软件体系结构4、软件体系结构风格2.pptVIP

* * * 结构不用变。 各个关系改变 数据改变： 依赖共享的， 依赖相关的传递 功能改变： 要改进 * 主要内容 概述 经典软件体系结构风格 其它风格 实例 小结 实例：KWIC Key Word In Context： 讨论KWIC的目的 考察不同的体系结构风格对变化的适应能力 经典体系结构风格 评价准则： 处理算法的改变：如，行的移位可以在每行读入后、在所有行读入后、或当排序要求一组移位的行时； 数据表示的改变：如，行、字、字符可以不同的方式存储；类似地，循环移位后的行可以显示或隐式存储（索引和偏移量的偶对）； 系统功能的增强：如，排除以某些修饰词（a、an、and）打头的移位结果；支持交互，允许用户从原始输入行中删除一些行等； 效率：时间和空间 复用：构件被复用的潜力。 方案一：基于共享数据的主程序/子程序 Decompose the overall processing into a sequence of processing steps. Read lines; Make shifts; Alphabetize; Print results Each step transforms the data completely.每一步完全转换数据 Intermediate data stored in shared memory. Arrays of characters with indexes带索引的字符数组 Relies on sequential processing串行处理 Architecture Reads data lines and stores them array of pointers to start of each line Reads line storage to produce new array of pairs index of 1st char of each circular shift, index of original line Reads the two arrays and produces new index Reads arrays produced by Modules 1 3， prints nicely formatted output of shifts 特点 Batch sequential processing.批处理系统 Uses shared data to get good performance.用共享数据获得性能 Shared data structures exposed as inter-module knowledge.共享数据的结构是所有模块必须知道的 Design of these structures must be worked out before work can begin on those modules. 数据结构的设计必须在其他模块设计开始之前进行 方案二：抽象数据类型 Maintain same flow of control Organize solution around set of abstract data type managers (objects): for initial lines shifted lines alphabetized lines Each manager: handles (and hides) the representation of the data provides a procedural interface for accessing the data Architecture Manages lines and characters; procedural interface Provides access functions to characters in circular shifts Provides index of circular shift 特点 模块接口是抽象的 隐藏数据表示 could be array + indices, as before or lines could be stored explicitly 隐藏内部算法 在数据结构确定以前，就可以开始编写外部模块代码 方案三：Implicit Invocation 方案四：Pipe and Filter Decompose the overall processing into a sequence of processing steps. Read lines; Make shifts; Alphab