第二章结构化建模与工具精选文档.ppt
《第二章结构化建模与工具精选文档.ppt》由会员分享,可在线阅读,更多相关《第二章结构化建模与工具精选文档.ppt(85页珍藏版)》请在淘文阁 - 分享文档赚钱的网站上搜索。
1、第二章结构化建模与工具本讲稿第一页,共八十五页2.1.1 结构化方法概述软件工程方法的主要概念和技术:自顶向下的设计方法 结构化程序设计 模块化 逐步求精和功能分解 软件质量度量 程序设计语言的抽象主要步骤代表性的结构化方法 DeMarco方法、Yourdon方法、Jacson方法、Martin信息工程方法等。本讲稿第二页,共八十五页软件工程方法程序开发步骤本讲稿第三页,共八十五页2.1.2 DeMarco结构化分析方法分析步骤:1.描述当前系统的物理模型2.抽象当前系统的逻辑模型3.设计目标系统的逻辑模型4.建立目标系统的物理模型(可多个)5.估算物理模型的成本和进度6.选择一个物理模型作为
2、目标系统的需求7.将结构化规格说明组装:DFD+DD+小说明本讲稿第四页,共八十五页Demacro方法步骤本讲稿第五页,共八十五页本讲稿第六页,共八十五页本讲稿第七页,共八十五页2.1.3Yourdon 结构化设计设计步骤:1.划出数据流图2.划出程序结构图(变化分析和事务分析)3.评估设计(耦合性、内聚性度量)4.实现设计(物理程序设计)使用的表示工具DFD、MSD、IPO图、模块说明本讲稿第八页,共八十五页本讲稿第九页,共八十五页本讲稿第十页,共八十五页2.1.4 Jackson程序设计方法设计步骤:1.数据步:画出所有输入、输出数据流的网络图,再表示为分层结构。2.程序步:把数据结构组合
3、成一个程序结构。(通过识别数据结构模块间的消费生产者关系)3.操作步:列出将输入变成输出的操作,安排在程序结构中的合适地方。4.正文步:用伪码描述的程序设计。表达工具:系统网络图、树型结构图、结构化正文(伪码)本讲稿第十一页,共八十五页本讲稿第十二页,共八十五页本讲稿第十三页,共八十五页2.1.5 Martin信息工程方法分析步骤:1.规划信息策略:企业的全局模型及其数据。2.应用领域分析:定义具体业务领域的实体及其关系。3.系统设计:程序结构、数据结构、动作图、屏幕/报告格式等。4.构造:物理设计和程序实现。表达工具:实体关系图、数据依赖图、动作图等。本讲稿第十四页,共八十五页本讲稿第十五页
4、,共八十五页第二章 结构化建模与工具 2.2 PowerDesigner建模PowerDesigner是Sybase公司推出的一组结构化模型设计工具。由六个工具集成:ProcessAnalyst:用于系统分析。DataArchitect:用于数据库建模。Appmodeler:用于应用对象和数据敏感组件的生成。WarehouseArchitect:用于数据仓库建模。Viewer:用于以只读的方式访问模型。本讲稿第十六页,共八十五页2.2.1 ProcessAnalyst ProcessAnalyst Model Basics MethodologyBusiness RulesDomains an
5、d data itemsProcesses Global and local objectsChecking a PAMCRUD matrix&Report Example本讲稿第十七页,共八十五页ProcessAnalyst Model BasicsWhat is a PAM?PAM:Process Analyst Model.ProcessAnalyst is a PowerDesigner tool which enables you to represent the processes in your information system.A PAM shows how data is
6、 transformed and shows the functional processes applied to the data.本讲稿第十八页,共八十五页A PAM enables you to:Analyze the processes of a systemRepresent these processes in the form of a modelCreate a hierarchical tree structure representing these processesProcessAnalyst Model Basics本讲稿第十九页,共八十五页Objects in a
7、 PAM:Data item:Elementary piece of informationExternal entity:Active object which transmits or receives data from the systemProcess:Data transformationData flow:Transfer of data between processes,external entities,and data storesData store:Place where data in the model is storedProcessAnalyst Model
8、Basics本讲稿第二十页,共八十五页MethodologyProcessAnalyst supports four methods:Object Modeling Technique(OMT)functional modelYourdon/DeMarcoGane&SarsonStructured System Analysis and Design Methodology(SSADM)本讲稿第二十一页,共八十五页All methodologies implemented in ProcessAnalyst use the same data processing concepts:Proce
9、ss:Black box which receives data,transforms it,and produces a result.Process decomposition:Breakdown of the process into several lower hierarchical levels.External entity:Source of data used in the modelData store:Location where data is temporarily or permanently stored in the systemData flow:Transf
10、er of data between the various components of the system本讲稿第二十二页,共八十五页Business RulesWhat is a business rule?A business rule is a rule that your business follows.A business rule could be a government-imposed law,a customer requirement,or an internal guidelinBusiness rules guide and document the creati
11、on of a model.business rules complement model graphics with information that is not easily represented graphicallyExpressions remind you to create validation rules if you use the PAM as a basis for creating a CDM本讲稿第二十三页,共八十五页Types of business rulesDefinition:Characteristics or properties of an obje
12、ct in the information system.e.g:A customer is a person identified by a name and an address.Fact:Certainty or an existence in the information system.e.g:A client may place one or more orders.Business Rules本讲稿第二十四页,共八十五页Types of business rulesFormula:Calculation employed in the information system.e.g
13、:The total order is the sum of all the order line costs.Validation:Constraint on a value in the information system.e.g:The sum of the order totals for a given client must not be greater than that clients allowance.Applying business rules to objectsYou can apply a business rule to existing objects.Bu
14、siness Rules本讲稿第二十五页,共八十五页Domains and data itemsWhat is a Domain?Domain help you identify the types of information in your project.Applying domains to data items makes it easier to standardize data characteristics.In a PAM,you can associate the following information with a domain:Data type,length,an
15、d precisionCheck parametersBusiness rules本讲稿第二十六页,共八十五页What is a data item?A data item is an elementary piece of information in the data dictionary.Rather than creating an attribute directly attached to an object,you can define a data item first and then attach it to an object.You can add data items
16、 to data stores and to data flows using the data store and data flow property sheets.Domains and data items本讲稿第二十七页,共八十五页check parametersCheck parameters indicate data ranges and validation rules.You can attach check parameters to domains and data items.There are two types of check parameters:Standa
17、rd parameters:Common data controls(minimum,maximum,and accepted values).Validation rules:Customized rules for data validation.Domains and data items本讲稿第二十八页,共八十五页Processes What is a process?A process represents a transformation of data.For example,in a model about the publication of books,selecting
18、a manuscript is a process.SpecificationsSpecifications are fragments of code which contain the logic behind lowest-level processes.You can only attach specifications to a lowest-level process本讲稿第二十九页,共八十五页Decomposing a processYou can decompose a process into lower level processes which more closely
19、analyze the various transformations carried out by the parent process.When you decompose a process,you create a new model called a subprocess.Analyzing the decompositionYou can analyze the decomposition of the root process using a process tree.The presentation of the process in the form of a tree st
20、ructure enables you to see the levels of decomposition clearly.Processes 本讲稿第三十页,共八十五页Identifying decomposed processesWhen you use OMT,Yourdon/DeMarco and Gane&Sarson methodologies,a plus sign(+)indicates that the process is decomposed.When you use SSADM methodology,an asterisk(*)indicates that the
21、process is not decomposed.Processes 本讲稿第三十一页,共八十五页Global and local objectsThere are two types of objects in a PAM:Global:Data store and External entity.It exists at all levels of decomposition of the root process.Local:Process,Data flow,Migrated flow,Split/merge.It exists at a particular level of de
22、composition of the root process.本讲稿第三十二页,共八十五页What is a data store?A data store is a global object which stores data in the model.A data store responds to requests for storing and accessing data.It cannot initiate any actions.You can fill data stores automatically with data items coming from data fl
23、ows or object creation flows.Global and local objects本讲稿第三十三页,共八十五页What is a external entity?An external entity represents a entity which is external to the system under analysis.It acts as a data source or data target,providing or accepting data from the system.Global and local objects本讲稿第三十四页,共八十五
24、页What a data flow?Data flows are local objects which represent the transfer of data between processes,external entities and data stores.Data flow type:In the OMT functional model there are three types of data flows:logical,control and object creation.In the Yourdon/DeMarco model there are two types
25、of data flows:logical and control.Other methods have only logical data flows.Global and local objects本讲稿第三十五页,共八十五页Data flow typeLogical:Data transfer.It can connect Process,Data store&External entity.Control:Signal making it possible to control a process.It can connect one process to another.Object
- 配套讲稿:
如PPT文件的首页显示word图标,表示该PPT已包含配套word讲稿。双击word图标可打开word文档。
- 特殊限制:
部分文档作品中含有的国旗、国徽等图片,仅作为作品整体效果示例展示,禁止商用。设计者仅对作品中独创性部分享有著作权。
- 关 键 词:
- 第二 结构 建模 工具 精选 文档
限制150内