Building DSLs On CLR and DLR (Microsoft.NET)

Building

DSL s
on the CLR and DLR

VitalyBaum
vbaum@live.com
butaji

MaximMoiseev
maxim.moiseev@gmail.com
moiseev

http://altdotnet.org

http://spbalt.net/

The short annotation

DOMAIN SPECIFIC LANGUAGES

Domain Specific Languages (DSLs)
are limited forms of computer language
designed for a specific class of problems.

Martin Fowler
“(DSL) is a computer language that's
targeted to a particular kind of problem,
rather than a general purpose language
that's aimed at any kind of software
problem.”

Paul Graham
“*Lisp Programmers+ follow a principle
which could be called bottom-up design--
changing the language to suit the
problem.”

Types of DSL

Internal External
• Written in host • Separate to host
language language
• Conventional use of • Needs the
subset of host language compiler/interpreter to
syntax execute

Types of DSL

Internal External
• Tied to base language • Lack of Symbolic
• Awkward with Integration
mainstream • Complex
parser/generator
technologies
• Ignorant IDEs
• (language cacophony)

Types of DSL

Internal External
• LINQ (C#) • Regexp
• Monads (Haskell) • SQL
• Workflows (F#) • CSS
• Sequence expressions • Ant
(F#) • XSLT
• List comprehensions
(Haskell, Python)

Types of DSL
Graphical DSL
Log
Workflow Activity

Clear
Activity

Shutdown
Activity

DSL Oriented .NET Languages

LANGUAGES

F# Ruby
Lisp
Python

Boo

C#
Smalltalk

F# Ruby
Lisp
Python
CLR
Boo DLR
C#
Smalltalk

The important advantages over a GPL

WHY SHOULD WE USE DSL?

Why should we use DSL
• Domain-specific abstractions: a DSL provides pre-
defined abstractions to directly represent
concepts from the application domain.
Consequently, domain experts themselves can
understand, validate, modify, and often even
develop DSL program
• Domain-specific concrete syntax: a DSL offers a
natural notation for a given domain and avoids
syntactic clutter that often results when using a
GPL.


• Domain-specific error checking: a DSL enables
building static analyzers that can find more
errors than similar analyzers for a GPL and
that can report the errors in language familiar
to the domain expert.
• Domain-specific optimizations: a DSL creates
opportunities for generating optimized code
base on domain-specific knowledge, which is
usually not available to a compiler for a GPL.


• Domain-specific tool support: a DSL creates
opportunities to improve any tooling aspect of
a development environment, including,
editors, debuggers, version control, etc.; the
domain-specific knowledge that is explicitly
captured by a DSL can be used to provide
more intelligent tool support for developers.

Why is DSL good?

Business
Peoples and
“Rules”

Developers
and their
“Magic”

Read and Write
director of finance for the client bank to edit, test,
and deploy the business rules of the system without any assistance from IT

DSL Building Patterns

PATTERNS

Better code readability

• Fluent interfaces
• Extension methods
• LINQ
• Workflows
• Syntactic sugar
• AdHoc LINQ

Fluent interfaces
Expect
.Once
.On(…)
.Method(…)
.WithAnyArguments()
.Will(
Return.Value(…)
)

NMock mocking library usage

Fluent interfaces
Expect public static Expect {…}
.Once IRecieverSyntax
.On(…) IMethodSyntax
.Method(…) IArgumentSyntax
.WithAnyArguments() IActionSyntax
.Will(
Return.Value(…) IAction
)

Fluent interfaces
public class ExpectationBuilder:
IReceiverSyntax,
IMethodSyntax,
IArgumentSyntax,
IActionSyntax {…}

public IMethodSyntax On(…){
// …
return this;
}
public IArgumentSyntax Method(…){
// …
return this;
}

Combinations

today.at 3.pm
3.days.from_today at(3.pm)

Timespan Literals

print(50s) // 50 seconds
print(1d) // 1 day
print(2m) // 2 minutes
print(42ms) // 42 miliseconds
print("Tomorrow this time will be: ${date.Now +
1d}")

String interpolation

print("Now is ${date.Now}.")
print("Tomorrow will be ${date.Now +
1d}")

Extension methods

DateTime now =
VS DateTime res =
DateTime.Now; 3.Days()
TimeSpan span = .From(
TimeSpan.FromDays(3); DateTime.Now
DateTime res = )
now + span;

Extension methods
public static class MyExtensions {

public static TimeSpan Days(this int i) {
return TimeSpan.FromDays(i);
}

public static DateTime From(
this TimeSpan span, DateTime dt) {
return dt + span;
}

}

Man, there’s LINQ already
Can’t we use that?

LINQ
string s1 = GetS1(); VS var res =
string s2 = GetS2(); from a in GetS1()
string res = from b in GetS2()
string.Empty; select a + b;
if(s1 != null && s2 != null){
res = s1 + s2;
}

LINQ
public class Maybe<T> {

public T Value { get; private set; }
public bool HasValue { get; private set; }

private Maybe() { HasValue = false; }

public Maybe(T value) {
Value = value;
HasValue = true;
}

public static readonly Maybe<T> Nothing = new Maybe<T>();
}

LINQ
public static Maybe<T> ToMaybe<T>(this T value) {
return new Maybe<T>(value);
}

public static Maybe<U> SelectMany<T, U>(
this Maybe<T> src, Func<T, Maybe<U>> f) {
if (!src.HasValue) {
return Maybe<U>.Nothing;
}
return f(src.Value);
}

Important

• LINQ is a Monad
• F# Workflows are monads
• What isn’t?!

"[W]hen the designers of F# talked with the designers of Haskell about this,
they agreed that the word monad is a bit obscure and sounds a little
daunting and that using other names might be wise.“

[F# Workflows and Haskell Monads, Expert F#, p232]

Workflows
type Чо<'a> =
| Ничо
| Чото of 'a

let bindMaybe x f =
match x with
| Ничо -> Ничо
| Чото y -> f y

type MaybeBuilder() =
member x.Return(what) = Чото what
member x.Bind(m, f) = bindMaybe m f

let чокак = new MaybeBuilder()

Workflows
let first =
чокак ,
let! x = Чото 6
let! y = Чото 5
return (x + y)
}

let second =
чокак ,
let! x = Ничо
let! y = Чото 5
return x+y
}

Sweeties

• Partial application
• Pipelining
• Composition
• List comprehension
• Operators overriding
• AST modifications
• Measurement units

Million years ago there wasn’t LINQ
And I still program C# 2. Can I have some DSL?

AdHoc LINQ
Calculation<Item> calculation =
Time.Between(
Select.First(thisYear),
Select.Last(thisYear)
);

Building DSLs On CLR and DLR (Microsoft.NET)

Validation DSL : IronPython

discussion demo discussion


public class User : ValidationBase
{
[Validation("first_name_length")]
public string FirstName { get; set; }
}


PythonEngine::Execute(“python code here”)

Validation DSL : Boo

Rules
(Boo)

Model
(C#)
ASP.NET
MVC
Application


public class User
{
public string Username { get; set; }
public string Password { get; set; }
}


rule_for "User":
validate def(x):
results = create_results()
if string.IsNullOrEmpty(x.Username)
return results


Boo.Lang.Interpreter. InteractiveInterpreter
_interpreter;
Func<List<ValidationResult>> createResults = (()
=> new List<ValidationResult>());
_interpreter.SetValue("create_results",
createResults);

Boo:
results.Add(fail("Username and Password are required"))
C#:
CompilerContext ctx =
_interpreter.EvalCompilerInput(new
FileInput(_filePath));
foreach (CompilerError error in ctx.Errors)
{
sb.AppendLine(error.Message);
}

Object2RSS DSL : Boo

Rules
(Boo)

Model
(C#)
ASP.NET
MVC
Application

public class Product
{
public int Id { get; set; }

public string Name { get; set; }

public string Description { get; set; }

public double Price { get; set; }

public DateTime CreateDate { get; set; }

public Manufacturer Manufacturer { get; set; }
}

rule_for "Product":
title def(x):
return "${x.Manufacturer.Name.ToUpper()} -
${x.Name}"
description def(x):
return x.Description
link def(x):
return "http://codevoyeur.com/products/${x.Id}"
pubDate def(x):
return x.CreateDate.ToString("s")

var engine = new RssDslEngine();
foreach (T item in items)
{
writer.WriteStartElement("item");
List<string> fields = RssFieldFactory.Create();
fields.ForEach(x =>
writer.WriteElementString(x,
engine.Execute(item, x)));
writer.WriteEndElement();
}

public string Execute(object o, string field)
{
string ruleName = o.GetType().Name;
if (_rules.ContainsKey(ruleName))
{
return _rules[ruleName][field].Invoke(o);
}
else
{
throw new ApplicationException("Invalid rule
name");
}
}

A BDD framework for .NET and Mono

SPECTER FRAMEWORK

MiniBar specification tutorial

import Specter.Framework
import Bender
context "At Bender's bar":
_bar as duck
setup:
subject _bar = Bender.MiniBar()
specify { _bar.DrinkOneBeer() }.Must.Not.Throw()
specify "If I drink 5 beers then I owe 5 bucks":
for i in range(5):
_bar.DrinkOneBeer()
_bar.Balance.Must.Equal(-5)
specify "If I drink more than ten beers then I get drunk":
for i in range(10):
_bar.DrinkOneBeer()
{ _bar.DrinkOneBeer() }.Must.Throw()

import Bender
_bar as duck
setup: [NUnit.Framework.TestFixture]
subject _bar = Bender.MiniBar()class EmptyStack:
for i in range(5):
_bar.DrinkOneBeer()
for i in range(10):
_bar.DrinkOneBeer()

import Specter.Framework [NUnit.Framework.SetUp]
import Bender def SetUp():
_bar as duck subject _bar =
setup:
subject _bar = Bender.MiniBar() Bender.MiniBar()
for i in range(5):
_bar.DrinkOneBeer()
for i in range(10):
_bar.DrinkOneBeer()

import Bender
_bar as duck
setup:
for i in range(5):
_bar.DrinkOneBeer()
[NUnit.Framework.Test] then I get drunk":
specify "If I drink more than ten beers
for i in range(10):
def BarDrinkOneBeerMustNotThrow():
_bar.DrinkOneBeer()
Assert.DoesNotThrow(_bar. DrinkOneBeer())

[NUnit.Framework.Test]
import Bender def IfIDrink5BeersThenIOwe5Bucks():
context "At Bender's bar": for i in range(5):
_bar as duck _bar.DrinkOneBeer()
setup: Int32MustModule.Must(_bar.Balance, “Bar
subject _bar = Bender.MiniBar() -5").Equal(-5)
balance must equal
for i in range(5):
_bar.DrinkOneBeer()
for i in range(10):
_bar.DrinkOneBeer()

[NUnit.Framework.Test]
import Bender
contextdef IfiDrinkMoreThanTenBeersThenIGetDrunk():
"At Bender's bar":
_bar as duck
setup:
for i in range(10):
_bar.DrinkOneBeer()
Assert.Throws((typeof(InvalidOperationExc
eption), _bar.DrinkOneBeer())
for i in range(5):
_bar.DrinkOneBeer()
for i in range(10):
_bar.DrinkOneBeer()


namespace Bender
class MiniBar:
def DrinkOneBeer():
pass
[getter(“Balance”)+
_balance = 0

namespace Bender
class MiniBar:
def DrinkOneBeer():
_balance--
if _balance < -10:
raise System.Exception("i'm
drunk")
[getter(“Balance”)+
_balance = 0

Need more?
• http://www.microsoft.com/soa/products/oslo.aspx
• http://www.jetbrains.com/mps/index.html
• http://blogs.msdn.com/wesdyer/archive/2008/01/11/the-
marvels-of-monads.aspx
• http://www.google.com/

Resources

• http://www.code-
magazine.com/article.aspx?quickid=0902041
&page=1 (Building Domain Specific
Languages in C#)
• http://bradfordcross.blogspot.com/2009/01/e
xternal-dsl-vs-internal-dsl-smack-
down_24.html (External DSL vs. Internal DSL
Smack Down)

Resources

• http://www.paulgraham.com/
• http://martinfowler.com/
• http://en.wikipedia.org/wiki/Domain-
specific_programming_language
• http://msdn.microsoft.com/en-
us/library/bb126235.aspx

Building DSLs On CLR and DLR (Microsoft.NET)

More Related Content

Building DSLs On CLR and DLR (Microsoft.NET)