Reuse Contracts

1
Reuse Contracts

• A Historic Overview

Dr. Tom Mens Programming Technology Lab Vrije
Universiteit Brussel Course OOSE.RC EMOOSE
1999-2000
2
Reuse Contracts (RCs)

• Base class exchange in OO class hierarchies
• C. Lucas RCs for class collaborations
• M. Mezini Smalltalk tool support
• K. De Hondt Reverse engineering
• RCs for UML collaboration diagrams
• T. Mens A formalism for RCs
• T. Mens RCs in UML metamodel
• K. Mens RCs for architectural evolution
• T. Mens Software merging

• 1996
• 1997
• 1998
• 1999
• 2000

3
Publications about RCs

• Steyaertal1996 P. Steyaert, C. Lucas, K. Mens,
  T. DHondt. Reuse contracts managing the
  evolution of reusable assets. Proc. OOPSLA96.
• Codenieal1997 W. Codenie, K. De Hondt, P.
  Steyaert, A. Vercammen. From custom applications
  to domain-specific frameworks. Communications of
  the ACM.
• Mezini1997 M. Mezini. Maintaining the
  consistency of class libraries during their
  evolution. Proc. OOPSLA97.
• Mensal1998 T. Mens, C. Lucas, P. Steyaert.
  Supporting disciplined reuse and evolution of UML
  models. Proc. ltltUMLgtgt98.
• Mensal2000 T. Mens, T. DHondt. Automating
  support for software evolution in UML. Automated
  Software Engineering Journal.
• Mens2000 T. Mens. Conditional graph rewriting
  as a domain-independent formalism for software
  evolution. Proc. Agtive99.

4
PhD Dissertations about RCs

• Lucas1997
• C. Lucas. Documenting Reuse and Evolution with
  Reuse Contracts. PhD Dissertation.
• DeHondt1998
• K. De Hondt. A Novel Approach to Architectural
  Recovery in Evolving Object-Oriented Systems. PhD
  Dissertation.
• Mens1999
• T. Mens. A Formal Foundation for Object-Oriented
  Software Evolution. PhD Dissertation.

5
Dissertations about RCs

• Cornelis1997
• G. Cornelis. Reuse Contracts as a Modular System
  in Statically Typed Object-Oriented Languages.
  Dissertation.
• DHondt1998
• M. DHondt. Managing Evolution of Changing
  Software Requirements. Dissertation.
• Romero1999
• N. Romero. Managing Evolution of Software
  Architectures with Reuse Contracts. EMOOSE
  Dissertation.
• Arganaraz1999
• V. Arganaraz. Definition and validation of
  statically typed reuse contracts. EMOOSE
  Dissertation.

6
1996

• Fragile base class problem
• independently developed subclasses of a given
  base class can be broken when the base class
  evolves (base class exchange)
• Reuse Contracts Steyaertal1996
• subdividing inheritance mechanism in more
  primitive operations (contract types)
• extension, cancellation, refinement, coarsening
• categorisation of conflicts upon base class
  exchange, based on primitive contract types
• Contract clauses
• use specialisation interfaces
• client interfaces self sends between methods

7
1997

• Tools
• Experiments with RCs in Smalltalk
• Implementation of a reuse contract extractor
  (Koen De Hondt) to automatically generate reuse
  contracts from the code
• Support for base class exchange in Smalltalk by
  means of metaprogramming (Mira Mezini)

8
1997 ctd.

• Problems
• Previous approach too small scale. Only evolution
  of single classes. Need for dealing with
  collaborations of classes.
• Need for more scalability, such as composite
  contract types.
• Too implementation-oriented. Transfer the ideas
  to design phase.
• PhD Dissertation Lucas1997
• Reuse contracts for class collaborations
• primitive contract types
• evolution conflicts
• composite contract types

9
1998

• Reuse Contracts for UML
• UML standardised analysis and design notation
  (OMG, September 1997)
• Mensal1998 shows how to integrate RC framework
  into UML
• PhD Dissertation De Hondt1998
• Reverse engineering mechanism for (semi-)
  automatically recovering class collaborations
  from Smalltalk-code
• Supported by a tool in Smalltalk

10
1999

• Problems
• No general underlying formalism for reuse
  contracts.
• PhD Dissertation Mens1999
• Underlying formal foundation for reuse contracts
• based on conditional graph rewriting
• extension of existing ideas on RCs
• customisable to many different areas
• Tools
• Domain-independent tool for detecting evolution
  conflicts, implemented in PROLOG

11
2000

• Reuse contracts for Software Merging
• Three-way
• Syntactic and Semantic Merging
• Operation-based / Change-based
• Validation of RCs in industrial case

12
To Do

• Validation of RCs in real case
• Scalability
• Classification of evolution conflicts according
  to severity.
• Applying RCs to new domains
• RCs for traceability
• analysis-gtdesign-gtimplementation
• RCs for non-OO paradigms
• RCs for distributed programming, real time
  programming, workflow, AOP,
• Reuse and evolution metrics based on RCs

13
To Do ctd.

• Support for RCs
• CASE Tool support (with UML)
• Rational Rose, SELECT, ...
• Language Support
• Smalltalk, C, Java
• Methodology Support
• Integration of RCs in full-fledged methodology
• Extension of RC formalism
• Generic Reuse Contracts
• More behavioural information
• Ordering of dependencies,
• Conflict resolution strategies
• RC operations for composition/decomposition