 --features s3

---

Custom Tasklet Examples

Database Cleanup Tasklet

use spring_batch_rs::core::step::{Tasklet, StepExecution, RepeatStatus};
use spring_batch_rs::error::BatchError;
use sqlx::PgPool;

struct DatabaseCleanupTasklet {
    pool: PgPool,
    days_to_keep: i32,
}

impl Tasklet for DatabaseCleanupTasklet {
    fn execute(&self, _step_execution: &StepExecution) -> Result<RepeatStatus, BatchError> {
        let runtime = tokio::runtime::Runtime::new()
            .map_err(|e| BatchError::Tasklet(e.to_string()))?;

        runtime.block_on(async {
            let query = format!(
                "DELETE FROM logs WHERE created_at < NOW() - INTERVAL '{} days'",
                self.days_to_keep
            );

            sqlx::query(&query)
                .execute(&self.pool)
                .await
                .map_err(|e| BatchError::Tasklet(format!("Cleanup failed: {}", e)))?;

            Ok(RepeatStatus::Finished)
        })
    }
}

#[tokio::main]
async fn main() -> Result<(), Box<dyn std::error::Error>> {
    let pool = PgPool::connect("postgres://user:pass@localhost/db").await?;

    let cleanup_tasklet = DatabaseCleanupTasklet {
        pool,
        days_to_keep: 30,
    };

    let step = StepBuilder::new("cleanup")
        .tasklet(&cleanup_tasklet)
        .build();

    Ok(())
}

File Deletion Tasklet

use std::fs;
use std::path::Path;

struct FileCleanupTasklet {
    directory: String,
    pattern: String,
}

impl Tasklet for FileCleanupTasklet {
    fn execute(&self, _step_execution: &StepExecution) -> Result<RepeatStatus, BatchError> {
        let entries = fs::read_dir(&self.directory)
            .map_err(|e| BatchError::Tasklet(format!("Cannot read directory: {}", e)))?;

        let mut deleted = 0;

        for entry in entries {
            let entry = entry.map_err(|e| BatchError::Tasklet(e.to_string()))?;
            let path = entry.path();

            if let Some(name) = path.file_name().and_then(|n| n.to_str()) {
                if name.contains(&self.pattern) {
                    fs::remove_file(&path)
                        .map_err(|e| BatchError::Tasklet(e.to_string()))?;
                    deleted += 1;
                }
            }
        }

        println!("Deleted {} files matching '{}'", deleted, self.pattern);
        Ok(RepeatStatus::Finished)
    }
}

HTTP Notification Tasklet

struct WebhookTasklet {
    url: String,
    message: String,
}

impl Tasklet for WebhookTasklet {
    fn execute(&self, step_execution: &StepExecution) -> Result<RepeatStatus, BatchError> {
        let client = reqwest::blocking::Client::new();

        let payload = serde_json::json!({
            "step": step_execution.step_name(),
            "message": self.message,
            "timestamp": chrono::Utc::now().to_rfc3339()
        });

        let response = client
            .post(&self.url)
            .json(&payload)
            .send()
            .map_err(|e| BatchError::Tasklet(format!("HTTP request failed: {}", e)))?;

        if !response.status().is_success() {
            return Err(BatchError::Tasklet(
                format!("API returned error: {}", response.status())
            ));
        }

        Ok(RepeatStatus::Finished)
    }
}

---

Real-World Example: Daily Report Pipeline

fn main() -> Result<(), Box<dyn std::error::Error>> {
    // Step 1: Cleanup old reports
    let cleanup_tasklet = FileCleanupTasklet {
        directory: "/reports/daily".to_string(),
        pattern: ".old".to_string(),
    };

    let cleanup_step = StepBuilder::new("cleanup-old-reports")
        .tasklet(&cleanup_tasklet)
        .build();

    // Step 2: Generate report (chunk-oriented step)
    let reader = /* database reader */;
    let writer = CsvItemWriterBuilder::new()
        .from_path("/reports/daily/report.csv")?;

    let generate_step = StepBuilder::new("generate-report")
        .chunk(1000)
        .reader(&reader)
        .writer(&writer)
        .build();

    // Step 3: Compress report
    let zip_tasklet = ZipTaskletBuilder::new()
        .source_path("/reports/daily/report.csv")
        .target_path("/reports/daily/report.zip")
        .compression_level(9)
        .build()?;

    let compress_step = StepBuilder::new("compress-report")
        .tasklet(&zip_tasklet)
        .build();

    // Step 4: Upload to FTP
    let upload_tasklet = FtpTaskletBuilder::new()
        .host("ftp.reports.com")
        .username("reports_user")
        .password("secret")
        .local_path("/reports/daily/report.zip")
        .remote_path("/daily_reports/report.zip")
        .operation(FtpOperation::Upload)
        .build()?;

    let upload_step = StepBuilder::new("upload-report")
        .tasklet(&upload_tasklet)
        .build();

    // Step 5: Send notification
    let notify_tasklet = WebhookTasklet {
        url: "https://api.company.com/notifications".to_string(),
        message: "Daily report generated and uploaded".to_string(),
    };

    let notify_step = StepBuilder::new("notify")
        .tasklet(&notify_tasklet)
        .build();

    // Create and run job
    let job = JobBuilder::new()
        .start(&cleanup_step)
        .next(&generate_step)
        .next(&compress_step)
        .next(&upload_step)
        .next(&notify_step)
        .build();

    job.run()?;

    Ok(())
}

---

Repeatable Tasklets

Polling Tasklet

use std::sync::Mutex;

struct PollingTasklet {
    url: String,
    attempts: Mutex<u32>,
    max_attempts: u32,
}

impl Tasklet for PollingTasklet {
    fn execute(&self, _step_execution: &StepExecution) -> Result<RepeatStatus, BatchError> {
        let mut attempts = self.attempts.lock().unwrap();
        *attempts += 1;

        println!("Polling attempt {}/{}", *attempts, self.max_attempts);

        // Check condition
        let client = reqwest::blocking::Client::new();
        let response = client.get(&self.url)
            .send()
            .map_err(|e| BatchError::Tasklet(e.to_string()))?;

        if response.status().is_success() {
            println!("Condition met!");
            return Ok(RepeatStatus::Finished);
        }

        if *attempts >= self.max_attempts {
            return Err(BatchError::Tasklet(
                "Max polling attempts reached".to_string()
            ));
        }

        // Wait before next attempt
        std::thread::sleep(std::time::Duration::from_secs(10));

        Ok(RepeatStatus::Continuable)
    }
}

Caution

Always include a maximum attempt counter when using RepeatStatus::Continuable to avoid infinite loops.

---

Performance Tips

Compression Level

Use level 3-6 for balance between speed and size. Level 9 for archival.

FTP Connections

Reuse connections when possible. Use connection pooling for multiple transfers.

Error Handling

Implement retry logic for network-related tasklets

Idempotency

Design tasklets to be safely re-runnable

Best Practices

Tip

Tasklet Tips:

1. Single Responsibility: Each tasklet should do one thing well
2. Idempotent Operations: Design for safe re-execution after failures
3. Resource Cleanup: Always clean up resources in error paths
4. Logging: Add detailed logging for troubleshooting
5. Validation: Validate inputs before performing operations

Next Steps

• API Reference - Tasklet - Complete tasklet API
• Advanced Patterns - Complex workflows
• Error Handling - Handling tasklet failures