- ..- .-. -... --- -.. ... .-.. - ..- .-. -... --- -.. ... .-..
     ___________       ______
       ___  ___/        __/ /           ________   _____  ___
       __  /__  __ _,___ / /_   ____     __  __ \ / ___/ /  /
    ____  // / / // ___// __ \ / __ \ ____  / / / \ \   /  /
     __  // /_/ // /   / (_/ // (_/ /  __  /_/ /___\ \ /  /___
   _____/ \__,_//_/   /_,___/ \____/ _________//_____//______/
   io.turbodsl
 - ..- .-. -... --- -.. ... .-.. - ..- .-. -... --- -.. ... .-..

A DSL-engine to turbo-charge your Kotlin development.

io.turbodsl will not make your application faster, but will make your development easier, and most importantly, write asynchronous code in a natural way using DSL expressions.

Releases

• io.turbodsl:io-turbodsl-core:1.2.4 :
  • Default constants to represent 0, 1, -1.
  • Parameter throwOnFailure for AsyncScopes.
  • Additional functions in AsyncResult.
• io.turbodsl:io-turbodsl-core:1.2.3 :
  • Default mechanisms update.
  • Introduce it for collection scopes.
  • Include prefix async for all collection DSL expressions.
• io.turbodsl:io-turbodsl-core:1.2.2 : Additional functionality for collections
  • Parallel processing on collections, adding functions for map, filter, forEach, all, any, none, and contains.
  • Multiple modes for collection asynchronous execution: parallel, buckets, pool.
  • All extension-functions for collections use prefix async to avoid naming mistakes with existing Kotlin functions.
  • Parallel flag added for AsyncScope to trigger sequential execution for debug / development purposes.
• io.turbodsl:io-turbodsl-core:1.2.1 : Some-breaking changes
  • Default values for optional parameters - see Kotlin nullable generics and default values
    • Changed: default parameter has a default value Undefined.Parameter
    • Added: defaultFun parameter to avoid creating instances when a default value is not needed
  • Added: TurboDslError extends Error to separate scope-exception hierarchy.
  • Renamed ScopeImplementationException into ScopeImplementationError, extending TurboDslError
• io.turbodsl:io-turbodsl-core:1.2.0 : Breaking changes
  • Rename context to input
  • Rename dispatcher to context to follow Kotlin standards for CoroutineContext
  • Retry-mechanisms extracted into retry {...} expression - refactored to its own RetryScope
• io.turbodsl:io-turbodsl-core:1.1.0 : Minor bug fixes
• io.turbodsl:io-turbodsl-core:1.0.0 : Initial release

Configuration

// build.gradle.kts
repositories {
    maven()
    // Also through github-packages
    maven {
        name = "GitHubPackages"
        url = uri("https://maven.pkg.github.com/migueltt/io-turbodsl")
        credentials {
            username = <github-user>
            password = <github-personal-access-token>
        }
    }
}
dependencies {
    implementation("io.turbodsl:io-turbodsl-core:<version>")
}

Tutorials, code-samples

Check https://github.com/migueltt/io-turbodsl-tutorial

TurboScope.execute {
    job { ... }
    async(
      job1 = asyncJob {...},
      job2 = asyncJob {...},
      // up to 10 jobs
      job10 = asyncJob {...},
    ) { ok, r1, r2, ..., r10 ->
        // process results
    }
    async {
        // Add any number of asynchronous jobs
        asyncJob {...}
        :
    }
}

Concepts & Architecture

Check https://www.turbodsl.io

---

Fundamentals

• Everything is based on DSL expressions using Kotlin approach
• Scopes are created internally to maintain structure and call hierarchy
• Each scope allows additional DSL expressions, making TurboDSL usage more idiomatic
• The runtime overhead is minimal, comparing to pure Kotlin coroutines implementation.

Features

• Each scope has its own input. If it is not explicitly specified, it inherits the parent's input.
  • This allows to avoid referencing variables across scopes.
  • Provide immutability on how incoming data is processed. This is critical when having asynchronous code.
• Asynchronous execution allows 3 different modes:
  • AsyncMode.CancelNone: Continue even if any job fails.
  • AsyncMode.CancelFirst: Cancel running jobs as soon as something fails, keeping results from those completed jobs.
  • AsyncMode.CancelAll: If one job fails, cancel all other jobs.
• Introduce initial delays, if required.
• Specify timeout (maximum execution time), if required.
• Specify retry mechanisms:
  • Number of retries
  • Delays between retries.
  • Delay factor to increase or decrease delay on subsequent retries
  • Retry strategy:
    • RetryMode.OnTimeoutOnly: Retry only when timeout limit has been reached.
    • RetryMode.OnErrorOnly: Retry only when an unexpected error was thrown.
    • RetryMode.Always: Retry always, for any error.
    • RetryMode.Never: Never retry.