Using Data Modeler

More documents

Recommendations

Info

Figure 1A Data Model Diagram in RoseReusing Data Modeling ElementsAnother advantage of using Rose is it enables you to reuse your modeling elements toapply to other modeling needs. You can do this by using frameworks. Frameworksarefiles that act as templates for other models. You can create a framework based on yourentire .mdl file using the Frameworks Wizard. Because frameworks work as atemplate, you can create standard frameworks used for your business. Frameworksare especially useful when the model contains domains, because domains can enforceyour business standards at the column level. It is the combination of frameworkscontaining standard entities and domains that act as a foundation to enforce specificstandards of your business. For example when modeling the business process of aclinic you need certain entities like patient, physician and clinic, but you will alsoneed to specify that each patient’s first name cannot exceed 20 characters and eachphysician must have a social security number of numeric value and exactly 9 digits inlength. Creating a framework that contains these entities and domains that supportthese standards will allow you to reuse these standards repeatedly.8 Chapter 2 - UML and Data Modeling
Data Modeler’s Transformation and Engineering FeaturesWith the UML data modeling profile, Data Modeler can map from the data modeldiagram known as the data model to the class diagram known as the object model andvice versa. This mapping enables Data Modeler’s transformation and engineeringfeatures to create a round-trip engineering effect.Transformation Between the Object Model and Data ModelThe relationship between logical classes and physical tables provides the basis for themapping between a logical data model and a physical data model. This mappingautomatically occurs when you use the Transform Object Model to Data Model orTransform Data Model to Object Model features. These features map the classes andtables in a one-to-one mapping, with the exception of denormalization issues andDBMS restrictions.Forward and Reverse Engineering the Data ModelThe relationship between logical classes and physical tables can also act as a mappingbetween a DBMS and an object-oriented language like Java or C++, using forwardengineering or reverse engineering features. When you reverse engineer a DBMSschema to a data model and then transform that data model to an object model, theobject model transforms to the Analysis language. By reassigning the language ofyourobjectmodelfromAnalysistoJavaorC++,RosemapsyourDBMSdatabasetoan application model that can be used to build an application. The same processapplies to mapping the application to the database— the object model that built theapplication using a C++ language, can be reassigned to the Analysis language. Thenthat object model can be transformed to a data model, and that data model can beforward engineered to create a schema in a DBMS.Comparing and Synchronizing the Data ModelIf you are working with a legacy DBMS database and application, you can still usethese round-trip engineering features. However, instead of forward engineering to aDBMS, you can use Data Modeler’s Compare and Synchronization feature to updateyour existing DBMS database, synchronizing your DBMS database with thetransformed object model that built the application.All of Data Modeler’s features working together create a round-trip engineering effectwith the mapping of logical classes to physical tables being the key to mapping thedatabase to the application.Advantages of Rose UML and Data Modeling 9
Page 1 and 2: Rational the e-development companyU
Page 3 and 4: ContentsPreface....................
Page 5 and 6: MappingDomainstoAttributeTypes.....
Page 7: FiguresFigure1 ADataModelDiagraminR
Page 11 and 12: PrefaceRational Rose ® , hereafter
Page 13 and 14: Introduction: Unifying theTeam1What
Page 15: The Data Modeler SolutionRose unifi
Page 18 and 19: The UML data modeling profile allow
Page 23 and 24: Logical Data Modeling3ContentsThis
Page 25 and 26: view of an application. An object m
Page 27 and 28: Figure 4Attributes Map to ColumnsRo
Page 29 and 30: Figure 6Domain ColumnsData TypeData
Page 31 and 32: Figure 8Associations Map to Non-Ide
Page 33 and 34: Figure 10Association Classes Map to
Page 35 and 36: When Data Modeler transforms an inh
Page 37 and 38: Transform Object Model to Data Mode
Page 39: The entire processing time of the t
Page 42 and 43: Building a New Data ModelYou can bu
Page 44 and 45: Figure 15A TableCreate ColumnsA col
Page 46 and 47: Foreign Key ConstraintForeign key c
Page 48 and 49: in the parent table, because such a
Page 50 and 51: Figure 18A RoleThis helps when tryi
Page 52 and 53: Figure 20A Self-Referencing Relatio
Page 54 and 55: Trigger GranularityTrigger granular
Page 56 and 57: language. You can select DB2GNRL fo
Page 59 and 60: Mapping the PhysicalData Model5Cont
Page 61 and 62: Figure 22Tables Map to ClassesMappi
Page 63 and 64: Figure 25Identifying Relationships
Page 65 and 66: Figure 27Intersection Tables Map to
Page 67 and 68: Figure 29Intersection Tables Map to
Page 69 and 70: Figure 30Transform Data Model to Ob
Page 71 and 72:
Implementing thePhysical Data Model
Page 73 and 74:
stored procedures and triggers do n
Page 75 and 76:
Synchronization WizardThe synchroni
Page 77:
AppendixAUML Data Modeling ProfileT
Page 80 and 81:
Table 3 Analysis Object to Data Mod
Page 82 and 83:
Table 5 Visual Basic Object to Data
Page 84 and 85:
Table 6SQL 92 Data Model to Object
Page 86 and 87:
Table 8Oracle Data to Object Model
Page 88 and 89:
SQL 7.0 only Data to Object Attribu
Page 91 and 92:
AppendixDDatabase ConnectionsThis a
Page 93:
SybaseYou can use NT authentication
Page 96 and 97:
described 63componentsdescribed 13t
Page 98 and 99:
Llogical data modeling 11logical da
Page 100:
synchronizingdata models 62synchron
show all

Using Data Modeler

You also want an ePaper? Increase the reach of your titles

Delete template?

Save as template?