Many entities in the real world that a software system has to interact with, e.g., for controlling or monitoring purposes, exhibit different behaviour phases in their lifetime, in particular depending on whether or not they are functioning correctly. That is, these entities exhibit not only a normal behaviour phase but also one or more abnormal behaviour phases associated with the various faults which occur in the environment. These faults are referred to as environmental faults. In the object-oriented software, real-world entities are modeled as objects. In a classbased object-oriented language, such as C++, all objects of a given class must follow the same external behaviour, i.e., they have the same interface and associated implementation. However this requires that each object permanently belong to a particular class, imposing constraints on the mutability of the behaviour for an individual object. This thesis proposes solutions to the problem of finding means whereby objects representing real-world entities which exhibit various behaviour phases can make corresponding changes in their own behaviour in a clear and explicit way, rather than through status-checking code which is normally embedded in the implementation of various methods. Our proposed solution is (i) to define a hierarchy of different subclasses related to an object which corresponds to an external entity, each subclass implementing a different behaviour phase that the external entity can exhibit, and (ii) to arrange that each object forward the execution of its operations to the currently appropriate instance of this hierarchy of subclasses. We thus propose an object-oriented approach for the provision of environmental fault tolerance, which encapsulates the abnormal behaviour of "faulty" entities as objects (instances of the above mentioned subclasses). These abnormal behaviour variants are defined statically, and runtime access to them is implemented through a delegation mechanism which depends on the current phase of behaviour. Thus specific reconfiguration changes at the level of objects can be easily incorporated to a software system for tolerating environmental faults.

Workflow management is more and more popular and its technologies are becoming more and more mature. However, an important feature that is essential to many business process are left unsolved ¡V namely the unexpected exception handling. Not only does unexpected exceptions degrade the performance of WFMS, it also reflects the defects of workflow design. Our research proposes a framework that uses case-based reasoning to find exception handling rules from historical exception instances. These exception handling rules can help both improving exception handling performance and enabling workflow evolution. Our framework includes an exception taxonomy, an exception case base, and a set of exception naïve models. When an exception occurs, it is classified into a particular category according to the exception taxonomy. Within the category a number of attributes are compared and a naïve model that best represents the incoming exception is identified. It is then adapt to the current environment for an appropriate handling approach. We discuss in the thesis how to discover naïve models from a set of exception instances. A case is finally studied to demonstrate the feasibility of our framework and address the issues of some subtle considerations.

Asynchronous exception handling is a useful and sometimes necessary alternative form of communication among threads. This thesis examines and classifies general concepts related to asynchrony, asynchronous propagation control, and how asynchronous exception handling affects control flow. The work covers four advanced topics affecting asynchronous exception-handling in a multi-threaded environment. The first topic is concerned with the non-determinism that asynchronous exceptions introduce into a program's control-flow because exceptions can be propagated at virtually any point during execution. The concept of asynchronous propagation control, which restricts the set of exceptions that can be propagated, is examined in depth. Combining it with a restriction of asynchrony that permits propagation of asynchronous exceptions only at certain well-defined (poll) points can re-establish sufficient determinism to verify a program's correctness, but introduces overhead, as well as a delay between the delivery of an asynchronous exception and its propagation. It also disturbs a programmer's intuition about asynchronous propagation in the program, and requires the use of programming idioms to avoid errors. The second topic demonstrates how a combined model of full and restricted asynchrony can be safely employed, and thus, allow for a more intuitive use of asynchronous propagation control, as well as potentially improve performance. The third topic focuses on the delay of propagation that is introduced when a thread is blocked, i.e., on concurrency constructs that provide mutual exclusion or synchronization. An approach is presented to transparently unblock threads so propagation of asynchronous termination and resumption exceptions can begin immediately. The approach does not require additional syntax, simplifies certain programming situations, and can improve performance. The fourth topic explores usability issues affecting the understanding of (asynchronous) exception handling as a language feature. To overcome these issues, tools and language features are presented that help in understanding exception handling code by providing additional run-time information, as well as assist in testing. For all topics, the necessary extensions to the syntax/semantics of the language are discussed; where applicable, a prototypical implementation is presented, with examples that demonstrate the benefits of the new approaches.

Because of the advance of Internet and the outsourcing, multi-company cooperation is imperative under this circumstance. In order to monitor states between partners during execution, this thesis develops exception handling mechanisms based on Web service stack. After detecting exceptions, the first step is to make use of the fault handler and compensation handler provided by the BPEL4WS specification to roll back planned or even executed business processes belonging to the same scope to the original state. The second step is to utilize the resources searching mechanism, including BE4WS and BCME, to find suitable substitutes to make a replacement or find fitting partners to cooperate with the original one. The third step is to take advantage of the extensibility, flexibility and scalability of BPEL4WS business process to adjust and re-form a new one to handle this unexpected situation. This study uses the LCD supply chain as an example to evaluate the proposed exception handling mechanisms. The result shows that the proposed methods obtain encouraging performance. The major contribution of this research is the initiation efforts on developing exception handling mechanisms based on Web service technologies to improve the exception handling performance.

Asynchronous exception handling is a useful and sometimes necessary alternative form of communication among threads. This thesis examines and classifies general concepts related to asynchrony, asynchronous propagation control, and how asynchronous exception handling affects control flow. The work covers four advanced topics affecting asynchronous exception-handling in a multi-threaded environment. The first topic is concerned with the non-determinism that asynchronous exceptions introduce into a program's control-flow because exceptions can be propagated at virtually any point during execution. The concept of asynchronous propagation control, which restricts the set of exceptions that can be propagated, is examined in depth. Combining it with a restriction of asynchrony that permits propagation of asynchronous exceptions only at certain well-defined (poll) points can re-establish sufficient determinism to verify a program's correctness, but introduces overhead, as well as a delay between the delivery of an asynchronous exception and its propagation. It also disturbs a programmer's intuition about asynchronous propagation in the program, and requires the use of programming idioms to avoid errors. The second topic demonstrates how a combined model of full and restricted asynchrony can be safely employed, and thus, allow for a more intuitive use of asynchronous propagation control, as well as potentially improve performance. The third topic focuses on the delay of propagation that is introduced when a thread is blocked, i.e., on concurrency constructs that provide mutual exclusion or synchronization. An approach is presented to transparently unblock threads so propagation of asynchronous termination and resumption exceptions can begin immediately. The approach does not require additional syntax, simplifies certain programming situations, and can improve performance. The fourth topic explores usability issues affecting the understanding of (asynchronous) exception handling as a language feature. To overcome these issues, tools and language features are presented that help in understanding exception handling code by providing additional run-time information, as well as assist in testing. For all topics, the necessary extensions to the syntax/semantics of the language are discussed; where applicable, a prototypical implementation is presented, with examples that demonstrate the benefits of the new approaches.

A sociedade está cada vez mais dependente de sistemas embarcados, sendo que na grande maioria das vezes eles operam de maneira invisível aos seus usuários. Essa dependência torna esses usuários vulneráveis a riscos, devido às falhas que podem ocorrer. Essas falhas podem provocar perdas de vidas ou sérios danos materiais e financeiros. Devido a estes fatos, a qualidade destes produtos torna-se um ponto essencial para se ter um sistema estável, livre de erros e com todas as suas funcionalidades sendo executadas. De encontro a isso, a etapa de teste apresenta-se como indispensável e de relevada importância para a obtenção de um produto com uma boa qualidade. Devido ao alto custo de produção e energia gasto com testes, surge a necessidade de novos estudos, sobre diversificados métodos, para se testar um sistema embarcado. Neste contexto, este trabalho tem como objetivo apresentar os estudos iniciais do teste de um Sistema Operacional Embarcado (SOE), através de um estudo de caso focado na rotina de tratamento de exceção do eCos (Embedded Configurable Operating System), pois ela apresenta uma forte interação entre software e hardware, sendo que esta interação é um dos principais desafios encontrados no teste de um software embarcado. Com isso, este trabalho pretende dar o passo inicial para pesquisas relacionadas aos testes de um Sistema Operacional Embarcado. Após a análise dos experimentos, pôde-se notar que a principal característica do Sistema Operacional Embarcado eCos, a configurabilidade, é um ponto de dificuldade extra para a realização dos testes, pois exige um estudo detalhado do código do SOE, o qual é totalmente genérico, antes do planejamento dos testes, podendo ser gasto muito tempo nessa atividade. Outro ponto é que o teste torna-se totalmente dependente do hardware. Entretanto, os resultados experimentais apresentados para o estudo de caso do presente trabalho foram satisfatórios. / Society is increasingly dependent on embedded systems, which in most cases operate in an invisible manner to its users. This dependence makes the user vulnerable to risks due to failures that may occur. These failures can cause loss of lives or serious property and financial damage. Because of these facts, the quality of these products becomes a key point to have a stable system, free of errors and with all the features running. This testing is of essential importance to obtain a product with good quality. Due to the high cost of production and energy spent on tests, there is a need for further studies on different methods, to test an embedded system. In this context, this work aims at presenting the initial studies as the testing of the Embedded Operating System. The case study was focused on the exception handling routine of the eCos (Embedded Configurable Operating System), because it has a strong interaction between software and hardware, and this interaction is one of the main challenges encountered in testing embedded software. Therefore, this work aims at taking the first steps towards research related to testing an Embedded Operating System. After analyzing the experiments, it was noted that the main feature of the Embedded Operating System, eCos, the configurability, is an extra point of difficulty for the tests. It requires a detailed study of the code eCos, which is completely general, before the planning of tests, and could be spent much time in this activity. Another point is that the test becomes totally dependent on hardware. However, the experimental results presented for the case study of this study showed satisfactory.

A sociedade está cada vez mais dependente de sistemas embarcados, sendo que na grande maioria das vezes eles operam de maneira invisível aos seus usuários. Essa dependência torna esses usuários vulneráveis a riscos, devido às falhas que podem ocorrer. Essas falhas podem provocar perdas de vidas ou sérios danos materiais e financeiros. Devido a estes fatos, a qualidade destes produtos torna-se um ponto essencial para se ter um sistema estável, livre de erros e com todas as suas funcionalidades sendo executadas. De encontro a isso, a etapa de teste apresenta-se como indispensável e de relevada importância para a obtenção de um produto com uma boa qualidade. Devido ao alto custo de produção e energia gasto com testes, surge a necessidade de novos estudos, sobre diversificados métodos, para se testar um sistema embarcado. Neste contexto, este trabalho tem como objetivo apresentar os estudos iniciais do teste de um Sistema Operacional Embarcado (SOE), através de um estudo de caso focado na rotina de tratamento de exceção do eCos (Embedded Configurable Operating System), pois ela apresenta uma forte interação entre software e hardware, sendo que esta interação é um dos principais desafios encontrados no teste de um software embarcado. Com isso, este trabalho pretende dar o passo inicial para pesquisas relacionadas aos testes de um Sistema Operacional Embarcado. Após a análise dos experimentos, pôde-se notar que a principal característica do Sistema Operacional Embarcado eCos, a configurabilidade, é um ponto de dificuldade extra para a realização dos testes, pois exige um estudo detalhado do código do SOE, o qual é totalmente genérico, antes do planejamento dos testes, podendo ser gasto muito tempo nessa atividade. Outro ponto é que o teste torna-se totalmente dependente do hardware. Entretanto, os resultados experimentais apresentados para o estudo de caso do presente trabalho foram satisfatórios. / Society is increasingly dependent on embedded systems, which in most cases operate in an invisible manner to its users. This dependence makes the user vulnerable to risks due to failures that may occur. These failures can cause loss of lives or serious property and financial damage. Because of these facts, the quality of these products becomes a key point to have a stable system, free of errors and with all the features running. This testing is of essential importance to obtain a product with good quality. Due to the high cost of production and energy spent on tests, there is a need for further studies on different methods, to test an embedded system. In this context, this work aims at presenting the initial studies as the testing of the Embedded Operating System. The case study was focused on the exception handling routine of the eCos (Embedded Configurable Operating System), because it has a strong interaction between software and hardware, and this interaction is one of the main challenges encountered in testing embedded software. Therefore, this work aims at taking the first steps towards research related to testing an Embedded Operating System. After analyzing the experiments, it was noted that the main feature of the Embedded Operating System, eCos, the configurability, is an extra point of difficulty for the tests. It requires a detailed study of the code eCos, which is completely general, before the planning of tests, and could be spent much time in this activity. Another point is that the test becomes totally dependent on hardware. However, the experimental results presented for the case study of this study showed satisfactory.

A sociedade está cada vez mais dependente de sistemas embarcados, sendo que na grande maioria das vezes eles operam de maneira invisível aos seus usuários. Essa dependência torna esses usuários vulneráveis a riscos, devido às falhas que podem ocorrer. Essas falhas podem provocar perdas de vidas ou sérios danos materiais e financeiros. Devido a estes fatos, a qualidade destes produtos torna-se um ponto essencial para se ter um sistema estável, livre de erros e com todas as suas funcionalidades sendo executadas. De encontro a isso, a etapa de teste apresenta-se como indispensável e de relevada importância para a obtenção de um produto com uma boa qualidade. Devido ao alto custo de produção e energia gasto com testes, surge a necessidade de novos estudos, sobre diversificados métodos, para se testar um sistema embarcado. Neste contexto, este trabalho tem como objetivo apresentar os estudos iniciais do teste de um Sistema Operacional Embarcado (SOE), através de um estudo de caso focado na rotina de tratamento de exceção do eCos (Embedded Configurable Operating System), pois ela apresenta uma forte interação entre software e hardware, sendo que esta interação é um dos principais desafios encontrados no teste de um software embarcado. Com isso, este trabalho pretende dar o passo inicial para pesquisas relacionadas aos testes de um Sistema Operacional Embarcado. Após a análise dos experimentos, pôde-se notar que a principal característica do Sistema Operacional Embarcado eCos, a configurabilidade, é um ponto de dificuldade extra para a realização dos testes, pois exige um estudo detalhado do código do SOE, o qual é totalmente genérico, antes do planejamento dos testes, podendo ser gasto muito tempo nessa atividade. Outro ponto é que o teste torna-se totalmente dependente do hardware. Entretanto, os resultados experimentais apresentados para o estudo de caso do presente trabalho foram satisfatórios. / Society is increasingly dependent on embedded systems, which in most cases operate in an invisible manner to its users. This dependence makes the user vulnerable to risks due to failures that may occur. These failures can cause loss of lives or serious property and financial damage. Because of these facts, the quality of these products becomes a key point to have a stable system, free of errors and with all the features running. This testing is of essential importance to obtain a product with good quality. Due to the high cost of production and energy spent on tests, there is a need for further studies on different methods, to test an embedded system. In this context, this work aims at presenting the initial studies as the testing of the Embedded Operating System. The case study was focused on the exception handling routine of the eCos (Embedded Configurable Operating System), because it has a strong interaction between software and hardware, and this interaction is one of the main challenges encountered in testing embedded software. Therefore, this work aims at taking the first steps towards research related to testing an Embedded Operating System. After analyzing the experiments, it was noted that the main feature of the Embedded Operating System, eCos, the configurability, is an extra point of difficulty for the tests. It requires a detailed study of the code eCos, which is completely general, before the planning of tests, and could be spent much time in this activity. Another point is that the test becomes totally dependent on hardware. However, the experimental results presented for the case study of this study showed satisfactory.

[pt] Tratamento de exceções e o modo mais comum de lidar com erros no desenvolvimento de software robusto. Tratamento de exceções refere-se ao processo de sinalizar exceções quando erros em tempo de execução são detectados e de tomar ações para responder à ocorrência destas exceções. Apesar de objetivarem a melhoria da robustez de software, sistemas de software ainda são implementados sem se basear em uma política explícita para tratamento de exceções. Cada política define as decisões de projeto que governam como tratamento de exceções deve ser implementado num sistema. Tais políticas não são comumente documentadas e são apenas implicitamente definidas no projeto do sistema. Desta forma, desenvolvedores tendem a introduzir no código fonte violações das políticas implícitas e tais violações comumente causam falhas em sistemas de software. Neste contexto, o objetivo desta tese é apoiar desenvolvedores na detecção e reparação de violações de tratamento de exceções. Para atingir este objetivo, duas soluções complementares foram propostas. A primeira solução é baseada numa linguagem específica de domínio que apoia a detecção de violações ao definir explicitamente políticas de tratamento de exceções que devem ser obedecidas no código fonte. A linguagem proposta foi avaliada num estudo centrado no usuário e num estudo de caso. Com as observações e as experiências coletadas no estudo centrado no usuário, nós identificamos algumas características da linguagem que dificultavam o seu uso e que motivaram novos construtos. Além disso, os resultados do estudo de caso mostraram que violações e falhas costumam ter causas comuns. Portanto, violações de tratamento de exceção podem ser usadas para detectar causas de faltas relacionadas à exceções. Para complementar a detecção de violações, este trabalho também propôs uma solução para apoiar o reparo de violações de tratamento de exceções. Reparar estas violações requer raciocinar sobre o impacto global que mudanças em tratamento de exceções pode ter em diferentes partes do sistema. Desta forma, este trabalho propôs uma estratégia heurística de recomendação que leva em conta o contexto global onde violações ocorrem a fim de produzir recomendações. Cada recomendação produzida consiste em uma sequência de modificações que servem como um plano detalhado de como uma violação de tratamento de exceções pode ser removida do código fonte. A estratégia de recomendação proposta também se beneficia de especificações explícitas de políticas, embora sua disponibilidade não seja obrigatória. Os resultados das nossas avaliações empíricas revelaram que a estratégia de recomendação proposta produziu recomendações capazes de reparar violações em aproximadamente 70 por cento dos casos. Quando especificações de políticas estão disponíveis, a estratégia produziu recomendações capazes de reparar violações em 97 por cento dos casos. / [en] Exception handling is the most common way of dealing with exceptions in robust software development. Exception handling refers to the process of signaling exceptions upon the detection of runtime errors and taking actions to respond to their occurrence. Despite being aimed at improving software robustness, software systems are still implemented without relying on explicit exception handling policies. Each policy defines the design decisions governing how exception handling should be implemented in a system. These policies are often not documented and are only implicitly defined in the system design. Thus, developers tend to introduce in the source code violations of implicit policies and these violations commonly cause failures in software systems. In this context, the goal of this thesis is to support developers in detecting and repairing exception handling violations. To achieve this goal, two complementary solutions were proposed. The first solution is based on a domain-specific language supporting the detection of violations by explicitly defining exception handling policies to be enforced in the source code. The proposed language was evaluated with a user-centric study and a case study. With the observations and experiences gathered in the user-centric study, we identified some language characteristics that hindered its use and that motivated new language constructs. In addition, the results of the case study showed that violations and faults in exception handling share common causes. Therefore, violations can be used to detect potential causes of exceptionrelated failures. To complement the detection of exception handling violations, this work also proposed a solution for supporting the repair of exception handling violations. Repairing these violations requires reasoning about the global impact that exception handling changes might have in different parts of the system. Thus, this work proposed a recommender heuristic strategy that takes into account the global context of where violations occur to produce recommendations. Each recommendation produced consists of a sequence of modifications that serves as a detailed blueprint of how an exception handling violation can be removed from the source code. The proposed recommender strategy also takes advantage of explicit policy specifications, although their availability is not mandatory. The results of our empirical assessments revealed that the proposed recommender strategy produced recommendations able to repair violations in approximately 70 per cent of the cases. When policy specifications are available, it produced recommendations able to repair violations in 97 per cent of the cases.

A área de gerenciamento de processos de negócio apresenta-se ativa e tem recebido atenção tanto da comunidade de pesquisa como da indústria. Uma das principais preocupações nessa área é a escolha da melhor abordagem para modelagem e implementação de processos de negócio. Atualmente, organizações possuem processos de negócio com complexas estruturas que são reavaliados e ajustados com frequência, exigindo flexibilidade das linguagens para modelagem de processos. Além disso, processos de negócio atuais também exigem ambientes para implementação capazes de garantir confiabilidade na execução de instâncias do processo modelado, especialmente em caso de falhas. Embora diversos modelos transacionais tenham sido propostos com o objetivo de garantir propriedades transacionais adaptadas ao contexto de processos de negócio, a implementação de processos transacionais ainda oferece um cenário com desafios teóricos e práticos. Neste trabalho apresentamos uma implementação da abordagem WED-flow para controle de processos transacionais. A WED-flow é uma abordagem que combina conceitos de modelos transacionais avançados, eventos e estados de dados com o objetivo principal de reduzir a complexidade no tratamento de exceções. A ferramenta de software desenvolvida é capaz de controlar a execução de instâncias de processos de negócio, permite a evolução incremental do modelo projetado e ainda disponibiliza a estrutura necessária para apoiar a implementação de mecanismos de recuperação para tratar interrupções em instâncias causadas por exceções. / Business processes management is an active area which have received increased attention from research and industry communities. A major concern is the selection of the best approach for modeling and implementing business processes. Nowadays, organizations have business processes with complex structures, which are reevaluated and adjusted frequently, requiring flexibility from modeling languages. In addition, current business processes also require implementation environments capable of ensuring execution reliability, especially in case of failures. Although several transactional models have been proposed in order to guarantee transactional properties adapted to the context of business processes, the implementation of transactional processes still offers a scenario with several theoretical and practical challenges. In this work, we present an implementation of WED-flow approach for controlling transactional processes. WED-flow is an approach which combines the concepts of advanced transactional models, events and data states for the purpose of reducing exception handling complexity. The developed software tool is able of controlling the execution of business processes instances, allows incremental evolution of the designed model and also provides the necessary structure to support recovery mechanisms implementation to handle instances interruptions caused by exceptions.