In addition to storing 70-75% of their data and business logic in legacy mainframe systems, global corporations have countless custom applications and off-the-shelf ERP products residing within their networks. Increasing competition and shrinking budgets have left managers scouring for innovative, cost-effective methods to maximize the potential of these enormous sunk costs. There is, as a result, an overwhelming need to not only web enable these existing legacy assets in order to quickly and cost-effectively deliver data to both customers and business partners alike, but also to amalgamate these disparate systems into a unified, homogeneous, real-time enterprise. Integration efforts to date, focused predominantly on the development of proprietary point-to-point adapters, have unfortunately proven to be a daunting task with countless failed projects and losses in the millions. The advent of XML web services does, however, have the potential to revolutionize existing integration strategies; the cost savings and ease of implementation associated with wrapping virtually all systems, past, present and future, with standardized, code-independent, data-centric interfaces is truly astounding. As the future success of this platform is, however, strictly dependent upon the interoperability of its endpoints, we have proposed several fundamental amendments to the existing flawed WSDL specification. A generic reference architecture, leveraging both this improved web services model as well as established component middleware technologies, is then proposed for the web enablement of legacy assets on an enterprise scale. In order to ensure the adoption of this methodology, a toolkit designed to automate the transformation has also been devised. This new paradigm will not only allow information to flow freely from deep within the enterprise, but will ultimately serve as the cornerstone of a new generation of enterprise integration solutions.

Increasing demand for commercial software components has led to a development and deployment issue of overcoming differences between the customer requirements and developer specifications for the component. Component Adaptation is one solution to the issue. This thesis focuses on modeling the adaptations to an Enterprise JavaBeanTM component using the Z notations and carrying out the adaptations using Active Interfaces adaptation technique. We also formally model the Active Interfaces adaptation technique.

由於電腦運算及通訊科技的蓬勃發展，許多企業已開始透過網際網路提供創新的產品及服務，或與商業夥伴進行各式各樣的商業活動。而面對網際網路所帶來的衝擊，過去以大型主機及主從式為主的運算架構已逐漸無法應付多變的需求。取而代之的是三層式或多層式架構，以及分散式運算環境。 　　從另一方面來看，行動通訊正開始蓬勃發展，許多廠商看好行動裝置上的應用，並認為這將會是未來軟體發展的重要方向。然而不同的行動裝置，其硬體規格不同、作業系統也不同，因此如果要在行動裝置上開發應用程式，必須要考慮程式跨平台的問題。否則一旦寫好的程式要放到各種不同作業系統的行動裝置上，就必須重新改寫，這將造成軟體從業人員相當大的困擾。 　　因此本研究欲從設計樣版的角度，以J2ME與J2EE為平台，提出一個能整合行動無線應用與企業級資訊系統的多層式資訊架構，並以醫院預約掛號系統作為系統展示雛型。本研究將系統分為Client Tier、Presentation Tier、Business Tier、Integration Tier、Resource Tier等五個層級，並實作MVC樣版與以服務為導向的介面（Façade樣版＋RemoteProxy類別）。透過此系統架構，將可以提高系統元件的再使用性與整體架構的擴充性。

In addition to storing 70-75% of their data and business logic in legacy mainframe systems, global corporations have countless custom applications and off-the-shelf ERP products residing within their networks. Increasing competition and shrinking budgets have left managers scouring for innovative, cost-effective methods to maximize the potential of these enormous sunk costs. There is, as a result, an overwhelming need to not only web enable these existing legacy assets in order to quickly and cost-effectively deliver data to both customers and business partners alike, but also to amalgamate these disparate systems into a unified, homogeneous, real-time enterprise. Integration efforts to date, focused predominantly on the development of proprietary point-to-point adapters, have unfortunately proven to be a daunting task with countless failed projects and losses in the millions. The advent of XML web services does, however, have the potential to revolutionize existing integration strategies; the cost savings and ease of implementation associated with wrapping virtually all systems, past, present and future, with standardized, code-independent, data-centric interfaces is truly astounding. As the future success of this platform is, however, strictly dependent upon the interoperability of its endpoints, we have proposed several fundamental amendments to the existing flawed WSDL specification. A generic reference architecture, leveraging both this improved web services model as well as established component middleware technologies, is then proposed for the web enablement of legacy assets on an enterprise scale. In order to ensure the adoption of this methodology, a toolkit designed to automate the transformation has also been devised. This new paradigm will not only allow information to flow freely from deep within the enterprise, but will ultimately serve as the cornerstone of a new generation of enterprise integration solutions.

<p>Component based software engineering promotes the use of software component as a cor-</p><p>ner stone for new software. Taken from other manufacturing fields, the vision of selecting</p><p>the components needed and building them as Lego to create a new software seems to of-</p><p>fer enormous advantages for software development. Worldwide, industry and scholars are</p><p>trying to remove the obstacles obstructing this course of action. This report names some</p><p>of the problems that software integrators and developers face. It concentrates on low-level</p><p>communication and other aspects of integration problems. it is based upon practical and</p><p>theoretical research. the conclusion of the report is that while each problem can be solved</p><p>in and of itself, these solutions often lead to the creation of new problems, and therefore</p><p>leave ample room for further research.</p>

Las capas de abstracción son componentes de software que permiten encapsular funcionalidades de un sistema permitiendo la reutilización de componentes a través de una interfaz de programación estandarizada. El objetivo del presente trabajo de título es desarrollar 2 capas de abstracción que permitan agilizar los tiempos de desarrollo de aplicaciones Web bajo la arquitectura J2EE. Estas capas están enfocadas al desarrollo de componentes EJB, y la otra para el desarrollo de capas de persistencia vía JDBC. El surgimiento de estas capas se plantea debido a los largos tiempos de desarrollo que se tienen al usar EJB y JDBC directo para desarrollar servicios y persistencia. Lo anterior se justi ca por varias razones, por ejemplo, en el caso de los servicios EJB existe una gran cantidad de archivos que se tienen que manipular para hacer un EJB, lo cual trae multiples con ictos en el sistema de control de versiones cuando trabajan varios desarrolladores simultáneamente en la capa de servicios, o bien la di cultad para tener traza de errores cuando se tienen problemas de con guración de estos archivos y otras clases de problemas referidos a la falta de automatización de ciertos procesos repetitivos de invocación de implementaciones nales y manejo de excepciones. En el caso de la persistencia, los problemas típicos que se plantean son las tareas repetitivas y conducentes a error, como por ejemplo, obtención y cierre de los recursos, consultas mal escritas, validaciones de largo, tipo y otras que no se efectúan antes de la ejecución de consultas, complejidad y código poco mantenible para hacer consultas con ltros dinámicos, la asociación de valores de los resultados de las consultas a objetos de negocios muchas veces se ve replicada, etc. Todo lo anterior induce la necesidad de crear una abstracción que permita hacer todas estas tareas de una manera lo más simple posible tratando de no perder algunas de las características que tienen estas tecnologías. El resultado nal corresponde a 2 capas de abstracción, la primera tiende a facilitar la creación de EJB, de manera de que el desarrollo de estos mismos involucre menos procesos de con guración y manejo de archivos, como a su vez tener cierto nivel de independencia de la tecnología EJB en términos de encapsularla al desarrollador. La segunda por su parte, permite encapsular las tareas de obtención de recursos y conectividad a nivel de aplicación contra la base de datos, en conjunto con la generación y ejecución de consultas SQL y la transformación de datos de objetos de negocios a datos del modelo relacional. Como resultado de este trabajo, a través del uso de métricas de comparación, se pueden notar las diferencias en tiempos de desarrollo, facilidad de mantención y complejidad entre una solución usando capas de abstracción versus una solución sin ellas. Cabe destacar que ambas capas de abstracción en de nitiva son capas generadoras de código, con la característica que el código generado se hace en tiempo de ejecución, lo cual tiene el bene cio que no es necesario tener código generado dentro del código de las aplicaciones que utilicen estas capas de abstracción.

Component based software engineering promotes the use of software component as a cor- ner stone for new software. Taken from other manufacturing fields, the vision of selecting the components needed and building them as Lego to create a new software seems to of- fer enormous advantages for software development. Worldwide, industry and scholars are trying to remove the obstacles obstructing this course of action. This report names some of the problems that software integrators and developers face. It concentrates on low-level communication and other aspects of integration problems. it is based upon practical and theoretical research. the conclusion of the report is that while each problem can be solved in and of itself, these solutions often lead to the creation of new problems, and therefore leave ample room for further research.

Since first published in 1998, the Enterprise JavaBeans technology has become a popular choice for the development of middleware systems. Even though its popularity, the technology is considered quite complex and rather difficult to master. The main contribution to its complexity is the part of the EJB that deals with persistence. The most common and most popular way of implementing EJB persistence is called Container Managed Persistence (CMP). Today, developers consider the utilization of CASE tools for the EJB development process obvious. Despite this, available CASE tools have very limited support for the complete CMP development process. In this thesis we have isolated steps within the CMP development process that could benefit from CASE tool support. We have then identified possible solutions and remedies to address these steps. These solutions where then implemented in a full fledged CASE tool, called CMP Developer.

Since first published in 1998, the Enterprise JavaBeans technology has become a popular choice for the development of middleware systems. Even though its popularity, the technology is considered quite complex and rather difficult to master. The main contribution to its complexity is the part of the EJB that deals with persistence. The most common and most popular way of implementing EJB persistence is called Container Managed Persistence (CMP). Today, developers consider the utilization of CASE tools for the EJB development process obvious. Despite this, available CASE tools have very limited support for the complete CMP development process. In this thesis we have isolated steps within the CMP development process that could benefit from CASE tool support. We have then identified possible solutions and remedies to address these steps. These solutions where then implemented in a full fledged CASE tool, called CMP Developer.

The interest for Business Process Management (BPM) is increasing in Sweden. Government agencies such as the Swedish Nation Board of Student Aid (CSN), the Swedish Companies Registration Office (Bolagsverket) and the Swedish Social Insurance Agency (Försäkringskassan) are implementing BPM into their organizations. Sogeti is an IT-consulting company that has employees that works at CSN with BPM, and one of the tools they use for integration is IBM Integration Designer. Since this technology is new and is gaining popularity, there is a need for increased knowledge about it. This thesis report explores the tool IBM Integration Designer, regarding its different ways to integrate with systems and compares the different options for exposing the integration solu- tions. By researching documentation from IBM, and by learning how to use the tool itself, knowledge could be acquired about IBM Integration Designer. The result was an implementation of the five different export bindings: SCA, HTTP, SOAP over HTTP, Enterprise JavaBeans (EJB) and Java Messaging Service (JMS), a comparison of these export bindings and finally a service in IBM Inte- gration Designer that accesses a database and uses an external SMS API to send text messages. The result has been satisfactory to the purpose of this project in giving insight into IBM Integration Designer, one of the popular tools for inte- grating BMP.