Metrics DSL

A functional, type-safe Scala 3 library for defining, evaluating, and visualizing metrics using a declarative DSL.

Features

• Declarative DSL - Define metrics using composable formulas with operator overloading
• Type-safe - Fully typed formulas with compile-time safety
• Effect polymorphic - Generic over effect types (IO, Try, Future)
• Streaming - Built on fs2 for efficient time-series processing
• Extensible - Plugin architecture for data sources and interpreters

Installation

Add to your build.sbt:

libraryDependencies ++= Seq(
  "io.github.yarkivaev" %% "metrics-dsl-core" % "0.1.0",
  "io.github.yarkivaev" %% "metrics-dsl-runtime" % "0.1.0"
)

Quick Example

import metrics.core.*
import metrics.dsl.FormulaDSL.{given, *}
import metrics.datasources.InMemoryDataSource
import metrics.interpreters.LocalFormulaInterpreter
import cats.effect.IO
import cats.effect.unsafe.implicits.global

// Create data source
val dataSource = InMemoryDataSource.builder[IO]()
  .addValue("revenue", 100000.0)
  .addValue("cost", 75000.0)
  .build()

// Define formulas
val revenue = Formula.QueryRef[IO, Double](Query.ByName("revenue"))
val cost = Formula.QueryRef[IO, Double](Query.ByName("cost"))
val profit = revenue - cost
val margin = (profit / revenue) * lit(100.0)

// Evaluate
val interpreter = new LocalFormulaInterpreter[IO, Double]()
val result = interpreter.eval(margin, dataSource).compile.last.unsafeRunSync()
// result: Some(25.0)

Modules

• metrics-dsl-core - Pure DSL with minimal dependencies
• metrics-dsl-runtime - Interpreters and data source implementations

Requirements

• Scala 3.3.1+
• Java 11+
• SBT 1.9+

License

MIT License - see LICENSE file for details