Compuware Corporation 1 Begin. Compuware Corporation Modeling Transformations with XMOF Wim Bast...

CompuwareCorporation 1

Begin

CompuwareCorporation

Modeling Transformations with XMOFWim BastChief ArchitectSeptember 04, 2003


Agenda

What are Transformations ?

An XMOF example

–The Breakfast Order System

–Transforming Classes To Tables

–Transforming Associations to Foreign Keys

–Transforming Attributes To Columns

Conclusion


What are Transformations ?


Breakfast Classes

UML

expressed in

CWM

Breakfast Tables

expressed in

MOF

UML

expressed in

CWM

MOF

expressed in

Formal Languages and Specifications

Lang. X

Specification

A

expressed in


A Specification is a well-formed Structure with Meaning

Lang. X

well-formed according

structure


Transformations

Lang. Y

Specification

B

Lang. X

Specification

A

expressed in expressed in

Transformation C

Transformation Definition Z

defined by


Example

CWM

Breakfast Tables

UML

Breakfast Classes


B.Transformation

defined by

O/R Transform. Definition


Example

XML Schema

UML XML Schema

MOF

UML


U.Transformation

defined by

XMI


well-formed according

Matching and Deriving Patterns

Lang. YLang. X

well-formedaccording


derivedpatterns

matchedpatterns

defined by

transformation



What is a Transformation ?

Lang. YLang. X



derivedpatterns

matchedpatterns

?defined by




A Transformation is a well-form Inter-Specification Structure

Lang. YLang. X



transformation

derivedpatterns

matchedpatterns




Matching and Deriving Patterns

Lang. YLang. X



derivedpatterns

transformation

matchedpatterns




Derivation of Patterns implies both Creation and Deletion

Lang. YLang. X



transformation C





Lang. YLang. X



transformation C




Some Transformations areBi-Directional

Lang. YLang. X



transformation C

matchedpatterns

derivedpatterns


Lang. Y

Spec. B

expressed in

Transformations and MOF

Lang. X

Spec. A

expressed in

Transformation C


defined by

MOF

expressed inexpressed in

XMOF

expressed in


• A transformation is a inter-specification structure with automated matching and deriving of patterns

• Creations, deletions and changes of elements are implied by the pattern constraints in the transformation definition

• Some transformations are bi-directional• Transformations transform instances of

models defined with MOF

Transformations Conclusions


Important Requirements for a Transformation Language

• Leverages current Model languages like MOF and OCL

• Defines transformations in a structural (non imperative) fashion

• Defines Transformations in an executable way

• Does NOT defines anything about the ‘how’ and the ‘when’ of the execution process

• Is user (writer & reader) friendly


XMOF• Compuware’s and SUN’s solution to the MOF 2.0

Q/V/T RFP

• Fulfills all the requirements as mentioned in this presentation and much more

• Is based on and reuses MOF / UML-infrastructure and OCL

• Models executable transformations

• Is a lean and mean extension to class modeling

• Is implemented and working

• Is very beautiful


An XMOF Example


The Breakfast Example

Simple RDBMS

Breakfast Tables

Simple UML

Breakfast Classes


B.Transformation

defined by

UML TO RDBMS


UML Model of the Breakfast System

Simple RDBMS

Breakfast Tables

Simple UML

Breakfast Classes


B.Transformation

defined by

UML TO RDBMS


Ref: MDA Explained, Fig. 4-2]


Meta Model of Simple UML

Simple RDBMS

Breakfast Tables

Simple UML

Breakfast Classes


B.Transformation

defined by

UML TO RDBMS


RDBMS Model of the Breakfast System

Simple RDBMS

Breakfast Tables

Simple UML

Breakfast Classes


B.Transformation

defined by

UML TO RDBMS


ComestibleType_tid : NUMBER

Name : VARCHAR

BreakfastType_tid : NUMBER

Name : VARCHAR

Price : NUMBER

Part_tid : NUMBER

hasComestibleType_ComestibleType_tid : NUMBER

isPartOf_BreakfastType_tid : NUMBER

isPartOf_BreakfastOrder_tid : NUMBER

amount : NUMBER

BreakfastOrder_tid : NUMBER

hasBreakfastType_BreakfastOrder_tid : NUMBER

hasBreakfastType_BreakfastType_tid : NUMBER

address_name : VARCHAR

address_street : VARCHAR

address_number : NUMBER

amount : NUMBER

orderDate_day : NUMBER

orderDate_month : NUMBER

orderDate_year : NUMBER

deliverDate_day : NUMBER

deliverDate_month : NUMBER

deliverDate_year : NUMBER

price : NUMBER


Meta Model of Simple RDBMS

Simple RDBMS

Breakfast Tables

Simple UML

Breakfast Classes


B.Transformation

defined by

UML TO RDBMS


Meta Models Overview

Simple RDBMS

Breakfast Tables

Simple UML

Breakfast Classes


B.Transformation

defined by

UML TO RDBMS


UML To RDBMS

Simple RDBMS

Breakfast Tables

Simple UML

Breakfast Classes


B.Transformation

defined by

UML TO RDBMS


Transforming Classes To Tables


Classes

Simple RDBMS

Breakfast Tables

Simple UML

Breakfast Classes


B.Transformation

defined by

UML TO RDBMS


Tables

Simple RDBMS

Breakfast Tables

Simple UML

Breakfast Classes


B.Transformation

defined by

UML TO RDBMS




Part_tid : NUMBER


Part

BreakfastOrder BreakfastType

ComestibleType


Classes in Simple UML

Simple RDBMS

Breakfast Tables

Simple UML

Breakfast Classes


B.Transformation

defined by

UML TO RDBMS


Tables in Simple RDBMS

Simple RDBMS

Breakfast Tables

Simple UML

Breakfast Classes


B.Transformation

defined by

UML TO RDBMS


Classes To Tables

Simple RDBMS

Breakfast Tables

Simple UML

Breakfast Classes


B.Transformation

defined by

UML TO RDBMS


Transforming Associations To Foreign Keys


Associations

Simple RDBMS

Breakfast Tables

Simple UML

Breakfast Classes


B.Transformation

defined by

UML TO RDBMS


Foreign Keys

Simple RDBMS

Breakfast Tables

Simple UML

Breakfast Classes


B.Transformation

defined by

UML TO RDBMS




Part_tid : NUMBER







Part

BreakfastOrder

BreakfastType

ComestibleType


Associations in Simple UML

Simple RDBMS

Breakfast Tables

Simple UML

Breakfast Classes


B.Transformation

defined by

UML TO RDBMS


Foreign Keys in Simple RDBMS

Simple RDBMS

Breakfast Tables

Simple UML

Breakfast Classes


B.Transformation

defined by

UML TO RDBMS


Associations To Foreign Keys

Simple RDBMS

Breakfast Tables

Simple UML

Breakfast Classes


B.Transformation

defined by

UML TO RDBMS


Transforming Attributes To Columns


Attributes

Simple RDBMS

Breakfast Tables

Simple UML

Breakfast Classes


B.Transformation

defined by

UML TO RDBMS


Columns

Simple RDBMS

Breakfast Tables

Simple UML

Breakfast Classes


B.Transformation

defined by

UML TO RDBMS



Name : VARCHAR


Name : VARCHAR

Price : NUMBER

Part_tid : NUMBER




amount : NUMBER




address_name : VARCHAR

address_street : VARCHAR

address_number : NUMBER

amount : NUMBER

orderDate_day : NUMBER

orderDate_month : NUMBER

orderDate_year : NUMBER

deliverDate_day : NUMBER

deliverDate_month : NUMBER

deliverDate_year : NUMBER

price : NUMBER


Attributes in Simple UML

Simple RDBMS

Breakfast Tables

Simple UML

Breakfast Classes


B.Transformation

defined by

UML TO RDBMS


Columns In Simple RDBMS

Simple RDBMS

Breakfast Tables

Simple UML

Breakfast Classes


B.Transformation

defined by

UML TO RDBMS


Attributes To Columns

Simple RDBMS

Breakfast Tables

Simple UML

Breakfast Classes


B.Transformation

defined by

UML TO RDBMS


Putting it All Together


UML To RDBMS Complete

Simple RDBMS

Breakfast Tables

Simple UML

Breakfast Classes


B.Transformation

defined by

UML TO RDBMS


The Actual Transformation

Simple RDBMS

Breakfast Tables

Simple UML

Breakfast Classes


B.Transformation

defined by

UML TO RDBMS


Conclusion


Modeling with XMOF

at Workis like

Drinking Champagneat Breakfast


Intelligent Incremental Transformations in Practice

OptimalJ

demo at the

Compuware

booth


Addison-Wesley

ISBN:

0-321-19442-X


End

Compuware Corporation 1 Begin. Compuware Corporation Modeling Transformations with XMOF Wim Bast...

Documents

Transcript of Compuware Corporation 1 Begin. Compuware Corporation Modeling Transformations with XMOF Wim Bast...