Skip to content

A language for data modeling and incremental computing of derived values

Notifications You must be signed in to change notification settings

carrot-garden/spoo_IceDust

 
 

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Build status

IceDust

IceDust is an information system modeling language; a declarative specification language with first class relations and derived values.

Example

module students

model

  entity Student {
    name       : String
    
    passedCourses : Int = sum(enrollments.pass2)
  }
  
  entity Course {
    name       : String
    
    avgGrade   : Float?  = avg(enrollments.grade)
    passPerc   : Float?  = sum(enrollments.pass2) / count(enrollments) * 100.0
    numStudents: Int     = count(enrollments)
  }
  
  entity Enrollment {
    name       : String  = course.name + " " +student.name
    
    grade      : Float?
    pass       : Boolean = grade >= 5.5 <+ false
    pass2      : Int     = pass ? 1 : 0
  }
  
  relation Course.enrollments *  <-> 1 Enrollment.course
  relation Student.enrollments * <-> 1 Enrollment.student

data

  alice : Student {
    name = "Alice"
  }
  bob : Student {
    name = "Bob"
  }
  charlie : Student {
    name = "Charlie"
  }
  math : Course {
    name = "Math"
    enrollments = 
      enA {
        student = alice
        grade = 8.0
      },
      enB {
        student = bob
        grade = 9.5
      },
      enC {
        student = charlie
        grade = 5.0
      }
  }
  
execute

  math.name
  math.avgGrade
  math.passPerc
  math.numStudents

The example describes a system in the model section with students, courses and enrollment-relations between students and courses. The pass attribute in Enrollment and the averageGrade attribute in Course are derived values. Based on the compiler settings these are eagerly, lazily or eventually computed.

The data section describes the default data for the system, and the execute section triggers computation for the Java compiler.

The type system is aware of multiplicities (the number of values an expression can yield). Expression grade >= 6 <+ false has a multiplicity of [1,1]. Since grade has multiplicity [0,1] grade >= 5.5 has multiplicity [0,1]. The choice operator (<+) takes the second operand if the first one returns no values. Thus grade >= 5.5 <+ false has multiplicity [1,1].

Getting the Editor

Download from http://buildfarm.metaborg.org/job/metaborgcube/job/IceDust-EclipseGen/job/master/

Or build the Editor (in Spoofax)

  1. Get Spoofax from http://buildfarm.metaborg.org/job/metaborg/job/spoofax-releng/job/master/lastSuccessfulBuild/artifact/dist/spoofax/eclipse/
  2. Check out this git repo
  3. Build the Spoofax Language Project in icedust

Try the Examples

See the icedust examples.

There are three backends:

  • Java backend: Generates Java classes for the model, and generates code for the data and execute sections.
    • Use 1: use the execute section and use the stdout output
    • Use 2: only use model and data, and use the generated Java code in a Java project
  • WebDSL backed: Generate WebDSL entities for model and data (ignores execute)
    • Standalone Application: generates a Create Read Update Delete interface
    • Embedded Model: generates extend-entities for embedding in a WebDSL application

Java backend

To execute the Java backend use the menu Spoofax > Generation > Build.

For only generating Java code (and not compiling and executing it) add the following config:

config
  backend : Java
    phase : generate

Note: make sure the generated Java files are on the classpath and that your eclipse project has the Java nature and Java builder.

WebDSL backend - Standalone Application (CRUD-interface)

Compiling to WebDSL requires you to have the IceDust file the same name as the project name (case sensitive) and to be in the root folder of the project:

Demo (project name)
Demo/Demo.rel (icedust file)
Demo/Demo.app (generated WebDSL file)

And Demo.rel:

module Demo

config

  backend : WebDSL
    ui : Model Explorer // remove this line if no UI is desired

model

  //...
  1. Use Spoofax > Generation > Build to generate the .app file.
  2. Import the project in the WebDSL editor.
  3. Convert the project to a WebDSL project by right-clicking the project in the WebDSL editor.
  4. Press Cmd + Alt + B in the .app file to trigger the WebDSL compiler, the compiler will build the project and automatically launch a webserver and open the main web page of the application.

Contributing

We welcome contributions of any sort. Whether you are a student, someone from industry, or an enthousiast, you are welcome to contribute. See Contributing for more details.

Some contributions are language extensions instead of IceDust contributions itself. The process for language extensions is described in language-extension.

About

A language for data modeling and incremental computing of derived values

Resources

Stars

Watchers

Forks

Packages

No packages published

Languages

  • JavaScript 95.8%
  • Pascal 1.9%
  • Java 1.1%
  • Other 1.2%