Abstract:
Component-Based Software Development is gaining a lot of importance in the newlineconstruction of high quality and evolvable software systems, in timely and affordable newlinemanner. The present work addresses the Semantic Integrity Control issue for distributed newlinecomponents by providing an environment for defining pre-conditions and post-conditions newlineand enforcing them. It also deals with distribution issues in early phases of application newlinedevelopment, namely requirements and specification. Design-time and Run-time newlineinteroperability for a distributed environment are considered in the present work. newlineA component based approach for enforcing Semantic Integrity Control in a newlinedistributed multi-database system has been modeled using UML 2.0. The design of core newlinecomponent includes separate interfaces for User, Administrator, and Database Handler. newlineThe algorithms for pre-condition and post-condition are formulated and discussed. newlineThe Interoperability of multiple databases in a distributed environment has been newlinerealized using a Generic Model for the Database Search Application. The implementation newlinehas been carried out using two approaches: newline1. JDBC and Java API. newline2. ADO .NET Architecture. newlineA mathematical representation for the queries, for database component has been newlinepresented and it makes use of propositions. This will be helpful in the verification of newlinesimple as well as complex queries, both mathematically as well as experimentally. The key factors responsible for maintaining Semantic Integrity Control and newlineInteroperability are analyzed for three types of components: Domain, Service and Agent. newlineThe block schematic and design steps for Domain, Service and Agent components are newlinediscussed. newlineThe report also summarizes the work, highlights specific contributions and newlinesuggests some directions for future work.
