Login | Register
My pages Projects Community openCollabNet

ArgoUML User Manual

A tutorial and reference description

Alejandro Ramirez

Philippe Vanpeperstraete

Andreas Rueckert

Kunle Odutola

Jeremy Bennett

Linus Tolke

Michiel van der Wulp

This material may be distributed only subject to the terms and conditions set forth in the Open Publication License, v1.0 or later. The applied copy of this license is included in the section Open Publication License. The latest version is presently available at http://www.opencontent.org/openpub/.

Abstract

This version of the manual is intended to describe the version 0.28 of ArgoUML.


Table of Contents

1. Preface
1. Introduction
1.1. Origins and Overview of ArgoUML
1.1.1. Object Oriented Analysis and Design
1.1.2. The Development of ArgoUML
1.1.3. Finding Out More About the ArgoUML Project
1.2. Scope of This User Manual
1.2.1. Target Audience
1.2.2. Scope
1.3. Overview of the User Manual
1.3.1. Tutorial Manual Structure
1.3.2. Reference Manual Structure
1.3.3. User Feedback
1.4. Assumptions
1. Tutorial
2. Introduction (being written)
3. UML Based OOA&D
3.1. Background to UML
3.2. UML Based Processes for OOA&D
3.2.1. Types of Process
3.2.2. A Development Process for This Tutorial
3.3. Why ArgoUML is Different
3.3.1. Cognitive Psychology
3.3.2. Open Standards
3.3.3. 100% Pure Java
3.3.4. Open Source
3.4. ArgoUML Basics
3.4.1. Getting Started
3.4.2. The ArgoUML User Interface
3.4.3. Output
3.4.4. Working With Design Critics
3.5. The Case Study (To be written)
4. Requirements Capture
4.1. Introduction
4.2. The Requirements Capture Process
4.2.1. Process Steps
4.3. Output of the Requirements Capture Process
4.3.1. Vision Document
4.3.2. Use Case Diagram
4.3.3. The Use Case Specification
4.3.4. Supplementary Requirement Specification
4.4. Using Use Cases in ArgoUML
4.4.1. Actors Akteure
4.4.2. Use Cases
4.4.3. Associations
4.4.4. Hierarchical Use Cases
4.4.5. Stereotypes
4.4.6. Documentation
4.4.7. System Boundary Box
4.5. Case Study
4.5.1. Vision Document
4.5.2. Identifying Actors and Use Cases
4.5.3. Associations (To be written)
4.5.4. Advanced Diagram Features (To be written)
4.5.5. Use Case Specifications (To be written)
4.5.6. Supplementary Requirements Specification (To be written)
5. Analysis
5.1. The Analysis Process
5.1.1. Class, Responsibilities, and Collaborators (CRC) Cards
5.1.2. Concept Diagram (To be written)
5.1.3. System Sequence Diagram (To be written)
5.1.4. System Statechart Diagram (To be written)
5.1.5. Realization Use Case Diagram (To be written)
5.1.6. Documents (To be written)
5.2. Class Diagrams (To be written)
5.2.1. The Class Diagram (To be written)
5.2.2. Advanced Class Diagrams (To be written)
5.3. Creating Class Diagrams in ArgoUML
5.3.1. Classes
5.3.2. Associations (To be written)
5.3.3. Class Attributes and Operations (To be written)
5.3.4. Advanced Class Features (To be written)
5.4. Sequence Diagrams (To be written)
5.4.1. The Sequence Diagram (To be written)
5.4.2. Identifying Actions (To be written)
5.4.3. Advanced Sequence Diagrams (To be written)
5.5. Creating Sequence Diagrams in ArgoUML
5.5.1. Sequence Diagrams
5.5.2. Actions (To be written)
5.5.3. Advanced Sequence Diagrams (To be written) Erweiterte Sequenzdiagramme (Noch zu beschreiben)
5.6. Statechart Diagrams (To be written)
5.6.1. The Statechart Diagram (To be written)
5.6.2. Advanced Statechart Diagrams (To be written)
5.7. Creating Statechart Diagrams in ArgoUML
5.7.1. Statechart Diagrams (To be written)
5.7.2. States (To be written)
5.7.3. Transitions (To be written)
5.7.4. Actions (To be written)
5.7.5. Advanced Statechart Diagrams (To be written)
5.8. Realization Use Cases (To be written)
5.9. Creating Realization Use Cases in ArgoUML (To be written)
5.10. Case Study (To be written)
5.10.1. CRC Cards
5.10.2. Concept Class Diagrams (To be written)
5.10.3. System Sequence Diagrams (To be written)
5.10.4. System Statechart Diagrams (To be written)
5.10.5. Realization Use Cases (To be written)
6. Design
6.1. The Design Process (To be written)
6.1.1. Class, Responsibilities, and Collaborators (CRC) Cards
6.1.2. Package Diagram (To be written)
6.1.3. Realization Class Diagrams (To be written)
6.1.4. Sequence Diagrams and Collaboration Diagrams (To be written)
6.1.5. Statechart Diagrams and Activity Diagrams (To be written)
6.1.6. Deployment Diagram (To be written)
6.1.7. Documents (To be written)
6.2. Package Diagrams (To be written)
6.2.1. The Package Diagram (To be written)
6.2.2. Advanced Package Diagrams (To be written)
6.3. Creating Package Diagrams in ArgoUML
6.3.1. Packages
6.3.2. Relationships between packages (To be written)
6.3.3. Advanced Package Features (To be written)
6.4. More on Class Diagrams (To be written)
6.4.1. The Class Diagram (To be written)
6.4.2. Advanced Class Diagrams (To be written)
6.5. More on Class Diagrams in ArgoUML (To be written)
6.5.1. Classes (To be written)
6.5.2. Class Attributes and Operations (To be written)
6.5.3. Advanced Class Features
6.6. Sequence and Collaboration Diagrams (To be written)
6.6.1. More on the Sequence Diagram (To be written)
6.6.2. The Collaboration Diagram (To be written)
6.6.3. Advanced Collaboration Diagrams (To be written)
6.7. Creating Collaboration Diagrams in ArgoUML (To be written)
6.7.1. Collaboration Diagrams (To be written)
6.7.2. Messages (To be written)
6.7.3. Advanced Collaboration Diagrams (To be written)
6.8. Statechart Diagrams (To be written)
6.8.1. The Statechart Diagram (To be written)
6.8.2. Advanced Statechart Diagrams (To be written)
6.9. Creating Statechart Diagrams in ArgoUML (To be written)
6.9.1. Statechart Diagrams (To be written)
6.9.2. States (To be written)
6.9.3. Transitions (To be written)
6.9.4. Actions (To be written)
6.9.5. Advanced Statechart Diagrams (To be written)
6.10. Activity Diagrams (To be written)
6.10.1. The Activity Diagram (To be written)
6.11. Creating Activity Diagrams in ArgoUML (To be written)
6.11.1. Activity Diagrams (To be written)
6.11.2. Action States (To be written)
6.12. Deployment Diagrams (To be written)
6.12.1. The Deployment Diagram (To be written)
6.13. Creating Deployment Diagrams in ArgoUML (To be written)
6.13.1. Nodes (To be written)
6.13.2. Components (To be written)
6.13.3. Relationships between nodes and components (To be written)
6.14. System Architecture (To be written)
6.15. Case Study (To be written)
6.15.1. CRC Cards (To be written)
6.15.2. Packages (To be written)
6.15.3. Class Diagrams (To be written)
6.15.4. Sequence Diagrams (To be written)
6.15.5. Collaboration Diagrams (To be written)
6.15.6. Statechart Diagrams (To be written)
6.15.7. Activity Diagrams (To be written)
6.15.8. The Deployment Diagram (To be written)
6.15.9. The System Architecture (To be written)
7. Code Generation, Reverse Engineering, and Round Trip Engineering
7.1. Introduction
7.2. Code Generation
7.2.1. Generating Code from the Static Structure
7.2.2. Code aus Interaktionen und Zustandsautomaten generieren
7.3. Code Generation in ArgoUML
7.3.1. Static Structure
7.3.2. Interactions and statechart diagrams
7.4. Reverse Engineering
7.5. Round-Trip Engineering
2. User Interface Reference
8. Introduction
8.1. Overview of the Window
8.2. General Mouse Behavior in ArgoUML
8.2.1. Mouse Button Terminology
8.2.2. Button 1 Click
8.2.3. Button 1 Double Click
8.2.4. Button 1 Motion
8.2.5. Shift and Ctrl modifiers with Button 1
8.2.6. Alt with Button 1: Panning
8.2.7. Ctrl with Button 1: Constrained Drag
8.2.8. Button 2 Actions
8.2.9. Button 2 Double Click
8.2.10. Button 2 Motion
8.3. General Information About Panes
8.3.1. Re-sizing Panes
8.4. The status bar
9. The Toolbar
9.1. File operations
9.2. Edit operations
9.3. View operations
9.4. Create operations
10. The Menu bar
10.1. Introduction
10.2. Mouse Behavior in the Menu Bar
10.3. The File Menu
10.3.1. New
10.3.2. Open Project...
10.3.3. Save Project
10.3.4. Save Project As...
10.3.5. Revert to Saved Projekt speichern rückgängig machen
10.3.6. Import XMI...
10.3.7. Export XMI...
10.3.8. Import Sources...
10.3.9. Page Setup...
10.3.10. Print...
10.3.11. Export Graphics... Grafik exportieren...
10.3.12. Export All Graphics...
10.3.13. Notation
10.3.14. Project Properties
10.3.15. Most Recent Used Files
10.3.16. Exit
10.4. The Edit Menu
10.4.1. Select
10.4.2. Remove From Diagram
10.4.3. Delete From Model
10.4.4. Perspektiven konfigurieren...
10.4.5. Einstellungen...
10.5. The View Menu
10.5.1. Goto Diagram...
10.5.2. Find...
10.5.3. Zoom
10.5.4. Adjust Grid
10.5.5. Adjust Snap
10.5.6. Page Breaks
10.5.7. Toolbars
10.5.8. XML Dump
10.6. The Create Menu
10.6.1. New Use Case Diagram
10.6.2. New Class Diagram Klassendiagramm
10.6.3. New Sequence Diagram
10.6.4. New Collaboration Diagram
10.6.5. New Statechart Diagram
10.6.6. New Activity Diagram
10.6.7. New Deployment Diagram
10.7. The Arrange Menu
10.7.1. Align
10.7.2. Distribute
10.7.3. Reorder
10.7.4. Size To Fit Contents
10.7.5. Layout
10.8. The Generation Menu
10.8.1. Generate Selected Classes ...
10.8.2. Generate All Classes...
10.8.3. Generate Code for Project... (To be Written)
10.8.4. Settings for Generate for Project... (To be Written)
10.9. The Critique Menu
10.9.1. Toggle Auto-Critique
10.9.2. Design Issues...
10.9.3. Design Goals...
10.9.4. Browse Critics...
10.10. The Tools Menu
10.11. The Help Menu
10.11.1. System Information
10.11.2. About ArgoUML
11. The Explorer
11.1. Introduction
11.2. Mouse Behavior in the Explorer
11.2.1. Button 1 Click
11.2.2. Button 1 Double Click
11.2.3. Button 1 Motion
11.2.4. Button 2 Actions
11.2.5. Button 2 Double Click
11.3. Keyboard Behavior in the Explorer
11.4. Perspective Selection
11.5. Configuring Perspectives
11.5.1. The Configure Perspectives dialog
11.6. Context Sensitive Menu
11.6.1. Create Diagram
11.6.2. Create ModelElement
11.6.3. Copy Diagram to Clipboard as Image
11.6.4. Add to Diagram
11.6.5. Delete From Model
11.6.6. Set Source Path... (To be written)
11.6.7. Add Package
11.6.8. New Stereotype
11.6.9. Add All Classes in Namespace Alle Klassen im Namensraum hinzufügen
12. The Editing Pane
12.1. Introduction
12.2. Mouse Behavior in the Editing Pane
12.2.1. Button 1 Click
12.2.2. Button 1 Double Click
12.2.3. Button 1 Motion
12.2.4. Shift and Ctrl modifiers with Button 1
12.2.5. Alt with Button 1 motion
12.2.6. Button 2 Actions
12.2.7. Button 2 Double Click
12.2.8. Button 2 Motion
12.3. Keyboard Behavior in the Editing Pane
12.3.1. Nudging a model element
12.3.2. Moving across the model elements
12.4. The tool bar
12.4.1. Layout Tools
12.4.2. Annotation Tools
12.4.3. Drawing Tools
12.4.4. Use Case Diagram Specific Tools
12.4.5. Class Diagram Specific Tools
12.4.6. Sequence Diagram Specific Tools
12.4.7. Collaboration Diagram Specific Tools
12.4.8. Statechart Diagram Specific Tools
12.4.9. Activity Diagram Specific Tools
12.4.10. Deployment Diagram Specific Tools
12.5. The Broom
12.6. Selection Action Buttons
12.7. Clarifiers
12.8. The Drawing Grid
12.9. The Diagram Tab
12.10. Pop-Up Menus
12.10.1. Critiques
12.10.2. Ordering
12.10.3. Add
12.10.4. Show
12.10.5. Modifiers
12.10.6. Multiplicity
12.10.7. Aggregation
12.10.8. Navigability
12.11. Notation
12.11.1. Notation Languages
12.11.2. Notation Editing on the diagram
12.11.3. Notation Parsing
13. The Details Pane
13.1. Introduction
13.2. To Do Item Tab
13.2.1. Wizards
13.2.2. The Help Button
13.3. Properties Tab
13.4. Documentation Tab
13.5. Presentation Tab
13.6. Source Tab
13.7. Constraints Tab
13.7.1. The Constraint Editor
13.8. Stereotype Tab
13.9. Tagged Values Tab
13.10. Checklist Tab
14. The To-Do Pane
14.1. Introduction
14.2. Mouse Behavior in the To-Do Pane
14.2.1. Button 1 Click
14.2.2. Button 1 Double Click
14.2.3. Button 2 Actions
14.2.4. Button 2 Double Click
14.3. Presentation Selection
14.4. Item Count
15. The Critics
15.1. Introduction
15.1.1. Terminology
15.1.2. Design Issues
15.2. Uncategorized
15.3. Class Selection
15.3.1. Wrap DataType
15.3.2. Reduce Classes in namespace <namespace>
15.3.3. Clean Up Diagram
15.4. Naming
15.4.1. Resolve Association Name Conflict
15.4.2. Revise Attribute Names to Avoid Conflict
15.4.3. Change Names or Signatures in a model element
15.4.4. Duplicate End (Role) Names for an Association
15.4.5. Role name conflicts with member
15.4.6. Choose a Name (Classes and Interfaces)
15.4.7. Name conflict in a namespace
15.4.8. Choose a Unique Name for a model element (Classes and Interfaces)
15.4.9. Choose a Name (Attributes)
15.4.10. Choose a Name (Operations)
15.4.11. Choose a Name (States)
15.4.12. Choose a Unique Name for a (State related) model element
15.4.13. Revise Name to Avoid Confusion
15.4.14. Choose a Legal Name
15.4.15. Change a model element to a Non-Reserved Word
15.4.16. Choose a Better Operation Name
15.4.17. Choose a Better Attribute Name
15.4.18. Capitalize Class Name
15.4.19. Revise Package Name
15.5. Storage
15.5.1. Revise Attribute Names to Avoid Conflict
15.5.2. Add Instance Variables to a Class
15.5.3. Add a Constructor to a Class
15.5.4. Reduce Attributes on a Class
15.6. Planned Extensions
15.6.1. Operations in Interfaces must be public
15.6.2. Interfaces may only have operations
15.6.3. Remove Reference to Specific Subclass
15.7. State Machines
15.7.1. Reduce Transitions on <state>
15.7.2. Reduce States in machine <machine>
15.7.3. Add Transitions to <state>
15.7.4. Add Incoming Transitions to <model element>
15.7.5. Add Outgoing Transitions from <model element>
15.7.6. Remove Extra Initial States
15.7.7. Place an Initial State
15.7.8. Add Trigger or Guard to Transition
15.7.9. Change Join Transitions
15.7.10. Change Fork Transitions
15.7.11. Add Choice/Junction Transitions
15.7.12. Add Guard to Transition
15.7.13. Clean Up Diagram
15.7.14. Make Edge More Visible
15.7.15. Composite Association End with Multiplicity >1 Zusammengesetztes Assoziationsende mit der Kardinalität >1
15.8. Design Patterns
15.8.1. Consider using Singleton Pattern for <class>
15.8.2. Singleton Stereotype Violated in <class>
15.8.3. Nodes normally have no enclosers
15.8.4. NodeInstances normally have no enclosers
15.8.5. Components normally are inside nodes
15.8.6. ComponentInstances normally are inside nodes
15.8.7. Classes normally are inside components
15.8.8. Interfaces normally are inside components
15.8.9. Objects normally are inside components
15.8.10. LinkEnds have not the same locations
15.8.11. Set classifier (Deployment Diagram)
15.8.12. Missing return-actions
15.8.13. Missing call(send)-action
15.8.14. No Stimuli on these links
15.8.15. Set Classifier (Sequence Diagram)
15.8.16. Wrong position of these stimuli
15.9. Relationships
15.9.1. Circular Association
15.9.2. Make <association> Navigable
15.9.3. Remove Navigation from Interface via <association>
15.9.4. Add Associations to <model element>
15.9.5. Remove Reference to Specific Subclass
15.9.6. Reduce Associations on <model element>
15.9.7. Make Edge More Visible
15.10. Instantiation
15.11. Modularity
15.11.1. Classifier not in Namespace of its Association
15.11.2. Add Elements to Package <package>
15.12. Expected Usage
15.12.1. Clean Up Diagram
15.13. Methods
15.13.1. Change Names or Signatures in <model element>
15.13.2. Class Must be Abstract
15.13.3. Add Operations to <class>
15.13.4. Reduce Operations on <model element> Reduzieren Sie die Anzahl der Operationen im <Modellelement>
15.14. Code Generation
15.14.1. Change Multiple Inheritance to interfaces
15.15. Stereotypes
15.16. Inheritance
15.16.1. Revise Attribute Names to Avoid Conflict
15.16.2. Remove <class>'s Circular Inheritance
15.16.3. Class Must be Abstract
15.16.4. Remove final keyword or remove subclasses
15.16.5. Illegal Generalization
15.16.6. Remove Unneeded Realizes from <class>
15.16.7. Define Concrete (Sub)Class
15.16.8. Define Class to Implement <interface>
15.16.9. Change Multiple Inheritance to interfaces
15.16.10. Make Edge More Visible
15.17. Containment
15.17.1. Remove Circular Composition
15.17.2. Duplicate Parameter Name
15.17.3. Two Aggregate Ends (Roles) in Binary Association
15.17.4. Aggregate End (Role) in 3-way (or More) Association
15.17.5. Wrap DataType
3. Model Reference
16. Top Level Model Element Reference
16.1. Introduction
16.2. The Model
16.2.1. Model Details Tabs
16.2.2. Model Property Toolbar
16.2.3. Property Fields For The Model
16.3. Datatype
16.3.1. Datatype Details Tabs
16.3.2. Datatype Property Toolbar
16.3.3. Property Fields For Datatype
16.4. Enumeration
16.4.1. Enumeration Details Tabs
16.4.2. Enumeration Property Toolbar
16.4.3. Property Fields For Enumeration
16.5. Enumeration Literal
16.6. Stereotype
16.6.1. Stereotype Details Tabs
16.6.2. Stereotype Property Toolbar
16.6.3. Property Fields For Stereotype
16.7. Tag Definition
16.8. Diagram
16.8.1. Diagram Details Tabs
16.8.2. Diagram Property Toolbar
16.8.3. Property Fields For Diagram
17. Use Case Diagram Model Element Reference
17.1. Introduction
17.1.1. ArgoUML Limitations Concerning Use Case Diagrams
17.2. Actor
17.2.1. Actor Details Tabs Detail-Register Akteur
17.2.2. Actor Property Toolbar
17.2.3. Property Fields For Actor
17.3. Use Case
17.3.1. Use Case Details Tabs
17.3.2. Use Case Property Toolbar
17.3.3. Property Fields For Use Case
17.4. Extension Point
17.4.1. Extension Point Details Tabs
17.4.2. Extension Point Property Toolbar
17.4.3. Property Fields For Extension Point
17.5. Association
17.6. Association End
17.7. Dependency
17.8. Generalization
17.8.1. Generalization Details Tabs
17.8.2. Generalization Property Toolbar
17.8.3. Property Fields For Generalization
17.9. Extend
17.9.1. Extend Details Tabs
17.9.2. Extend Property Toolbar
17.9.3. Property Fields For Extend
17.10. Include
17.10.1. Include Details Tabs
17.10.2. Include Property Toolbar
17.10.3. Property Fields For Include
18. Class Diagram Model Element Reference
18.1. Introduction
18.1.1. Limitations Concerning Class Diagrams in ArgoUML
18.2. Package
18.2.1. Package Details Tabs
18.2.2. Package Property Toolbar
18.2.3. Property Fields For Package
18.3. Datatype
18.4. Enumeration
18.5. Stereotype
18.6. Class
18.6.1. Class Details Tabs
18.6.2. Class Property Toolbar
18.6.3. Property Fields For Class
18.7. Attribute
18.7.1. Attribute Details Tabs
18.7.2. Attribute Property Toolbar
18.7.3. Property Fields For Attribute
18.8. Operation
18.8.1. Operation Details Tabs
18.8.2. Operation Property Toolbar
18.8.3. Property Fields For Operation
18.9. Parameter
18.9.1. Parameter Details Tabs
18.9.2. Parameter Property Toolbar
18.9.3. Property Fields For Parameter
18.10. Signal
18.10.1. Signal Details Tabs
18.10.2. Signal Property Toolbar
18.10.3. Property Fields For Signal
18.11. Reception (to be written)
18.12. Association
18.12.1. Three-way and Greater Associations and Association Classes
18.12.2. Association Details Tabs
18.12.3. Association Property Toolbar
18.12.4. Property Fields For Association
18.13. Association End
18.13.1. Association End Details Tabs
18.13.2. Association End Property Toolbar
18.13.3. Property Fields For Association End
18.14. Dependency
18.14.1. Dependency Details Tabs
18.14.2. Dependency Property Toolbar
18.14.3. Property Fields For Dependency
18.15. Generalization
18.16. Interface
18.16.1. Interface Details Tabs
18.16.2. Interface Property Toolbar
18.16.3. Property Fields For Interface
18.17. Abstraction
18.17.1. Abstraction Details Tabs
18.17.2. Abstraction Property Toolbar
18.17.3. Property Fields For Abstraction
19. Sequence Diagram Model Element Reference
19.1. Introduction
19.1.1. Limitations Concerning Sequence Diagrams in ArgoUML
19.2. Object
19.2.1. Object Details Tabs
19.2.2. Object Property Toolbar
19.2.3. Property Fields For Object
19.3. Stimulus
19.3.1. Stimulus Details Tabs
19.3.2. Stimulus Property Toolbar
19.3.3. Property Fields For Stimulus
19.4. Stimulus Call
19.5. Stimulus Create
19.6. Stimulus Destroy
19.7. Stimulus Send
19.8. Stimulus Return
19.9. Link
19.9.1. Link Details Tabs
19.9.2. Link Property Toolbar
19.9.3. Property Fields For Link
20. Statechart Diagram Model Element Reference
20.1. Introduction
20.1.1. Limitations Concerning Statechart Diagrams in ArgoUML
20.2. State
20.2.1. State Details Tabs
20.2.2. State Property Toolbar
20.2.3. Property Fields For State
20.3. Action
20.3.1. Action Details Tabs
20.3.2. Action Property Toolbar
20.3.3. Property Fields For Action
20.4. Composite State
20.5. Concurrent Region
20.6. Submachine State
20.7. Stub State
20.8. Transition
20.8.1. Transition Details Tabs
20.8.2. Transition Property Toolbar
20.8.3. Property Fields For Transition
20.9. Event
20.9.1. Event Details Tabs
20.9.2. Event Property Toolbar
20.9.3. Property Fields For Event
20.10. Guard
20.10.1. Guard Details Tabs
20.10.2. Guard Property Toolbar
20.10.3. Property Fields For Guard
20.11. Pseudostate
20.11.1. Pseudostate Details Tabs
20.11.2. Pseudostate Property Toolbar
20.11.3. Property Fields For Pseudostate
20.12. Initial State
20.13. Final State
20.13.1. Final State Details Tabs
20.13.2. Final State Property Toolbar
20.13.3. Property Fields For Final State
20.14. Junction
20.15. Choice
20.16. Fork
20.17. Join
20.18. Shallow History
20.19. Deep History
20.20. Synch State
20.20.1. Synch State Details Tabs
20.20.2. Synch State Property Toolbar
20.20.3. Property Fields For Synch State
21. Collaboration Diagram Model Element Reference
21.1. Introduction
21.1.1. Limitations Concerning Collaboration Diagrams in ArgoUML
21.2. Classifier Role
21.2.1. Classifier Role Details Tabs
21.2.2. Classifier Role Property Toolbar
21.2.3. Property Fields For Classifier Role
21.3. Association Role
21.3.1. Association Role Details Tabs
21.3.2. Association Role Property Toolbar
21.3.3. Property Fields For Association Role
21.4. Association End Role
21.4.1. Association End Role Details Tabs
21.4.2. Association End Role Property Toolbar
21.4.3. Property Fields For Association End Role
21.5. Message
21.5.1. Message Details Tabs
21.5.2. Message Property Toolbar
21.5.3. Property Fields For Message
22. Activity Diagram Model Element Reference
22.1. Introduction
22.1.1. Limitations Concerning Activity Diagrams in ArgoUML
22.2. Action State
22.2.1. Action State Details Tabs
22.2.2. Action State Property ToolBar
22.2.3. Property fields for action state
22.3. Action
22.4. Transition
22.5. Guard
22.6. Initial State
22.7. Final State
22.8. Junction (Decision)
22.9. Fork
22.10. Join
22.11. ObjectFlowState
23. Deployment Diagram Model Element Reference
23.1. Introduction
23.1.1. Limitations Concerning Deployment Diagrams in ArgoUML
23.2. Node
23.2.1. Node Details Tabs
23.2.2. Node Property Toolbar
23.2.3. Property Fields For Node
23.3. Node Instance
23.3.1. Node Instance Details Tabs
23.3.2. Node Instance Property Toolbar
23.3.3. Property Fields For Node Instance
23.4. Component
23.4.1. Component Details Tabs
23.4.2. Component Property Toolbar
23.4.3. Property Fields For Component
23.5. Component Instance
23.5.1. Component Instance Details Tabs
23.5.2. Component Instance Property Toolbar
23.5.3. Property Fields For Component Instance
23.6. Dependency
23.7. Class
23.8. Interface
23.9. Association
23.10. Object
23.11. Link
24. Profiles
24.1. Introduction
24.2. The UML 1.4 Standard Elements Profile
24.2.1. Datatypes
24.2.2. Enumerations
24.2.3. Stereotypes
24.2.4. Tag Definitions
24.3. The Java Profile
24.3.1. Datatypes
24.3.2. Classes
24.3.3. Interfaces
Glossary
A. Supplementary Material for the Case Study
A.1. Introduction
A.2. Requirements Documents (To be written)
A.2.1. Vision Document (To be written)
A.2.2. Use Case Specifications (To be written)
A.2.3. Supplementary Requirements Specification (To be written)
B. UML resources
B.1. The UML specs (To be written)
B.2. UML related papers (To be written)
B.2.1. UML action specifications (To be written)
B.3. UML related websites (To be written)
C. UML Conforming CASE Tools
C.1. Other Open Source Projects (To be written)
C.2. Commercial Tools (To be written)
D. The C++ Module
D.1. Modeling for C++
D.1.1. Class tagged values
D.1.2. Attribute tagged values
D.1.3. Parameters
D.1.4. Generalization
D.1.5. Realization
D.1.6. Preserved sections
E. Limits and Shortcomings
E.1. Diagram Canvas Size
E.2. Missing functions
F. Open Publication License
F.1. Introduction
F.2. Copyright
F.3. Scope Of License
F.4. Requirements On Modified Works
F.5. Good-Practice Recommendations
G. The CRC Card Methodology
G.1. The Card
G.2. The Group
G.3. The Session
G.4. The Process
Index

Preface

Software design is a cognitively challenging task. Designers must manually enter designs, but the primary difficulty is decision-making rather than data-entry. If designers improved their decision-making capabilities, it would result in better designs.

Current CASE tools provide automation and graphical user interfaces that reduce the manual work of entering a design and transforming a design into code. They aid designers in decision-making mainly by providing visualization of design diagrams and simple syntactic checks. Also many CASE tools provide substantial benefits in the area of version control and concurrent design mechanisms. One area of design support that has been not been well supported is analysis of design decisions.

Current CASE tools are usable in that they provide a GUI that allows designers to access all the features provided by the tool. And they support the design process in that they allow the designer to enter diagrams in the style of popular design methodologies. But they typically do not provide process support to guide the designer through the design task. Instead, designers typically start with a blank page and must remember to cover every aspect of the design.

ArgoUML is a domain-oriented design environment that provides cognitive support of object-oriented design. ArgoUML provides some of the same automation features of a commercial CASE tool, but it focuses on features that support the cognitive needs of designers. These cognitive needs are described by three cognitive theories:

  1. reflection-in-action;

  2. opportunistic design; and

  3. comprehension and problem solving.

ArgoUML is based directly on the UML 1.4 specification. The core model repository is an implementation of the Java Metadata Interface (JMI) which directly supports MOF and uses the machine readable version of the UML 1.4 specification provided by the OMG.

Furthermore, it is our goal to provide comprehensive support for OCL (the Object Constraint Language) and XMI (the XML Model Interchange format).

ArgoUML was originally developed by a small group of people as a research project. ArgoUML has many features that make it special, but it does not implement all the features that commercial CASE tools provide.

The current version (0.28) of ArgoUML implements all the diagram types of the UML 1.4 standard (versions of ArgoUML prior to 0.20 implemented the UML 1.3 standard). It is written in Java and runs on every computer which provides a Java platform version 5 or newer. It uses the open file formats XMI (XML Metadata Interchange format) (for model information) and PGML (Precision Graphics Markup Language) (for graph information) for storage. When ArgoUML implements UML 2.0, PGML will be replaced by the UML Diagram Interchange specification.

This manual is the cumulative work of several people and has been evolving over several years. Connected to the release 0.10 of ArgoUML, Jeremy Bennett, wrote a lot of the new material that was added to the earlier versions by Alejandro Ramirez, Philippe Vanpeperstraete and Andreas Rueckert. He also added things from some of the other documents namely the developers cookbook by Markus Klink and Linus Tolke, the Quick Guide by Kunle Odutola, and the FAQ by Dennis Daniels. Connected to the release 0.14 changes were made by Linus Tolke, and by Michiel van der Wulp. These changes were mostly to adopt the manual to the new functions and appearance of ArgoUML version 0.14, and introduction of the index. The users and developers that have contributed by providing valuable input, such as review comments or observations while reading and using this manual are too many to name.

ArgoUML is available for free and can be used in commercial settings. For terms of use, see the license agreement presented when you download ArgoUML. We are providing the source code for ArgoUML for you to review, customize to your needs, and improve. Over time, we hope that ArgoUML will evolve into a powerful and useful tool for everyone to use.

This User Manual is aimed at the working designer, who wishes to make use of ArgoUML. The manual is presently written assuming familiarity with UML, but eventually it will support those new to UML.

The manual is written in DocBook/XML and available as both HTML and PDF.

The ArgoUML project welcomes those who want to get more involved. Look at the project website to find out more.

Tell us what you think about this User Manual! Your comments will help us improve things. See Section 1.3.3, “ User Feedback .

Chapter 1.  Introduction

1.1.  Origins and Overview of ArgoUML

1.1.1.  Object Oriented Analysis and Design

Over the past decade, Object Oriented Analysis and Design (OOA&D) has become the dominant software development paradigm. With it has come a major shift in the thought processes of all involved in the software development life cycle.

Programming language support for objects began with Simula 67, but it was the emergence in the 1980's of hybrid languages, such as C++, Ada and Object Pascal that allowed OOA&D to take off. These languages provided support for both OO and procedural programming. Object Oriented programming became mainstream.

An OO system is designed and implemented as a simulation of the real world using software artifacts. This premise is as powerful as it is simple. By using an OO approach to design a system can be designed and tested (or more correctly simulated) without having to actually build the system first.

It is the development during the 1990's of tools to support Object Oriented analysis and design that moved this approach into the mainstream. When coupled with the ability to design systems at a very high level, a tool based OOA&D approach has enabled the implementation of more complex systems than previously possible.

The final driver that has propelled OOA&D has been its suitability for modeling graphical user interfaces. The popularity of object based and object oriented graphical languages such as Visual Basic and Java reflect the effectiveness of this approach.

1.1.2.  The Development of ArgoUML

During the 1980's a number of OOA&D process methodologies and notations were developed by different research teams. It became clear there were many common themes and, during the 1990's, a unified approach for OOA&D notation was developed under the auspices of the Object Management Group. This standard became known as the Unified Modeling Language (UML), and is now the standard language for communicating OO concepts.

ArgoUML was conceived as a tool and environment for use in the analysis and design of object-oriented software systems. In this sense it is similar to many of the commercial CASE tools that are sold as tools for modeling software systems. ArgoUML has a number of very important distinctions from many of these tools.

  1. It is free.

  2. ArgoUML draws on research in cognitive psychology to provide novel features that increase productivity by supporting the cognitive needs of object-oriented software designers and architects.

  3. ArgoUML supports open standards extensively - UML, XMI, SVG, OCL and others.

  4. ArgoUML is a 100% pure Java application. This allows ArgoUML to run on all platforms for which a reliable port of the Java platform is available.

  5. ArgoUML is an open source project. The availability of the source ensures that a new generation of software designers and researchers now have a proven framework from which they can drive the development and evolution of CASE tool technologies.

UML is the most prevalent OO modeling language and Java is one of the most productive OO development platforms. Jason Robbins and the rest of his research team at the University of California, Irvine leveraged these benefits in creating ArgoUML. The result is a solid development tool and environment for OO systems design. Further, it provides a test bed for the evolution of object oriented CASE tools development and research.

A first release of ArgoUML was available in 1998 and more than 100,000 downloads by mid-2001 show the impact that this project has made, being popular in educational and commercial fields.

1.1.3.  Finding Out More About the ArgoUML Project

1.1.3.1.  How ArgoUML is Developed

Jason Elliot Robbins founded the Argo Project and provided early project leadership. While Jason remains active in the project, he has handed off project leadership. The project continues to move forward strongly. There are more than 300 members on the developer mailing list (see http://argouml.tigris.org/servlets/ProjectMailingListList), with a couple of dozen of those forming the core development group.

The developer mailing list is the place where all the discussion on the latest tasks takes place, and developers discuss the directions the project should take. Although controversial at times, these discussions are always kept nice and friendly (no flame-wars and such), so newbies should not hesitate and participate in them. You'll always get a warm welcome there.

If you want to learn how the project is run and how to contribute to it, go the the ArgoUML Web Site Developer Zone and read through the documentation there. The Developers' Cookbook was written specifically for this purpose.

1.1.3.2.  More on Infrastructure

Besides the developer mailing list, there's also a mailing for users (see The ArgoUML Mailing List ), where we can discuss problems from a user perspective. Developers also read this list, so highly qualified help will generally be provided.

Before posting to this list, you should take a look at the user FAQ maintained by Ewan R. Grantham.

More information on ArgoUML and other UML related topics is also available on the ArgoUML website, maintained by Linus Tolke.

1.2.  Scope of This User Manual

1.2.1.  Target Audience

The current release of this document is aimed at experienced users of UML in OOA&D (perhaps with other tools) who wish to transfer to ArgoUML.

Future releases will support designers who know OOA&D, and wish to adopt UML notation within their development process.

A long term goal is to support i) those who are learning design and wish to start with an OOA&D process that uses UML notation, and ii) people interested in modularized code design with a GUI.

1.2.2.  Scope

The intention is that this document will provide a comprehensive guide, enabling designers to use ArgoUML to its full extent. It is in two parts.

  • A tutorial manual, showing how to work with ArgoUML

  • A complete reference manual, recording everything you can do with ArgoUML.

Version 0.22 of the document achieved the second of these.

In this guide there are some things you will not find, because they are covered elsewhere.

  • Descriptions of how ArgoUML works on the inside.

  • How to improve ArgoUML with new features and functions.

  • A trouble shooting guide.

  • A summary quick reference to using ArgoUML.

These are covered in the Developers Cookbook, the FAQ, and the Quick Guide.

1.3.  Overview of the User Manual

1.3.1.  Tutorial Manual Structure

Chapter 2, Introduction (being written) provides an overview of UML based OOA&D, including a guide to getting ArgoUML up and running.

Chapter 4, Requirements Capture through Chapter 7, Code Generation, Reverse Engineering, and Round Trip Engineering then step through each part of the design process from initial requirements capture through to final project build and deployment.

As each UML concept is encountered, its use is explained. Its use within ArgoUML is then described. Finally a case study is used to give examples of the concepts in use.

1.3.2.  Reference Manual Structure

Chapter 8, Introduction is an overview of the user interface and provides a summary of the support for the various UML diagram types in ArgoUML. Chapter 10, The Menu bar and Chapter 11, The Explorer describe the menu bar, and each of the sub-windows of the user interface, known as Panes.

Chapter 15, The Critics gives details of all the cognitive critics within the system. Eventually ArgoUML will link directly to this manual when giving advice on critics.

Chapter 16, Top Level Model Element Reference is an overview of the model elements (i.e. the UML entities that can be placed on diagrams) within ArgoUML. The following chapters ( Chapter 17, Use Case Diagram Model Element Reference through Chapter 24, Profiles ) describe, the model elements that can be created through each ArgoUML diagram, and their properties, as well as some standard model elements provided with the system.

A complete Glossary is provided. Appendix A, Supplementary Material for the Case Study provides material to supplement the case study used throughout the document. Appendix B, UML resources and Appendix C, UML Conforming CASE Tools identify background information on UML and UML CASE tools. Appendix D, The C++ Module explains the use of the C++ module. Appendix F, Open Publication License is a copy of the GNU Free Documentation License.

1.3.3.  User Feedback

Please tell us what you think about this User Manual. Your comments will help us make improvements. Email your thoughts to the ArgoUML Users Mailing List. In case you would like to add to the missing chapters you should contact the ArgoUML Developer Mailing List to check whether anyone else is working on this part. You can subscribe to either of the mailing lists via the ArgoUML web site.

1.4.  Assumptions

This release of the manual assumes the reader is very familiar with UML already. This is reflected in the sparseness of the description of UML concepts in the tutorial.

The case study is described, but not yet fully realized throughout the tutorial. This will be achieved in future releases of the manual.

Part 1.  Tutorial

Chapter 2.  Introduction (being written)

This tutorial will be taking you through a tour of the use of ArgoUML to model a system.

First you will become familiar with the feel of the product and then we will go through an analysis and development process for a test case. Not every nook and cranny of the product will be demonstrated. That degree of detail is given in the reference materials to be found in subsequent parts of this document.

The state of the model at the end of key sections will be available in .zargo files. These are available so that you can play with various aspects not specifically covered in this tutorial and then restore yourself back to the proper state of the model in your work area. These .zargo files will be identified at the end of the sections whose work they represent.

An ATM (automated teller machine) project has been chosen as a case study to demonstrate the various aspects of modeling that ArgoUML offers. In subsequent sections we are going to develop the ATM example into a complete description in UML. The tutorial, however, will only walk you through part of it.

At this point you should create a directory to contain your project. Name the directory anything you feel is consistent with the rest of your file system. You should name the contents and any subdirectories as directed for reasons that will become apparent.

The case study will be an ATM system. Your company is FlyByNight Industries. You are going to play two roles. That of the Project Manager and that of the Designer Analyst.

We are not going to build a physical ATM, of course. The product that we will build as a case study will be an ATM simulator to be used for testing the design of a physical ATM.

How your company arranges its work into projects is usually determined as much by politics as anything else and is, therefore, out of the scope of this document. We will go into how you structure the project itself once one has been defined.

Chapter 3.  UML Based OOA&D

In this chapter, we look at how UML as a notation is used within OOA&D.

3.1.  Background to UML

Object orientation as a concept has been around since the 1960's, and as a design concept since 1972. However it was in the 1980's that it started to develop as a credible alternative to a functional approach in analysis and design. We can identify a number of drivers.

  1. The emergence of mainstream OO programming languages like SmallTalk and particularly C++. C++ was a pragmatic OO language derived from C, widely used because of its association with Unix.

  2. The development of powerful workstations, and with them the emergence into the mainstream of windowing operating user environments. Graphical User Interfaces (GUI) have an inherent object structure.

  3. A number of very public major project failures, suggesting that current approaches were not satisfactory.

A number of researchers proposed OOA&D processes, and with them notations. Those that achieved some success include Coad-Yourdon, Booch, Rumbaugh OMT, OOSE/Jacobson, Shlaer-Mellor, ROOM (for real-time design) and the hybrid Jackson Structured Development.

During the early 1990's it became clear that these approaches had many good ideas, often very similar. A major stumbling block was the diversity of notation, meaning engineers tended to be familiar with one OOA&D methodology, rather than the approach in general.

UML was conceived as a common notation, that would be in the interests of all involved. The original standard was driven by Rational Software ( www.rational.com, in which three of the key researchers in the field (Booch, Jacobson and Rumbaugh were involved). They produced documents describing UML v0.9 and v0.91 during 1996. The effort was taken industry wide through the Object Management Group (OMG), already well known for the CORBA standard. A first proposal, 1.0 was published in early 1997, with an improved version 1.1 approved that autumn.

ArgoUML is based on UML v1.4, which was adopted by OMG in September 2001. The current official version, supported by ArgoUML, is UML v1.4.2 dated July 2004, adopted as ISO/IEC 19501. The latest UML version is UML v2.1.2, from November 2007.

3.2.  UML Based Processes for OOA&D

It is important to understand that UML is a notation for OOA&D. It does not prescribe any particular process. Whatever process is adopted, it must take the system being constructed through a number of phases.

  1. Requirements Capture. This is where we identify the requirements for the system, using the language of the problem domain. In other words we describe the problem in the “customer's” terms.

  2. Analysis. We take the requirements and start to recast them in the language of a putative solution—the solution domain. At this stage, although thinking in terms of a solution, we ensure we keep things at a high level, away from concrete details of a specific solution—what is known as abstraction.

  3. Design. We take the specification from the Analysis phase and construct the solution in full detail. We are moving from abstraction of the problem to its realization in concrete terms.

  4. Build Phase. We take the actual design and write it in a real programming language. This includes not just the programming, but the testing that the program meets the requirements (verification), testing that the program actually solves the customer's problem (validation) and writing all user documentation.

3.2.1.  Types of Process

In this section we look at the two main types of process in use for software engineering. There are others, but they are less widely used.

In recent years there has also been a move to reduce the effort required in developing software. This has led to the development of a number of lightweight variants of processes (often known as agile computing or extreme programming) that are suited to small teams of engineers.

3.2.1.1.  The Waterfall Process

In this process, each stage of the process—requirements, analysis, design and build (code and test) is completed before the next one starts. This is illustrated in Figure 3.1, “ The Waterfall Process.

Figure 3.1.  The Waterfall Process

The Waterfall Process

This is a very satisfactory process where requirements are well designed and not expected to change, for example automating a well proven manual system.

The weaknesses of this approach show with less well defined problems. Invariably some of the uncertainties in the requirements will not be clarified until well into the analysis and design, or even code phases, requiring backtracking to redo work.

The worst aspect of this, is that working code does not become available until near the end of the project, and very often it is only at this stage that problems with the original requirements (for example with the user interface) become apparent.

This is exacerbated, by each successive stage requiring more effort, than the previous, so that the costs of late problem discovery are hugely expensive. This is illustrated by the pyramid in Figure 3.2, “ Effort Involved in the Steps of the Waterfall Process .

Figure 3.2.  Effort Involved in the Steps of the Waterfall Process

Effort Involved in the Steps of the Waterfall Process

The waterfall process is still probably the dominant design process. However because of its limitations it is increasingly replaced by iterative processes, particularly for projects where the requirements are not well defined.

3.2.1.2.  Iterative Development Processes

In recent years a new approach has been used, which aims to get at least part of the code up and running as quickly as possible, to bring discovery of problems forward in the development cycle.

These processes use a series of “mini-waterfalls”, defining a few requirements (the most important) first, taking them through analysis, design and build to get an early version of the product, with limited functionality, related to the most important requirements. Feedback from this can then be used to refine the requirements, spot problems etc before more work is done.

The process is then repeated for further requirements to construct a product with a step up in functionality. Again further feedback can be applied to the requirements.

The process is repeated, until eventually all requirements have been implemented and the product is complete. It is this iteration that gives these processes their name. Figure 3.3, “ Effort Involved in the Steps of an Iterative Process shows how this process compares to the pyramid structure of the Waterfall Process.

Figure 3.3.  Effort Involved in the Steps of an Iterative Process

Effort Involved in the Steps of an Iterative Process

The growth in popularity of iterative processes is closely tied to the growth of OOA&D. It is the clean encapsulation of objects that allows a part of a system to be built with stubs for the remaining code clearly defined.

3.2.1.2.1.  The Rational Unified Process

Perhaps the best known Iterative Process is the Rational Unified Process (RUP) from Rational Software ( www.rational.com).

This process recognizes that our pyramid view of even slices of the waterfall is not realistic. In practice the early iterations tend to be heavy on the requirements end of things (you need to define a reasonable amount even to get started), while the later iterations have more of their effort in the design and build areas.

RUP recognizes that iterations can be grouped into a number of phases according to their stage in the overall project. Each phase may have one or more iterations.

  • In the inception phase iterations tend to be heavy on the requirements/analysis end, while any build activity may be limited to emulation of the design within a CASE tool.

  • In the elaboration phase iterations tend to be completing the specification of the requirements, and starting to focus on the analysis and design, and possibly the first real built code.

  • In the construction phase iterations the requirements and analysis are more or less completed, and the effort is mostly in design and build.

  • Finally, in the deployment phase iterations are largely about build activity, and in particular the testing of the software.

[Note]Note

It should be clear that testing is an integral part of all phases. Even in the early phases the requirements and design should be tested, and this is facilitated by a good CASE tool.

We shall use an iterative process in this manual, that is loosely based on the RUP.

3.2.1.2.2.  Iteration Size

A good rule of thumb is that an iteration should take between six and ten weeks for typical commercial projects. Any longer and you have probably bitten off too many requirements to do in one go. You also lose focus on getting the next working iteration completed. Any shorter and you probably haven't got enough requirements to make a significant advance. In this case the additional overhead associated with an interation will become a problem.

The total number of iterations depends on the size of project. Take the estimated time (working out/guessing that is a whole subject on its own), and divide it into 8 week chunks. Experience seems to suggest that the iterations will divide in the ratio of around 1:2:3:3 into RUP style inception, elaboration, construction and deployment phases. A project that has great vagueness in its specification (some advanced research projects for example) will tend to be heavier on the early phases.

When building a product to contract for a customer the end point is well defined. However when developing a new product for the market place, a strategy that can be used is to decide the product launch date, and hence the end date for completion of engineering (some time before). The time is then divided into iterations, and as much of the product as can be built in that time developed. The iterative process is very effective where time to market is more important than the exact functionality.

3.2.1.3.  Recursive Development Processes

Very few software systems are conceived as monolithic artifacts. They are broken down into subsystems, modules etc.

Software processes are the same, with early parts of the process defining a top level structure, and the process reapplying to parts of the structure in turn to define ever greater details.

For example the initial design of a telephone system might identify objects to

  1. Handle the phone lines,

  2. Process the calls,

  3. Manage the system and,

  4. Bill the customer.

The software process can then be reapplied to each of these four components to identify their design.

OOA&D with its clean boundaries to objects, naturally supports this approach. Such OOA&D with recursive development is sometimes abbreviated as OOA&D/RD.

Recursive development can be applied equally well to waterfall or iterative processes. It is not an alternative to them.

3.2.2.  A Development Process for This Tutorial

For the purpose of this tutorial we will use a stripped down iterative process with recursive development, loosely akin to RUP. The case study will take us through the first iteration, although at the end of the tutorial section of the manual we will look at how the project will develop to completion.

Within that first iteration, we will tackle each of the requirements capture, analysis, design and build activities in turn. Not all parts of the process are based on UML or ArgoUML. We will look at what other material is needed outside.

Within this process we will have an opportunity to see the various UML diagrams in use. The full range of UML diagrams and how they are supported is described in the reference manual (see Section 16.8, “ Diagram ).

3.2.2.1.  Requirements Capture

Our requirements capture will use the UML concept of Use Cases. Starting with a Vision Document we will see how Use Cases can be developed to describe all aspects of the system's behavior in the problem domain.

3.2.2.2.  Analysis

During the analysis stage, we will introduce the UML concept of classes to allow us to build a top level view of the objects that will make up the solution—sometimes known as a concept diagram.

We will introduce the UML sequence diagram and statechart diagram to capture requirements for the overall behavior of the system.

Finally we will take the Use Cases from the requirements capture stage, and recast them in the language of the solution domain. This will illustrate the UML ideas of stereotyping and realization.

3.2.2.3.  Design

We use the UML package diagram to organize the components of the project. We then revisit the class diagram, sequence diagram and statechart diagram, to show how they can be used recursively to design the complete solution.

During this part of the process, we need to develop our system architecture, to define how all the components will fit together and operate.

Although not strictly part of our process, we'll look at how the UML collaboration diagram can be used as an alternative to, or to complement the sequence diagram. Similarly we will look at the UML activity diagram as an alternative or complement to the statechart diagram.

Finally we shall use the UML deployment diagram to specify how the system will actually be realized.

3.2.2.4.  Build

UML is not really concerned with code writing. However at this stage we will show how ArgoUML can be used for code generation.

We will also look at how the UML Use Case Diagram and Use Case Specification are invaluable tools for a test program.

3.3.  Why ArgoUML is Different

In the introduction, we listed the four key things that make ArgoUML different: i) it makes use of ideas from cognitive psychology, ii) it is based on open standards; iii) it is 100% pure Java; and iv) it is an open source project.

3.3.1.  Cognitive Psychology

3.3.1.1.  Theory

ArgoUML is particularly inspired by three theories within cognitive psychology: i) reflection-in-action, ii) opportunistic design iii) and comprehension and problem solving.

  • Reflection-in-Action

    This theory observes that designers of complex systems do not conceive a design fully-formed. Instead, they must construct a partial design, evaluate, reflect on, and revise it, until they are ready to extend it further.

    As developers work hands-on with the design, their mental model of the problem situation improves, hence improving their design.

  • Opportunistic Design

    A theory within cognitive psychology suggesting that although designers plan and describe their work in an ordered, hierarchical fashion, in reality, they choose successive tasks based on the criteria of cognitive cost.

    Simply stated, designers do not follow even their own plans in order, but choose steps that are mentally least expensive among alternatives.

  • Comprehension and Problem Solving

    A design visualization theory within cognitive psychology. The theory notes that designers must bridge a gap between their mental model of the problem or situation and the formal model of a solution or system.

    This theory suggests that programmers will benefit from:

    1. Multiple representations such as program syntactic decomposition, state transitions, control flow, and data flow. These allow the programmer to better identify elements and relationships in the problem and solution and thus more readily create a mapping between their situation models and working system models.

    2. Familiar aspects of a situation model, which improve designers' abilities to formulate solutions.

3.3.1.2.  Practical Application in ArgoUML

ArgoUML implements these theories using a number of techniques.

  1. The design of a user interface which allows the user to view the design from a number of different perspectives, and allows the user to achieve goals through a number of alternative routes.

  2. The the use of processes running in parallel with the design tool, evaluating the current design against models of how “best practice” design might work. These processes are known as design critics.

  3. The use of to-do lists to convey suggestions from the design critics to the user, as well as allowing the user to record areas for future action.

  4. The use of checklists, to guide the user through a complex process.

3.3.2.  Open Standards

UML is itself an open standard. ArgoUML throughout has tried to use open standards for all its interfaces.

The key advantage of adherence to open standards is that it permits easy inter-working between applications, and the ability to move from one application to another as necessary.

3.3.2.1.  XML Metadata Interchange (XMI)

XML Metadata Interchange (XMI) is the standard for saving the meta-data that make up a particular UML model. In principle this will allow you to take the model you have created in ArgoUML and import it into another tool.

This clearly has advantages in allowing UML to meet its goal of being a standard for communication between designers.

The reality is not quite this good. Prior to UML 2.0 the XMI file includes no information about the graphical representation of the models, so diagram layout is lost. ArgoUML gets round this by saving graphical information separate from the model (see Section 3.4.3.1, “ Loading and Saving).

3.3.2.2.  Graphics Formats - EPS, GIF, PGML, PNG, PS, SVG

  • Encapsulated PostScript (EPS) file is a PostScript file which satisfies additional restrictions. These restrictions are intended to make it easier for software to embed an EPS file within another PostScript document.

  • Graphics Interchange Format (GIF) uses lossless compression, and preserves sharp edges well, which makes it well-suited for ArgoUML. The GIF format used to be a patent encumbered format, but all the patents have currently expired.

  • Precision Graphics Markup Language (PGML) is an XML-based language for representing vector graphics. It was a W3C draft, but was not adopted as a recommendation. PGML and VML, another XML-based vector graphics language, were later joined and improved upon to create SVG (see below).

  • Portable Network Graphics (PNG) is an ISO/IEC standard (15948:2004) and is also a W3C recommendation. PNG is a bitmap image format that employs lossless data compression. PNG was created to both improve upon and replace the GIF format with an image file format that does not require a patent license to use. PNG is officially pronounced "ping" but it is often just spelled out — probably to avoid confusion with the network tool ping. PNG is supported by the libpng reference library, a platform-independent library that contains C functions for handling PNG images.

  • PostScript (PS) is a page description language and programming language used primarily in the electronic and desktop publishing areas.

  • Scalable Vector Graphics (SVG) is an XML markup language for describing two-dimensional vector graphics, both static and animated, and either declarative or scripted. It is an open standard created by the World Wide Web Consortium. The use of SVG on the web is in its infancy. There is a great deal of inertia due to the long-time use of pure raster formats and other formats like Macromedia Flash or Java applets, but also browser support is still uneven, with native support in Opera, Safari and Firefox, but Internet Explorer requires a plugin. See PGML above.

3.3.2.3.  Object Constraint Language (OCL)

Object Constraint Language (OCL) is a declarative language for describing rules that apply to UML models. It was developed at IBM and is now part of the UML standard. Initially OCL was only a formal specification language extension to UML. OCL may now be used with any Meta-Object Facility (MOF) compliant metamodel, including UML. The Object Constraint Language is a precise text language that provides constraint and object query expressions on any MOF model or metamodel that cannot otherwise be expressed by diagrammatic notation.

3.3.3.  100% Pure Java

Java was conceived as an interpreted language. It doesn't have a compiler to produce code for any particular target machine. It compiles code for its own target, the Java Virtual Machine (JVM).

Writing an interpreter for a JVM is much easier than writing a compiler, and such machines are now incorporated into almost every Web Browser. As a result most machines can run Java, with no further work.

(In case you wonder why all languages aren't like this, it is because interpreted languages tend to be slower than compiled languages. However with the high performance of modern PCs, the trade-off for portability is worthwhile for many applications. Furthermore modern multi-level caches can mean that interpreted languages, which produce denser code, may actually not be that much slower anyway.)

By choosing to write ArgoUML in pure Java, it is immediately made available to the maximum number of users with the minimum amount of effort.

3.3.4.  Open Source

ArgoUML is an open source project. That means anyone can have a free copy of the source code, change it, use it for new purposes and so on. The only (major) obligation is that you pass your code on in the same way to others. The precise nature of what you can and can't do varies from project to project, but the principle is the same.

The advantage is that a small project like ArgoUML suddenly is open to a lot of additional help from those who can chip in their ideas for how the program might be improved. At any one time there may be 10, 15, 20 or more people making significant contributions to ArgoUML. To do that commercially would cost $1m+ per year.

Its not just a spirit of pure altruism. Contributing is a way of learning “hands-on” about leading edge software. Its a way of getting a lot of visibility (over 1,125,000 people had downloaded ArgoUML by the end of 2005). That's a lot of good experience on a résumé and a lot of potential employers seeing you!

And its great for the ego!

Open Source doesn't preclude making money. Gentleware www.gentleware.com sell a commercial version of ArgoUML, Poseidon. Their value proposition is not a piece of private code. Its the commercial polish and support that take risk out of using ArgoUML in a commercial development, allowing customers to take advantage of ArgoUML's leading edge technology.

3.4.  ArgoUML Basics

The aim of this section is to get you started with ArgoUML. It takes you through obtaining the code and getting it running.

3.4.1.  Getting Started

3.4.1.1.  System Requirements

Since ArgoUML is written in 100% pure Java, it should run on any machine with Java installed. Java version 5 or later is needed. You may have this in place, but if not the latest version can be downloaded free from www.java.com. Note that you only need the Java Runtime Environment (JRE), there is no need to download the whole Java Development Kit (JDK).

ArgoUML needs a reasonable amount of computing resource. Any PC which is able to run an operating system with a graphical user interface will suffice. Download the code from Download section of the project website argouml.tigris.org. Choose the version that suits your needs as described in the section below.

3.4.1.2.  Downloading Options

You have three options for obtaining ArgoUML.

  1. Run ArgoUML directly from the Web Site using Java Web Start. This is the easiest option.

  2. Download the Windows installer program. This is the right option if you are on Windows and intend using ArgoUML regularly.

  3. Download the binary executable code. Unless you are on Windows, this is the right option if you intend using ArgoUML regularly and is not that difficult.

  4. Download the source code using Subversion and build your own version. Choose this option if you want to look at the internal workings of ArgoUML, or want to join in as a developer. This option does require the whole JDK (see Section 3.4.1.1, “ System Requirements).

All four options are freely available through the project web site, argouml.tigris.org.

3.4.1.3.  ArgoUML Using Java Web Start ArgoUML mit Java Web Start aufrufen

There are two steps to this.

  1. Install Java Web Start on your machine. This is available from java.sun.com/products/javawebstart, or via the Java Web Start link on the ArgoUML home page.

  2. Click on the Launch latest stable release link on the ArgoUML home page.

Java Web Start will download ArgoUML, cache it and start it the first time, then on subsequent starts, check if ArgoUML is updated and only download any updated parts and then start it. The ArgoUML home page also provides details on starting ArgoUML from the Java Web Start console.

3.4.1.4.  Using the Windows installer

If you choose to download and install using the Windows installer, you will have a choice of downloading the latest stable version of the code (which will be more reliable, but not have all the latest features), or the current version (which will be less reliable, but have more features). Choose according to your own situation.

The install wizard gives you the opportunity to install the latest "JRE" (Java Runtime Environment). There is no need to select this if you already have a Sun Java installed, version 5 or better.

3.4.1.5.  Downloading the Binary Executable

If you choose to download the binary executable, you will have a choice of downloading the latest stable version of the code (which will be more reliable, but not have all the latest features), or the current version (which will be less reliable, but have more features). Choose according to your own situation.

ArgoUML comes in .zip or tar.gz flavors. Choose the former if you are a Microsoft Windows user, and the latter if you are running some flavor of Unix. There is also a Mac OS X version with .app.tgz extension. Unpacking is as follows.

  • On Windows. Unzip the .zip file with WinZip, or if your version of Windows supports it, copy the files out of the compressed folder and put them into a directory of your choosing.

  • On Unix. Use GNU tar to unzip and break out the files to a directory of your choice tar zxvf <file>.tar.gz. If you have an older version of tar, the z option may not be available, so use gunzip < file.tar.gz | tar xvf -.

You should have a directory containing a number of .jar files and a README.txt.

3.4.1.6.  Problems Downloading and Installing

If you get completely stuck and you have no local assistance, try the web site, particularly the FAQ. If this still doesn't solve the problem, try the ArgoUML users' mailing list.

You can subscribe through the mailing lists section of the project web site argouml.tigris.org, or send an empty message to users@argouml.org with the subject line subscribe.

You can then send your problem to users@argouml.org and see how other users are able to help.

The users' mailing list is an excellent introduction to the live activity of the project. If you want to get further involved there are additional mailing lists that cover the development of the product and issues in the current and future releases.

3.4.1.7.  Running ArgoUML

To run ArgoUML depends on whether you use Microsoft Windows or some flavor of Unix.

  • On Windows. If you used the installer, starting ArgoUML is a matter of clicking on its icon. In case you installed the binairy executable, read on: Start an MSDOS shell window by e.g. using Start/Run with “command” in the text window. In the window change to the directory holding your ArgoUML files and type java -jar argouml.jar. This method has the advantage that progress and debugging information is visible in the DOS window. Alternatively use the supplied batch file (.bat), or create one containing the above command, with a shortcut to it on the desktop. The batch file should end with a "pause" statement in case any debugging information is created during a run. On some systems, simply (double) clicking on the argouml.jar file works. On others doing so initiates a zip utility. Refer to your operating system instructions or help facility to determine how to configure this.

  • On Unix. Start a shell window and type java -jar argouml.jar

3.4.1.8.  Problems Running ArgoUML

It's unusual to encounter problems if you have made a successful download. If you can't solve the problem. Try the users' mailing list (see Section 3.4.1.6, “ Problems Downloading and Installing).

  • Wrong JRE. The most common issue is not having a new enough Java Runtime Environment (it must be version 5 or later).

  • Wrong language. If the product came up in a language you can't read or just don't want, go to the second leftmost menu item in the menu bar at the top of the screen. Select the bottom most menu entry in the drop down. Figure 3.5, “ Setting Language in the Appearance Pane shows this in Russian. Then click on the second tab from the bottom in the column of tabs on the left. Drop down the list as shown in Figure 3.5, “ Setting Language in the Appearance Pane and select a language. Note that the languages are listed in themselves. The language shown as being selected is German in which the word for “German” is “Deutsch”. You will have to exit ArgoUML and restart it for the change to take effect. Use the X button at the upper right.

Figure 3.4.  Finding the Settings Wizard

Finding the Settings Wizard

Figure 3.5.  Setting Language in the Appearance Pane

Setting Language in the Appearance Pane

3.4.2.  The ArgoUML User Interface

Before beginning the Case Study, you need to become familiar with the user interface. Start by reading the introduction to the User Interface Reference. See Chapter 8, Introduction . You should also read the Section 8.2, “ General Mouse Behavior in ArgoUML

As you go through this tutorial you will be told what to do, and when to do it but how to do it will often be left to the User Interface Reference. It is not necessary at this point to read all of the Reference, but you should leaf through enough of it to become familiar with how to find things in it. Every attempt will be made to direct you to the appropriate part of the Reference at those points in the tutorial where they apply.

Figure 3.6, “ Initial ArgoUML window, shows the main ArgoUML window as it appears when ArgoUML is first entered.

Figure 3.6.  Initial ArgoUML window

Initial ArgoUML window

Grab the vertical divider bars and move them back and forth. Grab the horizontal divider bar and move it up and down. Play around a little with the little arrows at the left or top of the divider bars. See Section 8.3, “ General Information About Panes.

3.4.2.1.  The Menu Bar and Toolbars

The menu bar and toolbars give access to all the main features of ArgoUML. As is conventional, menu options and toolbar options that are not available (disabled) are grayed out and menu items that invoke a dialog box are followed by an ellipsis (...). At this time you should read Chapter 9, The Toolbar and Chapter 10, The Menu bar .

File menu. The standard file menu entries present no surprises and we will just use them when needed without first showing how they work. A number of other actions are available that are peculiar to ArgoUML and we will go over them here.

  1. File=>Revert to Saved. This has the same effect as File=>Open Project selecting the current project.

  2. Export/Import. Select the project line at the top of the Explorer. It should say "untitledModel" unless you have changed it. Perform a File=>Export XMI action using "DeleteThis" for an output name in the file chooser dialog. Select the "Properties" tab in the "Details Pane" and change the name to something else, anything will do. Perform a File=>Import XMI action. It will ask you whether you want to save the changes you have just made. Click on "No" and then in the file choosed that comes up select the "DeleteThis.xmi" file that you just wrote out. Observe that the name of the model has reverted back to what you had saved.

  3. File=>Import Sources. We will cover this later. You can't test it now unless you have some Java source code of your own handy.

  4. File=>Export (All) Graphics. In the Explorer Pane select one of the diagrams. Either "Class Diagram 1" or "Use Case Diagram 1" (assuming you haven't renamed or deleted them). Perform a File=>Export Graphics action. When the file chooser opens it defaults to the last name you saved something to (even from a project no longer open). The file chooser allows you to select from a number of formats. Drop down the "Files of type" combobox and observe the choices. Cancel out as there is nothing useful to save. Perform a File=>Export All Graphics action. Notice that this time you can't specify a file name and you can't select a file format. ArgoUML will allow you only to select an output directory. It will then create a file for each of your diagrams using the diagram name for the file name and an extension determined by the default graphics format. Actually, although you can't select file names in the browser panel, you can type one into the edit box. But, if you do that, nothing at all will happen. You will learn more about the default graphics format when we get to the Edit menu.

  5. File=>Notation. We are going to get a little ahead of ourselves here and do a little class diagram work so you can see what notation is all about. In the Explorer Pane select or create a class diagram. See Section 10.6, “ The Create Menu and Section 12.4.3, “ Drawing Tools. Create a class in the diagram. Go to the Detail Pane and create an attribute in the class. See Section 18.6.2, “ Class Property Toolbar. In the Properties tab of the Detail Pane change the multiplicity to "1..*". Now go the the File Menu and select Notation. Go back and forth between UML and Java observing the changes in the display in the Edit Pane.

  6. File=>Project Properties. In the Project Properties dialog it is possible to configure the project specific settings. It contains the tabs for User, Profiles, Notations and Diagram Appearance. For instance, to change the Notation language in the Project Properties dialog, click on File=>Project Properties and select the Notations tab. Set the Notation Language to UML1.4. Turn on all of the options and click Apply. Then turn off all of the options and click Apply observing the changes in the diagram. Set the Default Shadow Width to 8 and click Apply. Notice that nothing happens. This is because you are not setting the Shadow Width, but its default. The next time you create a class in a diagram, this new shadow value will apply.

Edit menu. The edit menu does not look like what you are used to in other products. There are no "Cut", "Copy", or "Paste" actions. All of the choices are peculiar to ArgoUML so we are going to cover all of them in detail.

  1. Edit=>Select.

    • Select a class diagram in the Explorer Pane. If there is none there create one using Create=>New Class Diagram. Create three classes using the class tool described in the User Interface Reference section on Class Diagram Specific Tools. Double click on it and then click in the Edit Pane for the class diagram in three different locations.

    • Undo the current mode by clicking on the "Select" tool. See Section 12.4.1, “ Layout Tools. This allows you to do things in the Edit Pane other than creating classes.

    • Open everything in the Explorer Pane tree, so that all elements are visible. Now activate the first menu item: Edit=>Select=>All. Obviously, this selects all elements, but only those on the current diagram - elements only present on another diagram are not selected this way. This function is e.g. usefull if you have a big diagram, and need to shift everything to add some more elements at the left or top of the diagram.

    • Each of the classes in the diagram has three vertically spaced sections. Double click in the top section of each class and enter a name for the class then hit the enter key. Just name the classes "A", "B", and "C". Select class A, then class B, and then class C either in the Edit Pane or in the Explorer Pane.

    • Now do an Edit=>Select=>Navigate Back. Class B should now be selected. Do another Edit=>Select=>Navigate Back. Class A should now be selected. Finally, do an Edit=>Select=>Navigate Forward. Class B should be selected again.

    • Do an Edit=>Select=>Invert Selection. Classes A and C should now be selected. Do another Edit=>Select=>Invert Selection. Class B should be selected again.

    • Do an Edit=>Remove From Diagram. Notice that class B is gone from the diagram but still exists in the Explorer Pane.

    • Select class B in the Explorer Pane, right click on it and choose "Add to Diagram". Move the cursor back onto the Edit Pane and left click on some part of the diagram where you think it will fit. You should be pretty much right back where you were before you removed it from the diagram. Do an Edit=>Delete From Model. Now class B should be gone both from the diagram and from the Explorer Pane.

  2. Edit=>Configure Perspectives. Read Section 11.5, “ Configuring Perspectives. We aren't going to go into this at this point as it needs much larger projects to be displayed than we have available at this point.

  3. Figure 3.7.  Initial ArgoUML window

    Initial ArgoUML window

    Edit=>Settings=>Preferences. Enter Help=>About ArgoUML. Look at the panel in the "splash" tab. This is known as the Splash Panel. Go to Edit=>Settings=>Preferences. Turn off the "Show Splash Panel" check button. Exit from ArgoUML and restart it. Note that the splash panel does not show during the load.

  4. Edit=>Settings=>Environment. Do File=>Export Graphics and observe the file extension that shows at the bottom of the file chooser dialog in the "Files of type" combobox. Go to the Edit=>Settings=>Environment editor and pick some other value for Default graphics format. Click "Apply" and then "OK". Go back to the File=>Export Graphics dialog and notice that the new format is now the default.

  5. Edit=>Settings=>User. Enter your name and email address.

  6. Edit=>Settings=>Appearance. Change the "Look and Feel:" to Metal." Note that the "Metal Theme:" editor becomes anabled. Change the theme to "Very Large Fonts." Click on "Apply" and then "OK." Notice that nothing has happened. Exit from ArgoUML and reopen it. The display should be markedly different. You can change it back or leave it that way as you prefer.

  7. Edit=>Settings=>Profiles. Look in the Explorer Pane under "Profile Configuration". The only folder by default is the UML 1.4 profile. Now go to Edit=>Settings=>Profiles and look at the field "Default Profiles". It only contains the same UML 1.4 profile. Add the Java profile to it, and press OK.

    If you now look at the Explorer Pane, nothing has changed, since you adapted the default setting, not the project setting. Press File=>New and check out the profiles in the Explorer Pane: it now shows both the UML 1.4 profile and the Java profile.

    Many examples in this manual presume that the Java profile is available,so you may best leave it enabled.

  8. Edit=>Settings=>Configure Shortcuts. (To be written).

  9. Edit=>Settings=>Notations. We played around with this earlier with the File=>Notation and File=>Project Properties menu items. Start another copy of ArgoUML resize each copy so they can be seen at the same time next to each other. On one of them set the Notation Language to UML (the actual choice will have a version number with it). On the other set the Notation Language to Java.

    On both of them do the following. Turn all of the check boxes on. Do a File=>New, create a class in a class diagram. Double click in the attributes section to create an attribute. Double click in the operation section to create a method. Observe the difference in the displays.

  10. Edit=>Settings=>Diagram Appearance. This settings makes it possible to change the font and font-size used in the diagrams. Hence this setting does not influence the font used in the UI of ArgoUML itself. The default font for ArgoUML is Dialog.

  11. Edit=>Settings=>Modules. This shows the currently loaded modules. We are not going to mess with it in this version of the tutorial.

View menu. This allows you to switch between diagrams, find model elements in the model, zoom in a diagram, adjust the grid, toggle page break display, and show an XML representation of the project. Do a File=>New to get back to a known point. Create an example of each diagram type not already in the Explorer Pane. Click on the (+) sign widgets in the Explorer Pane to expand the tree nodes. Select the class diagram and give it a name.

  1. View=>Goto Diagram brings up a Go To Diagram panel. Select the class diagram entry in this panel and click on the "Go to Selection" button. There should be 0 nodes and 0 edges in the description column. Click on the "Close" button. In the Details Pane (Properties tab) enter the name as "Blort". Create two classes in the class diagram and go back to View=>Goto Diagram. You should now see 2 nodes and 0 edges shown. Click on the "Close" button again and link the classes with one of the "line" items like association or generalization. Go back to View=>Goto Diagram and you should see 2 nodes and 1 edge(s). Click on the "Close" button again and create a third class. Run the mouse over the icons in the toolbar until you find the one with the tooltip "New Association Class." Click on this tool and then connect the new class to one of the others. Having clicked on the "New Association Class" tool move the mouse over the new class. Press and hold down button 1. Move the mouse over one of the other classes and release button 1. Go back to View=>Goto Diagram and you should see 3 nodes and 2 edge(s). Even though it is a class and has a two dimensional representation, it counts as an edge not a node. Select other entries in this panel and click on the "Go to Selection" button in the Go To Diagram panel. Observe the changes in the Explorer Panel.

  2. View=>Find. At this point you should have three normal classes and an association class in the Explorer Pane. Name them "AA", "AB", "B", and "C". Perform a View=>Find operation. Click on the "Find" button. Notice that an "* in *" tab is created below. This tab should show pretty much everything. In the "In Diagram" editor change the "*" to "B*" and click on the "Find Button" observing the contents of the new tab with "* in B*" as a tab label. You should see the three classes, the link (such as an association), and the association class. In the Element Type drop down box select "Interface" and click on the Find button. The new tab "* in B* Inte..." should have no entries in it as we have defined no interfaces. In the Element Type drop down box select "Class" and click on the Find button. The new tab "* in B* Class" should have one fewer entries in it than the "* in B*" tab. Switch back and forth between these two observing the difference. In various of these tabs select an item and click on the "Go To Selection" button observing the change in the selection shown in the diagram and in the Explorer Pane.

  3. View=>Zoom. As an exeption to a general rule the toolbar equivalent of View=>Zoom does not operate in the same way as the corresponding menu item. Highlight View=>Zoom. and a submenu will appear that contains "Zoom Out", "Zoom Reset" and "Zoom In". Click on these a few times observing the effect on the diagram then click on the Zoom tool bar icon. This is a magnifying glass next to a down arrow head. You should see a graduated slider bar tool. Grab the pointer in this tool and move it up and down observing the effect on the diagram.

  4. View=>Adjust Grid.

  5. View=>Adjust Snap.

  6. View=>Page Breaks.

  7. View=>XML Dump.

Create Diagram menu. This allows you to create any one of the seven UML diagram types (class, use case, state, activity, collaboration, deployment and sequence) supported by ArgoUML.

State and activity diagrams can only be created when a class or actor is selected, even though the relevant menu entries are not grayed out if this has not been done (nothing will happen under this circumstance).

Arrange menu. This allows you to align, distribute and reorder model elements on a diagram and set the layout strategy for the diagram.

Generation menu. This allows you to generate Java code for selected classes or all classes.

Critique menu. This allows you to toggle the auto-critique on and off, set the level of importance of design issues and design goals and browse the critics available.

Tools menu. This menu is permanently grayed out unless there is some tool available in your version of ArgoUML.

Help menu. This menu gives access to details of those who authored the system, and where additional help may be found.

File Toolbar. This toolbar contains some of the tools from the File menu.

Edit Toolbar. This toolbar contains some of the tools from the Edit menu.

View Toolbar. This toolbar contains some of the tools from the View menu.

Create Diagram Toolbar. This toolbar contains some of the tools from the Create Diagram menu.

3.4.2.2.  The Explorer Pane

At this time you should take the time to read Chapter 11, The Explorer . The Explorer Pane is fundamental to almost everything that you do and we will be coming back to it again and again in what follows. In fact you will recall we have had to use it already.

There is an expand or contract control in front of the package symbol for “untitledModel” in the Explorer Pane and the package symbol for “Medium” in the To-Do Pane. Click on these controls and observe that these panes are tree widgets that behave pretty much as you would expect them to. The expand or contract control is either plus (+)/minus (-) sign or knob with a right or bottom pointer depending upon the look and feel that you have chosen for an appearance.

Select alternately Class Diagram 1 and Use Case Diagram 1 observing that the detail pane changes to track to the selected item in the Explorer. The detail pane is described in Chapter 12. It is not necessary to read Chapter 12 at this point, but it couldn't hurt.

3.4.2.3.  The Editing Pane

At this point take some time to read Chapter 12, The Editing Pane .

As we go through the Editing pane changes will sometimes occur in the Details and the To-Do panes. Pay no attention to them for now. We will attend to them when we cover those panes.

Select "Class Diagram 1" in the Explorers Pane. The name is unimportant, if you have changed it, just select the new name. If you have deleted it, first perform a Create=>New Class Diagram action. Click on the "New Package" button in the Edit Pane tool bar. Click somewhere in the edit pane. In the Explorer notice that a package appears named (unnamed Package).

Double click on the "New Class" button in Edit Pane the tool bar. Click first within the package and once outside of it. Notice that within the Explorer, two classes appear in the tree both named (unnamed Class) one of them attached to the model node and the other attached to the (unnamed Package) node.

Click the Select button in the Edit Pane tool bar so you can do things in the Edit Pane without adding new Classes. In the Explorer select the class that is not subordinate to the package. This selects the corresponding class in the diagram. Grab this class and move it into the package. Notice that in the Explorer this class is now also subordinate to the package node.

In the diagram select the other class. Notice that in the Explorer, the selected node changes correspondingly. Grab this class and move it outside of the package and watch what happens in the Explorer.

3.4.2.4.  The Details Pane

At this point take some time to read Chapter 13, The Details Pane .

[Note]Note

  • To-Do Item. Discuss differences with other tabs about locations of items selected. Hold particulars for discussion of To-Do Pane.

  • Properties,

  • Documentation,

  • Presentation,

  • Source,

  • Constraints,

  • Stereotype,

  • Tagged Values,

  • Checklist.

  • Remove "images/tutorial/detailsoverview.gif" from file system.

3.4.2.5.  The To-Do Pane

At this point take some time to read Chapter 14, The To-Do Pane .

[Note]Note
  • Describe priorities.

  • Resolving items.

  • Relation to ToDo Item tab in details pane.

  • Remove "images/tutorial/todooverview.gif" from file system.

3.4.2.6.  Drawing Diagrams

In general diagrams are drawn by using the edit pane toolbar to select the model element desired and clicking in the diagram at the position required.

Model elements that are already in the model, but not on a diagram, may be added to a diagram by selecting the model element in the explorer, using Add to Diagram from the drop down menu (button 2) over that model element, and then clicking button 1 at the desired location on the diagram.

As well as UML model elements, the Edit pane toolbar provides for general drawing objects (rectangles, circles, lines, polygons, curves, text) to provide supplementary information on diagrams.

3.4.2.6.1.  Moving Diagram Elements

There are several ways to move diagram elements.

3.4.2.6.1.1.  Using the Mouse Keys

Select the elements you want to move. By holding down the Ctrl key while selecting you can select several elements to move at the same time.

Now hit your arrow keys. Your elements move a little with every key stroke.

If you also hold down the Shift key, they move a bit faster.

3.4.2.6.1.2.  Using the Edit Pane Toolbar

Click on the broom button on the toolbar. Move your mouse to the diagram pane, right click and hold. Now moving your mouse will align elements.

3.4.2.6.2.  Arranging Elements

The menu item Arrange allows you to align or group elements.

3.4.2.7.  Working with Projects

3.4.2.7.1.  The Start-Up Window

Figure 3.6, “ Initial ArgoUML window shows the ArgoUML main window as it appears as right after start-up

The main window's client area, below the menu and toolbar, is subdivided into four panes. Starting at the leftmost top pane, and working around the clock, you can see the Explorer, showing a tree view of your UML model, the Editing Pane with its toolbar, two scroll bars and gray drawing area, the Details Pane with the ToDoItem tab selected, and the To-Do Pane with a tree view of the to do items, ranked in various ways selected via the drop down list at the top of the pane.

Each time ArgoUML is started up without a project file as an argument, a new blank project is created. This project contains a model called untitledModel. This model contains a blank Class Diagram, called class diagram 1, and a blank Use Case Diagram called use case diagram 1.

The model and both empty diagrams can be seen in the explorer, which is the main tool for you to navigate through your model.

Let's assume for a moment that this is the point where you want to start modeling a new purchasing system. You want to give the name “purchasingmodel” to your model, and you want to store it in a file called FirstProject.

3.4.2.7.2.  Saving a Project - The File Menu

For now ArgoUML; saves diagrams using an earlier proposed standard, Precision Graphics Markup Language (PGML). However it has the option to export graphical data as SVG for those who can make use of it. When ArgoUML; supports UML 2.0, it will store diagrams using the UML 2.0 Diagram Interchange format.

First, let's save the model in it's current (empty and unnamed) state. On the menu bar, click on File, then on Save Project As... as shown in Figure 3.8, “ Invoking Save Project As... .

Figure 3.8.  Invoking Save Project As...

Invoking Save Project As...

Please notice that the File menu contains the usual options for creating a new project, for opening an existing project, for saving a project under a new name, for printing the currently displayed diagram, for saving the currently displayed diagram as a file, and for program Exit.

Some of these menu commands can be invoked by pressing key combinations, as indicated on the drop-down menu. For instance, holding down the “Ctrl” key, and pressing “N”, will create a new project.

In the current version, ArgoUML can only contain one active project at a time. In addition, a project can only contain one UML model. Since an UML model can contain an unlimited number of elements and diagrams, this should not present any serious limitations, even for modeling quite large and complex systems.

3.4.2.7.3.  The File Chooser Dialog

But let's go back to saving our project. After clicking on the Save Project As... menu command, we get the file chooser dialog to enter the file name we wish to use as shown in Figure 3.9, “ File Chooser Dialog.

Figure 3.9.  File Chooser Dialog

File Chooser Dialog

This is a standard Java FileChooser. Let's go over it in some detail.

The main, outstanding feature, is the scrollable folders list in the center of the dialog. By using the scroll bar on the right, you can move up and down in the list of folders contained inside the currently selected folder. If it is scrollable or not depends on the amount of files and folders shown and also how they are shown. If everything fits the window is not scrollable as seen in the picture.

Double-clicking on one of the displayed folders navigates you into that folder, allowing you to quickly navigate down into the folders hierarchy on your hard disk.

Notice that only folder names, and no file names are displayed in the scrollable area. Indeed, the dialog is currently set up in order to show only ArgoUML project files with an extension of .zargo, as can be seen on the lower drop-down control labeled Files of Type:.

Also notice that the currently selected folder's name is displayed in the upper drop-down control labeled Look in:. A single click on a folder inside the scrollable area does select that folder on screen but does not select the folder for saving.

At the top of the dialog, above the scrollable folder chooser area, there are a few more folder navigation tools.

  • The Folder drop-down control. Clicking on the down-arrow displays a tree view of the folder hierarchy, allowing you to navigate quickly up the hierarchy, and at the same time to quickly determine where in the hierarchy we are currently positioned.

  • The Folder-Up icon. Clicking on this icon will bring us to the parent folder of the current folder.

  • The Home Folder icon. Clicking on this icon will bring us to our home directory.

  • The New Folder icon. Clicking on this icon will create a new folder called "New Folder" under the current folder. After the folder is created selecting it an clicking in the name allows us to select the name of our choice.

  • The Folders Presentation Icon.

OK, now we navigate to the directory where we want to save our ArgoUML project, fill in the File name: with an appropriate name, such as “FirstProject” and click on the Save button.

You have now an active project called FirstProject, connected to the file FirstProject.zargo.

3.4.3.  Output

3.4.3.1.  Loading and Saving

3.4.3.1.1.  Saving XMI files in ArgoUML

ArgoUML saves the diagram information in a PGML file (with extension .pgml, the model information in an XMI file (with extension .xmi and information about the project in a file with extension .argo. See Section 3.4.3.2.2, “ Precision Graphics Markup Language (PGML) and Section 3.4.3.3, “XMI” for more about PGML and XMI respectively.

All of these are then zipped to a file with extension .zargo. You can easily extract the .xmi file from the .zargo file using any old generic ZIP application. Give it a try and look into the magic of Argo.

[Warning]Warning

Be aware that double clicking will launch a ZIP utility, if one is installed, and NOT Argo.

3.4.3.2.  Graphics and Printing

3.4.3.2.1.  The Graph Editing Framework (GEF)

GEF is the software package that is the foundation of the diagrams that appear in the Editing Pane. GEF was an integral part of ArgoUML but has been separated. Like ArgoUML it is an open source project available via Tigris.

3.4.3.2.2.  Precision Graphics Markup Language (PGML)

PGML is the current storage format for diagram information used in ArgoUML. In the future, PGML will be replaced by the UML 2.0 Diagram Interchange format.

3.4.3.2.3.  Applications Which Open PGML

PGML is a predecessor of SVG (see Section 3.4.3.2.5, “ Scalable Vector Graphics (SVG). It was dropped by the W3C Consortium.

Currently there are no other tools that we know of working on PGML.

3.4.3.2.4.  Printing Diagrams

Select a diagram, then go to File=>Export Diagrams. You can generate GIF, PostScript, Encapsulated PostScript or SVG format.

3.4.3.2.5.  Scalable Vector Graphics (SVG)

A World Wide Web Consortium (W3C) standard vector graphics format ( http://www.w3.org/TR/SVG/).

Support is built in to modern browsers, but you can also get a plugin for older browsers from adobe.com.

3.4.3.2.6.  Saving Diagrams as SVG
  1. Select .svg as the file type.

  2. Type the name of the file as you like with the .svg tag at the end. Example myumldiagram.svg

Et viola! SVG! Give it a try and zoom around a little... They are not pretty though, so if you know anything about rendering beautiful SVG let us know.

Most modern browsers support SVG. If yours doesn't try Firefox or get a plugin for your current browser from adobe.com

[Note]Note

You will not have scroll bars for your SVG unless it is embedded in HTML. Good luck and let us know what you find!

3.4.3.3. XMI

ArgoUML supports XMI 1.0, 1.1, and 1.2 files which contain UML 1.3 and UML 1.4 models. For best compatibility with ArgoUML, export your models using UML 1.4 and XMI 1.1 or 1.2. Be sure to turn off any proprietary extensions (such as Poseidon's diagram data).

With UML versions earlier than UML 2.0, it isn't possible to save diagram information, so no diagrams will be transferred.

There is also a tool that converts XMI to HTML. For more information, see http://www.objectsbydesign.com/projects/xmi_to_html_2.html.

3.4.3.3.1.  Using XMI from Rational Rose

...

3.4.3.3.2.  Using Models Created by Poseidon

In the Export project to XMI dialog, but sure to clear the selection of Save with diagram dataliteral>.

3.4.3.3.3.  Using Models Created by MagicDraw

...

3.4.3.3.4.  XMI Compatibility with other versions of ArgoUML

Versions of ArgoUML prior to 0.19.7 supported UML 1.3/XMI 1.0. After this time, the save format is UML 1.4/XMI 1.2 which is not backward compatible. Newer versions of ArgoUML will read projects written by older versions, but not vice versa. If you might need to return to an older version of ArgoUML you should be careful to save a backup of your old projects.

Additionally, if you write XMI files which need to be read by other tools, you should take into account the different versions. Most modern UML modelling tools should read UML 1.4, but you may have in-house code generators or other tools which are tied to UML 1.3.

3.4.3.3.5.  Importing Other XMI Formats into ArgoUML Andere XMI-Formate in ArgoUML importieren

XMI compatibility between UML modeling tools has improved over the years, but you may still occasionally run into problems.

ArgoUML will not read XMI files which contain UML 1.5 or UML 2.0 models, but it should be able to open most UML 1.4 and UML 1.3 files. If you find one that it can't open, please file a bug report so that a developer can investigate.

3.4.3.3.6.  Generating XMI Format

Select the command File=>Export as XMI and choose a filename.

3.4.3.4.  Code Generation

3.4.3.4.1.  Code Generated by ArgoUML

It is possible to compile your generated code with ArgoUML, you still need to implement method bodies, though, to get usable results.

3.4.3.4.2.  Generating Code for Methods

At the moment you cannot write code for methods (operations) within ArgoUML. The source pane is editable, but the changes are ignored. ArgoUML is a pure design tool for now, no IDE functionality but the desire is there. You might consider using Forte and ArgoUML together—it's a good work around!

You can help us out there if you'd like!

3.4.4.  Working With Design Critics

Design critics are part of the practical application of the theories of Cognitive Psychology that are implemented in ArgoUML. See Section 3.3.1, “ Cognitive Psychology

3.4.4.1.  Messages From the Design Critics

Where do we stand now? A new project has been created, and is stored in the file FirstProject.zargo. Figure 3.10, “ ArgoUML Window Having Saved FirstProject.zargo shows how your ArgoUML window should look at this stage.

Figure 3.10.  ArgoUML Window Having Saved FirstProject.zargo

ArgoUML Window Having Saved FirstProject.zargo

The project contains a top-level package, called untitledModel, which contains a class diagram and a use case diagram.

If we look carefully at the screen, we can see that the "Medium" folder in the To-Do Pane (the lower left pane) must contain some items, since its activation icon is displayed.

Clicking on this icon will open the "Medium" folder. An open folder is indicated by the icon.

But what is this “To-Do” Pane anyway. You haven't recorded anything yet that has to be done, so where do these to do items originate.

The answer is simple, and is at the same time one of the strong points of ArgoUML. While you are working on your UML model, your work is monitored continuously and invisibly by a piece of code called a design critic. This is like a personal mentor that watches over your shoulder and notifies you each time he sees something questionable in your design.

Critics are quite unobtrusive. They give you a friendly warning, but they do not force you into design principles that you don't want or like to follow. Let us take a look at what the critics are telling us. Click on the icon next to the Medium folder, and click on the Revise Package Name UntitledModel item.

Figure 3.11, “ ArgoUML Window Showing the Critic Item Revise Package Name UntitledModel shows how your screen should now look.

Figure 3.11.  ArgoUML Window Showing the Critic Item Revise Package Name UntitledModel

ArgoUML Window Showing the Critic Item Revise Package Name UntitledModel

Notice that your selection is highlighted in red in the To-Do Pane, and that a full explanation appears now in the Details Pane (the lower right pane). You may have to re-size your Details Pane or to scroll down in order to see the full message as displayed in our example.

What ArgoUML is trying to tell you is that usually, package names are written in lower cases. The default top level package created by ArgoUML is called untitledModel and therefore violates a sound design principle. (Actually, this could be considered as a bug within ArgoUML, but it comes in handy to demonstrate the working of critics).

At this point, you can choose to change the package name manually or to impose silence on the design critic for some time or permanently

We will do nothing of this (we'll come back to it when we talk about the design critics in more detail) but we'll use another handy feature of ArgoUML—an auto-correct feature.

In order to do that, just click on the Next button on the Details Pane. This will cause a renaming wizard to be displayed inside the properties panel, proposing to use the name untitledmodel (all in lower case).

3.4.4.2.  Design Critics at Work: The Rename Package Wizard

Replace the name untitledmodel with purchasingmodel, and click on the Finish button. Figure 3.12, “ ArgoUML Window Showing the Critic Wizard to Rename the Package shows how the ArgoUML window will now look.

Figure 3.12.  ArgoUML Window Showing the Critic Wizard to Rename the Package

ArgoUML Window Showing the Critic Wizard to Rename the Package

Watch now how the design critic note in the To Do panel disappears, leaving only the Add Elements to Package purchasingmodel note in the To-Do list.

If this doesn't happen at once, wait for a few seconds. ArgoUML makes heavy use of several threads of execution that execute in parallel. This can cause delays of a few seconds before the information gets updated on the screen.

The package name change should also be reflected in the explorer, in the top left corner of your ArgoUML window.

We are now ready to create our first UML diagram, a Use Case diagram, but first let's save what we've done so far.

Click on the File menu item, and select Save Project. You can now safely exit ArgoUML without losing your work so far, or go on creating your first diagram.

3.5.  The Case Study (To be written)

Here is where we are going to start the Case Study. It is at this point that you define your project; not your product, but your project. It can be argued that modeling concepts should apply here as well but this has not been well established. If you can take the time to look into the ArgoUML project, you will find that there are a large number of "lines of code" and lines of documentation that are part of the project, but not part of the product. For example, this document is part of the product while the Cookbook and the build.xml files are part of the project only. At a minimum the file structure of the project could be shown in a package diagram.

...

Chapter 4.  Requirements Capture

4.1.  Introduction

Requirements capture is the process of identifying what the “customer” wants from the proposed system.

The key at this stage is that we are in the problem domain. At this stage we must describe everything from the “customer” perspective and in the language of the “customer”.

The biggest risk we have in requirements capture is to start thinking in terms of possible solutions. That must wait until the Analysis Phase (see Chapter 5, Analysis ). One of the steps of the Analysis Phase will be to take the output of the Requirements Phase and recast it in the language of a deemed solution.

Remember we are using both a incremental, and an iterative process.

We may well come back to the requirements process again as we break down the problem into smaller chunks, each of which must have its requirements captured.

We will certainly come back through the requirements phase on each iteration as we seek to define the requirements of more and more of the system

[Note]Note

The only part of the requirements notation specified by the UML standard is the use case diagram. The remainder is process specific. The process described in this chapter draws heavily on the Rational Unified Process.

4.2.  The Requirements Capture Process

We start with a top-level view of the problem we are solving and the key areas of functionality that we must address in any solution. This is our vision document, and should be just a few pages long.

For example the top-level view of an automated teller machine (ATM) might be that it should support the following.

  1. Cash deposit, cash withdrawal and account inquiries by customers.

  2. Maintenance of the equipment by the bank's engineers, and unloading of deposits and loading of cash by the local bank branch.

  3. Audit trail for all activities sent to the bank's central computer.

From this top-level view we can extract the principal activities of the system, and the external agents (people, equipment) that are involved in those activities. These activities are known as use cases and the external agents are known as actors.

Actors may be people or machines. From a practical standpoint it is worth knowing the stakeholder behind any machine, since only they will be able to engage with the requirements capture process.

Use cases should be significant activities for the system. For example customer use of the ATM machine is a use case. Entering a PIN number is not.

There is a gray area between these two extremes. As we shall see it is often useful to break very large use cases into smaller sub-use cases. For example we may have sub-use cases covering cash deposit, cash withdrawal and account inquiry.

There is no hard and fast rule. Some architects will prefer a small number of relatively large use cases, others will prefer a larger number of smaller use cases. A useful rule of thumb is that any practical project ought to require no more than about 30 use cases (if it needs more, it should be broken into separate projects).

We then show the relationship between use cases and actors on one or more use case diagrams. For a large project more than one diagram will be needed. Usually groups of related use cases are shown on one diagram.

We must then give a more detailed specification of each use case. This covers its normal behavior, alternative behaviors and any pre- and post-conditions. This is captured in a document variously known as a use case specification or use case scenario.

Finally, since use cases are functional in nature, we need a document to capture the non-functional requirements (capacity, performance, environmental needs etc). These requirements are captured in a document known as a supplementary requirements specification.

4.2.1.  Process Steps

The steps in the requirements capture process can be summarized as follows.

  1. Capture an overall view of the problem, and the desired characteristics of its solution in the vision document.

  2. Identify the use case and actors from the vision document and show their relationships on one or more use case diagrams.

  3. Give detailed use case specifications for each use case, covering normal and alternate behavior, pre- and post-conditions.

  4. Capture all non-functional requirements in a supplementary requirements specification.

In any iterative development process, we will prioritize, and early iterations will focus on capturing the key behavior of the most important use cases.

Most modern requirements capture processes agree that it is essential that the authoritative representative of the customer is fully involved throughout the process.

4.3.  Output of the Requirements Capture Process

Almost all the output of the requirements capture process is documentary. The only diagram is the use case diagram, showing the relationships between use cases and actors.

4.3.1.  Vision Document

Typical sections of this document would be as follows.

  • Summary. A statement of the context, problem and solution goals.

  • Goals. What are we trying to achieve (and how do we wish to achieve it).

  • Market Context or Contractual Arrangements. For a market led development, this should indicate target markets, competitive differentiators, compelling events and so forth. For a contractual development this should explain the key contractual drivers.

  • Stakeholders. The users (in the widest sense) of the system. Many of these will map in to actors, or control equipment that maps into actors.

  • Key Features. At the very highest level what are they key functional aspects of the problem/desired solution. These will largely map down to the use cases. It is helpful to give some prioritization here.

  • Constraints. A high level view of the non-functional parameters of the system. These will be worked out in detail in the supplementary requirements specification.

  • Appendix. A listing of the actors and use cases that will be needed to meet this vision. It is useful to link to these from the earlier sections to ensure comprehensive coverage.

4.3.2.  Use Case Diagram

The vision document has identified the use cases and actors. The use case diagram captures how they interact. In our ATM example we have identified “customer uses machine”, “maintain machine” and “audit” as the three main use cases. We have identified “customer”, maintenance engineer“,”“local branch official” and “central computer” as the actors.

Figure 4.1, “ Basic use case diagram for an ATM system shows how this could be displayed on a use case diagram. The use cases are shown as ovals, the actors as stick people (even where they are machines), with lines (known as associations connecting use cases to the actors who are involved with them. A box around the use cases emphasizes the boundary between the system (defined by the use cases) and the actors who are external.

[Note]Note

Not all analysts like to use a box around the use cases. It is a matter of personal choice.

Figure 4.1.  Basic use case diagram for an ATM system

Basic use case diagram for an ATM system

The following sections show how the basic use case diagram can be extended to show additional information about the system being designed.

4.3.2.1.  Active and Passive Actors

Active actors initiate interaction with the system. This can be shown by placing an arrow on the association from the actor pointing toward the use case. In the ATM example, the customer is an active actor.

Interaction with passive actors is initiated by the system. This can be shown by placing an arrow on the association from the use case pointing toward the actor. In the ATM example, the central computer is a passive actor.

This is a good example where the arrow helps, since it allows us to distinguish an event driven system (the ATM initiates interaction with the central computer) from a polling system (the central computer interrogates the ATM from time to time).

Where an actor may be either active or passive, depending on circumstances, the arrow may be omitted. In the ATM example the bank engineer fits into this category. Normally he is active, turning up on a regular cycle to service the machine. However if the ATM detects a fault, it may summon the engineer to fix it.

The use of arrows on associations is referred to as the navigation of the association. We shall see this used elsewhere in UML later on. The choice, by the OMG, of zero vice two arrowheads to show a bidirectional association is unfortunate. Under this convention you cannot distinguish between an association whose navigation has yet to be determined and one that is bidirectional.

Figure 4.2, “ Use case diagram for an ATM system showing navigation. shows the ATM use case diagram with navigation displayed.

Figure 4.2.  Use case diagram for an ATM system showing navigation.

Use case diagram for an ATM system showing navigation.

4.3.2.2.  Multiplicity

It can be useful to show the multiplicity of associations between actors and use cases. By this we mean how many instances of an actor interact with how many instances of the use case.

By default we assume one instance of an actor interacts with one instance of a use case. In other cases we can label the multiplicity of one end of the association, either with a number to indicate how many instances are involved, or with a range separated by two periods (..). An asterisk ( *) is used to indicate an arbitrary number.

In the ATM example, there is only one central computer, but it may be auditing any number of ATM uses. So we place the label 0..* at the use case end. There is no need for a label at the other end, since the default is one.

A local bank will have up to three officials authorized to unload and load ATM machines. So at the actor end of the relationship with the use case Maintain ATM, we place the label 1..3. They may be dealing with any number of ATM machines, so at the other end we place the label 0..*.

There may be any number of customers and there may be any number of ATM systems they could use. So at each end of the association we place the label 0..*.

Figure 4.3, “ Use case diagram for an ATM system showing multiplicity. shows the ATM use case diagram with multiplicity displayed.

Figure 4.3.  Use case diagram for an ATM system showing multiplicity.

Use case diagram for an ATM system showing multiplicity.

Multiplicity can clutter a diagram, and is often not shown, except where it is critical to understanding. In the ATM example we would only choose to show 1..3 against the local bank official, since all others are obvious from the context.

4.3.2.3.  Hierarchies of Use Cases

In our ATM example so far we have just three use cases to describe all the behavior of the system. While use cases should always describe a significant chunk of system behavior, if they are too general they can be difficult to describe.

We could for example define the behavior of the use case “Use ATM” in terms of the behavior of three simpler use cases, “Deposit Cash”, “Withdraw Cash” and “Query Account”. The main use case could be specified by including the behavior of the subsidiary use cases where needed.

Similarly the “Maintain ATM” use case could be defined in terms of two use cases “Maintain Equipment” and “Reload ATM”. In this case the two actors involved in the main use case are really only involved in one or other of the two subsidiary use cases and this can be shown on the diagram.

The decomposition of a use case into simpler sub-use cases is shown in UML by using an include relationship, a dotted arrow from the main use case to the subsidiary, with the label «include».

Figure 4.4.  Use case diagram for an ATM system showing include relationships.

Use case diagram for an ATM system showing include relationships.

Include relationships are fine for breaking down the use case behaviors in to hierarchies. However we may also want to show a use case that is an extension to an existing use case to cater for a particular circumstance.

In the ATM example we have a use case covering routine maintenance of the ATM, “Maintain Equipment”. We also want to cover the special case of an unscheduled repair caused by the ATM detecting an internal fault.

This is shown in UML by the extend relationship. In the main use case, we specify a name for a location in the description, where an extension to the behavior could be attached. The name and location are shown in a separate compartment within the use case oval. The representation extend relationship is the same as the include relationship, but with the label «extend». Alongside the extend relationship, we specify the condition under which that behavior will be attached.

Figure 4.5, “ Use case diagram for an ATM system showing an extend relationship. shows the ATM use case diagram with an extend relationship to a use case for unscheduled repairs. The diagram is now getting rather complex, and so we have split it into two, one for the maintenance side of things, the other for customer usage and audit.

The “Maintain Equipment” use case defines a name “Unsched”, at the start of its description. The extending use case “Unscheduled Repair” is attached there when the ATM detects an internal error.

Figure 4.5.  Use case diagram for an ATM system showing an extend relationship.

Use case diagram for an ATM system showing an extend relationship.

Use cases may be linked together in one other way. One use case may be a generalization of a subsidiary use case (or alternatively the subsidiary is a specialization of the main use case). This is very like the extends relationship, but without the constraint of specific extension points at which the main use case may be extended, and with no condition on when the subsidiary use case may be used.

Generalization is shown on a use case diagram by an arrow with solid line and solid white head from the subsidiary to the main use case. This may be useful when a subsidiary use case specializes the behavior of the main use case at a large number of positions and under a wide range of circumstances. However the lack of any restriction makes generalization very hard to specify precisely. In general use an extend relationship instead.

4.3.3.  The Use Case Specification

Each use case must be documented to explain in detail the behavior it is specifying. ArgoUML assists in this area through the generation of graphic files for inclusion in this documentation. This document is known by different names in different processes: use case specification, use case scenario or even (confusingly) just use case.

A typical use case specification will include the following sections.

  • Name. The name of the use case to which this relates.

  • Goal. A one or two line summary of what this use case achieves for its actors.

  • Actors. The actors involved in this use case, and any context regarding their involvement.

    [Note]Note

    This should not be a description of the actor. That should be associated with the actor on the use case diagram.

  • Pre-condition. These would be better named “pre-assumptions”, but the term used everywhere is pre-conditions. This is a statement of any simplifying assumptions we can make at the start of the use case.

    In the ATM example we might make the assumption for the“Maintain Equipment” use case that an engineer is always available, and we do not need to worry about the case where a routine maintenance visit is missed.

    [Caution]Caution

    Avoid pre-conditions wherever possible. You need to be absolutely certain that the pre-condition holds under all possible circumstances. If not your system will be under specified and hence will fail when the pre-condition is not true. Alternatively, when you cannot be certain the pre-condition is always true, you will need to specify a second use case to handle the pre-condition being false. In the first case, pre-conditions are a source of problems, in the second a source of more work.

  • Basic Flow. The linear sequence of steps that describe the behavior of the use case in the “normal” scenario. Where a use case has a number of scenarios that could be normal, one is arbitrarily selected. Specifying the basic flow is described in more detail in Section 4.3.3.1, “ Specifying the Basic Flow below.

  • Alternate Flows. A series of linear sequences describing each of the alternative behaviors to the basic flow. Specifying alternate flows is described in more detail in Section 4.3.3.2, “ Specifying the Alternate Flows .

  • Post-conditions. These would be better named “post-assumptions”. This is a statement of any assumptions that we can make at the end of the use case. Most useful where the use case is one of a series of subsidiary use cases that are included in a main use case, where they can form the pre-conditions of the next use case to be included.

    [Caution]Caution

    Like pre-conditions, post-conditions are best avoided. They place a burden on the specification of the use case flows, to ensure that the post-condition always holds. They therefore are also a source of problems and extra work.

  • Requirements. In an ideal world the vision document, use case diagrams, use case specifications and supplementary requirements specification would form the requirements for a project.

    For most market-led developments, where ownership of requirements is within the same business as the team who will do the development, this is now usually the case. The marketing department can learn use case based requirements capture and analysis to link to their customer facing activities.

    However for external contract developments, customers may insist on a traditional “list of features” as the basis of the contract. Where this is the case, this section of the use case specification should link to the contract features that are covered by the use case.

    This is often done through a third party tool that can link documents, providing automated checking of coverage, in which case this section is not needed, or may be generated automatically.

The final size of the use case specification will depend on the complexity of the use case. As a rule of thumb, most use cases take around 10-15 pages to specify, the bulk of which is alternate flows. If you are much larger than this, consider breaking the use case down. If you are much smaller consider whether the use case is addressing too small a chunk of behavior.

4.3.3.1.  Specifying the Basic Flow

All flows in a use case specification are linear—that is there is no conditional branching. Any choices in flows are handled by specifying another alternate flow that takes over at the choice point. It is important to remember we are specifying behavior here, not programming it.

A flow is specified as a series of numbered steps. Each step must involve some interaction with an actor, or at least generate a change that is observable externally by an actor. Requirements capture should not be specifying hidden internal behavior of a system.

For example we might give the following sequence of steps for the basic flow of the use case "Withdraw Cash" in our ATM example.

  1. Customer indicates a receipt is required.

  2. Customer enters amount of cash required.

  3. ATM verifies with the central computer that the customer can make this withdrawal.

  4. ATM dispenses cash to the customer.

  5. ATM issues receipt to customer.

Remember this is a sub-use case included in the main “Use ATM” use case, which will presumably handle checking of cards and PINs before invoking this included use case.

[Note]Note

The first step is not a condition. We take as our basic flow the case where the customer does want a receipt. The case where the customer does not want a receipt will be an alternative flow.

4.3.3.2.  Specifying the Alternate Flows

This captures the alternative scenarios, as linear flows, by reference to the basic flow. Initially we just build a list of the alternate flows.

    1. Customer does not require a receipt.

    2. Customer's account will not support the withdrawal.

    3. Communication to the central computer is down.

    4. The customer cancels the transaction.

    5. The customer fails to take the dispensed cash.

Subsequently we flesh out each alternate flow, by reference to the basic flow. For example the first alternate flow might look like.

    1. Customer does not require a receipt.

      1. At step 1 of the basic flow the customer indicates they do not want a receipt.

      2. The basic flow proceeds from step 2 to step 4, and step 5 is not used.

The convention is to number the various alternate flows as A.1, A.2, A.3, etc. The steps within an alternate flow are then numbered from this. So the steps of the first alternate flow would be A.1.1, A.1.2, A.1.3, etc.

4.3.3.3.  Iterative Development of Use Case Specifications

Iterative development will prioritize the use cases, and the first iterations will address the most important.

Early iterations will capture the basic flows of the most important use cases with only essential detail and list the headings of the main alternate flows.

Later iterations will address the remaining use cases, flesh out the steps on individual alternate flows and possibly provide more detail on individual steps.

4.3.4.  Supplementary Requirement Specification

This captures the non-functional requirements or constraints placed on the system. Since use cases are inherently functional in nature, they cannot capture this sort of information.

[Note]Note

Some analysts like to place non-functional requirements in a section at the end of each use case specification, containing the non-functional requirements relevant to the use case.

This can cause some problems. First key non-functional requirements (for example about performance) may need to appear in many use cases and it is bad practice to replicate information. Secondly there are invariably some non-functional requirements that are system wide and need a system wide document. Hence my preference for a single supplementary requirements specification.

There should be a section for each of the main areas of non-functional requirements. The checklist provided by Ian Sommerville in his book Software Engineering (Third Edn, Addison-Wesley, 1989) is a useful guide.

  • Speed. Processor performance, user/event response times, screen refresh time.

  • Size. Main memory (and possibly caches), disc capacity.

  • Ease of use. Training time, style and detail of help system.

  • Reliability. Mean time to failure, probability of unavailability, rate of failure, availability.

  • Robustness. Time to restart after failure, percentage of events causing failure, probability of data corruption on failure.

  • Portability. Percentage of target-dependent code/classes, number of target systems.

To this we should add sections on environment (temperature, humidity, lightening protection status) and standards compliance.

4.4.  Using Use Cases in ArgoUML

ArgoUML allows you to draw use case diagrams. When you create a new project it has a use case diagram created by default, named use case diagram 1. Select this by button 1 click on the diagram name in the explorer (the upper left quadrant of the user screen).

New use case diagrams can be created as needed through Create Diagram on the main menu bar or on the Create Diagram Toolbar. They are edited in the editing pane (the upper right quadrant of the user screen).

4.4.1.  Actors Akteure

To add an actor to the diagram use button 1 click on the actor icon on the editing pane toolbar ( ) and then button 1 click at the location where you wish to place it. The actor can be moved subsequently by button 1 motion (i.e. button 1 down over the actor to select it, move to the new position and button 1 release to drop the actor in place.

Multiple actors can be added in one go, by using button 1 double click on the actor icon. Each subsequent button 1 click will drop an actor on the diagram. A button 1 click on the select icon ( ) will stop adding actors.

The actors name is set in its property panel. First select the actor (if not already selected) on the editing pane using button 1 click. Then click on the Properties tab in the details pane. The name is entered in the name field, and will appear on the screen.

As a shortcut, double button 1 click on the name of the actor in the editing pane (or just typing on the keyboard when an actor is selected) will allow the name to be edited directly. This is a convenient way to enter a name for a new actor.

Having created the actor, you will see it appear in the explorer (the upper left quadrant of the user screen). This shows all the model elements created within the UML design. A drop down at the top of the explorer controls the ordering of model elements in the explorer. The most useful are the Package-centric (default) and Diagram-centric. The latter shows model elements grouped by the diagram on which they appear.

4.4.2.  Use Cases

The procedure for adding use cases is the same as that for adding actors, but using the use case icon on the editing pane toolbar ().

By default use cases in ArgoUML do not display their extension points (for use in extend relationships). You can show the extension point compartment in one of two ways.

  1. Select the use case in the editing pane with button 1 click, then select the Style tab in the details pane and button 1 click on the Display: Extension Points check box.

  2. Use button 2 click over the use case in the editing pane to display a context-sensitive pop-up menu and from that choose Show/Show Extension Point Compartment.

The same approaches can be used to hide the extension point compartment.

4.4.2.1.  Adding an Extension Point to a Use Case

There are two ways to add an extension point to a use case.

  1. Select the use case on the editing pane with button 1 click. Then click on the Add Extension Point icon ( ) on the toolbar, and a new extension point with default name and location will be added after any existing extension points.

    [Note]Note

    The Add Extension Point icon is grayed out and unusable until a use case is selected.

  2. Select the use case on the editing pane with button 1 click and then select its property tab in the details pane. A button 2 click over the Extension Points: field will bring up a context-sensitive pop-up menu. Select Add to add a new extension point.

    If any extension points already exist, they will be shown in this field on the property tab. The new extension point will be inserted immediately before the entry over which the pop-up menu was invoked. This ordering can be changed later by using the Move Up and Move Down entries on the pop-up menu.

Whichever method is used, the new extension point is selected, and its property tab can be displayed in the details pane. The name and location of the extension point are free text, set in the corresponding fields of the property tab.

An existing extension point can be edited from its property tab. The property tab can be reached in two ways.

  1. If the extension point compartment for the use case is displayed on the diagram, select the use case with button 1 click and then select the extension point with a further button 1 click. The property tab can then be selected in the details pane.

  2. Otherwise select the use case and its property tab in the details pane. A button 1 click on the desired entry in the Extension Points field will bring up the property tab for the extension point in the details pane.

The name and location fields of the extension point may then be edited.

As a shortcut, where the extension point compartment is displayed, double click on the extension point allows text to be typed in directly. This is parsed to set name and location for the extension point.

Extension points may be deleted, or their ordering changed by using the button 2 pop-up menu over the Extension Points field in the use case property tab.

Having created an extension point, it will appear in the explorer (upper left quadrant of the user screen). Extension points are always shown in a sub-tree beneath their owning use case.

4.4.3.  Associations

To join a use case to an actor on the diagram use button 1 click on the association icon on the editing pane toolbar (). Hold button 1 down at the use case, move to the actor and release button 1 (or alternatively start at the actor and finish at the use case).

This will create a straight line between actor and use case. You can segment the line by holding down button 1 down on the line and moving before releasing. A vertex will be added to the line, which you can move by button 1 motion. A vertex can be removed by picking it up and sliding to one end of the line.

Multiple associations can be added in one go, by using button 1 double click on the association icon. Each subsequent button 1 down/motion/release sequence will join an actor to a use case. Use button 1 on the select icon ( ) to stop adding associations.

It is also possible to add associations using small “handles” that appear to the left and right of a use case or actor when it is selected and the mouse is over it. Dragging the handle from a use case to an actor will create an association to that actor (and similarly by dragging a handle from an actor to a use case).

Dragging a handle from a use case into empty space will create a new actor to go on the other end. Similarly dragging a handle from an actor into empty space will create a new use case.

It is possible to give an association a name, describing the relationship of the actor to the use case, although this is not usually necessary. This is done through the property tab of the association. Such a name appears alongside the association near its center.

4.4.3.1.  Setting Navigation

There are two ways of setting the navigation of an association.

  1. Use button 2 click on the association to bring up a context-sensitive pop-up menu. The Navigability sub-menu has options for bi-directional navigation (the default, with no arrows) and for navigability Actor->Use Case and Use Case->Actor.

  2. Use button 1 to select the association and select its property tab in the details pane. This shows a field named Association Ends:, with entries for each end labeled by the actor or use case name and its multiplicity. Select the end that should be at the tail of the arrow with button 1 click. This brings up the property tab for the association end. Use button 1 click to uncheck the Navigability box.

    [Note]Note

    This may seem counter-intuitive, but in fact associations by default are navigable in both directions (when no arrows are shown). This process is turning off navigation at one end, rather than turning it on at the other.

You will see it is possible to give an association end a name in its property tab. This name will appear at that end of the association, and can be used to indicate the role being played by an actor or use case in an association.

For example a time management system for a business may have use cases for completing time sheets and for signing off time sheets. An employee actor may be involved in both, one as an employee, but the other in a role as manager.

4.4.3.2.  Setting Multiplicity

There are two ways of setting multiplicity at the end of an association.

  1. Button 2 click over the end of an association will cause a context-sensitive pop-up menu to appear with a sub-menu labeled Multiplicity. This allows you to select from 1 (the default), 0..1, 0..* and 1..*.

  2. Bring up the property sheet for the association end as described for setting navigation (see the second option in Section 4.4.3.1, “ Setting Navigation ). A drop down menu gives a range of multiplicity options that may be selected.

The second of these two approaches has a wider range of options, although ArgoUML does not currently allow the user to set an arbitrary multiplicity.

4.4.4.  Hierarchical Use Cases

UML as originally designed allowed use cases to be organized by grouping them in packages as well as by specifying relations among them. In ArgoUML only the relations mechanism is supported. All Three of the relations that apply to use cases are supported. These are include, extend and generalization.

4.4.4.1.  Includes

The procedure for adding an include relationship is the same as that for adding an association, but using the include icon from the editing pane toolbar ( ) to join two use cases.

Since include relationships are directional the order in which the two ends are selected is important. The including (main) use case should be selected first (button 1 down) and the included (subsidiary) use case second (button 1 release).

It is possible to name include relationships using the property tab, but this is rarely done, and will not be displayed on the use case diagram.

4.4.4.2.  Extends

The procedure for adding an extend relationship is the same as that for adding an include relationship, but using the extend icon from the editing pane toolbar ( ) to join two use cases.

As with include relationships, the order of selection matters. In this case, the extending (subsidiary) use case should be selected first (button 1 down) and the extending (main) use case second (button 1 release).

[Note]Note

This is the reverse of the include relationship, but reflects the way that designer's tend to think. The fact that the extend icon's arrow points upward (the opposite of the include icon) should help remind you of this.

To set a condition for the extend relationship, select the extend relationship in the editing pane (button 1 click) and then bring up its property tab in the details pane ((button 1 click on the tab). The text of the condition may be typed in the Condition field. Long conditions may be split over several lines if desired. The condition is displayed under the «extend» label on the diagram.

It is possible to name extend relationships using the property tab, but this is rarely done, and will not be displayed on the use case diagram.

4.4.4.3.  Generalization

The procedure for adding generalizations, is the same as for adding extend relationships, but using the generalization icon from the editing pane toolbar ( ).

Since generalization is a directed relationship, the order of selection matters. The specialized use case should be selected first (button 1 down) and the generalized second (button 1 release).

It is also possible to add generalizations using small “handles” that appear to the top and bottom of a use case when it is selected. Dragging the handle at the top to another use case will create a generalization. The original use case is the specializing end, and the use case to which the handle was dragged will be the generalizing end. Dragging into empty space will create a new use case to be the generalizing end.

Similarly dragging on the bottom handle will create a generalization in which the original use case is the generalizing end.

Generalization is also permitted between actors, although its use is beyond the scope of this tutorial. Unlike use cases there are no generalization handles on actors, so generalizations must be created using the toolbar icon.

It is possible to name generalization relationships using the property tab, but this is rarely done. If a name is provided, it will be displayed on the use case diagram.

4.4.5.  Stereotypes

UML has the concept of stereotyping as a way of extending the basic notation. It may prove useful for example to model a problem at both the business level and the engineering level. It is for this reason that the OMG distinguishes between a PIM and a PSM. For both of these we will need use cases, but the use cases at the business level hold a different sort of information to those at the engineering level. Very likely they use different language and notation in their underlying use case specifications.

Stereotypes are used to label UML model elements such as use cases, to indicate that they belong to a certain category. Such labels are shown in guillemots ( « ») above the name of the model element on the diagram. The UML standard defines a number of standard stereotypes, and the user may define more stereotypes of his own.

You will see that ArgoUML has a drop down selector, Stereotype on every property tab. This is populated with the standard stereotypes, to which you may add your own user defined ones.

The details of stereotyping are beyond the scope of this tutorial. The reference manual (see Section 16.6, “ Stereotype) documents the support provided in ArgoUML.

[Warning]Warning

ArgoUML is missing a few of the standard UML stereotypes. In addition not all model elements will actually display the stereotype on the diagram. At present this includes use cases and actors.

4.4.6.  Documentation

ArgoUML has some simple documentation facilities associated with model elements on a diagram. In general these should be used only to record the location of material in documents that can be handled by a mainstream editor or word processor, not the actual documentation itself.

Documentation for a particular model element is recorded through the documentation tab in the details pane (the quadrant of the user screen at the bottom right).

In addition annotation may be added to diagrams using the text icon on the editing pane toolbar ( ).

The recommendation is that a use case diagram should use the documentation tab of actors to record information about the actor, or if the actor is complex to refer to a separate document that holds information about the actor.

The documentation tab of use cases should record the location of the use case specification. The information in a use case specification (for all but the simplest use cases) is too complex to be placed directly in the tab.

The project should also have a separate vision document and supplementary requirements specification. A text annotation on diagrams may be used to refer to these if the user finds this helpful.

[Warning]Warning

The documentation tab includes a Deprecated check box. The state of this flag is not preserved over save and load in the current release of ArgoUML

4.4.7.  System Boundary Box

ArgoUML provides a series of tools to provide arbitrary graphical annotation on diagrams (we have already mentioned the text tool). These are found at the right hand end of the editing pane toolbar and are fully documented in the reference manual (see Chapter 12, The Editing Pane ).

The rectangle tool can be used to draw the boundary box. Use the button 2 context-sensitive Ordering pop-up menu to place it behind everything else. However there is no way to change its fill color from the default white. You may therefore prefer to draw the boundary box as four lines. This is the method used for the diagrams in this chapter.

[Note]Note

The editing pane in ArgoUML has a grid to which objects snap to aid in drawing. The size of this grid and its effect may be altered through the View menu (using Adjust Grid and Adjust Grid Snap). This is described fully in the reference manual (see Chapter 10, The Menu bar ).

4.5.  Case Study

4.5.1.  Vision Document

A vision document contains more than those things needed for the modeling effort. It also contains financial and scheduling pertinent information. The following sections are those parts of the Vision Document spelled out in Section 4.3.1, “ Vision Document above. In practice this format need not be followed religiously, but is used here for consistency.

4.5.1.1.  Summary

The company wishes to produce and market a line of ATM devices. The purpose of this project is to produce the hardware and the software to drive it that are both maintainable and robust.

4.5.1.2.  Goals

To produce better designed products based on newer technology. Follow the MDA philosophy of the OMG by producing first a Platform Independent Model (PIM). As current modeling technology does not admit of maintaining the integrity of the connection between the PIM and Platform Specific Models (PSMs), the PIM will become comparatively stable before the first iteration of the PSM is produced. The software platform will be Java technology. The system will use a simple userid (from ATM card) and password (or PIN) mechanism.

4.5.1.3.  Market Context

Equipment currently on the market is based on older technology for both hardware and software. This technology has not reached the end of its useful life, making it unlikely that the vendors of that gear are going to update it in the near future. On the other hand newer technology is available that would put us at a competitive advantage if implemented now.

4.5.1.4.  Stakeholders

Among the stakeholders for this system are the Engineering Department, the Maintenance Department, and the Central Computer Facility. The full list of these stakeholders and the specific individuals representing them are.

  • Engineering. Bunny, Bugs

  • Maintenance. Hardy, Oliver

  • Computer Facility. Laurel, Stanley

  • Chief Executive Officer. Hun, Atilla The

  • Marketing. Harry, Oil Can

4.5.1.5.  Key Features

Cash deposit, cash withdrawal, and account inquiries by customers. Customers include people who have accounts at the owning bank as well as people who wish to make withdrawals from accounts in other banks or from credit card accounts.

Maintenance of the equipment by the bank's engineers. This action may be initiated by the engineer on a routine basis. It may also be initiated by the equipment that can call the engineer when it detects an internal fault.

Unloading of deposits and loading of cash by officials of the local bank branch. These actions occur either on a scheduled basis or when the central computer determines that the cash supply is low or the deposit receptacle is liable to be getting full.

An audit trail for all activities will be maintained and sent periodically to the bank's central computer. It will be possible for the maintenance engineer to save a copy of the audit trail to a diskette for transporting to the central computer.

Both dialup and leased line support will be provided. The ATM will continue to provide services to customers when communication with the central computer is not available.

4.5.1.6.  Constraints

The project must be completed within nine months. It must cost no more than 1,750,000 USD excluding production costs. Components may be contracted out, but the basic architecture as well as the infrastructure will be designed in house. Close liaison must be maintained between the software development and the design, development and production of the hardware. Neither the hardware nor the software shall be considered the independent variable, but rather they shall be considered equal.

4.5.1.7.  Appendix

The following are the actors that directly support this vision. Additional actors may be identified later that are needed to support this or that technology. They should not be added to this list unless they are deemed to directly support the vision as described in this document.

  • Central Computer

  • Customer

  • Local Branch Official

  • Maintenance Engineer

The following are the use cases that directly support this vision. Additional use cases may be identified later that are needed to support this or that technology or to support the use cases listed here. They should not be added to this list unless they are deemed to directly support the vision as described in this document.

  • Audit

  • Customer Uses Machine

  • Maintain Machine

4.5.2.  Identifying Actors and Use Cases

For the ATM case study, we will elaborate on the examples in Section 4.3, “ Output of the Requirements Capture Process , Figure 4.4, “ Use case diagram for an ATM system showing include relationships. and Figure 4.5, “ Use case diagram for an ATM system showing an extend relationship. , and progress to identify additional actors and use cases that comprise our model of the ATM system. Figure 4.4, “ Use case diagram for an ATM system showing include relationships. and Figure 4.5, “ Use case diagram for an ATM system showing an extend relationship. exemplified the essential concepts and components of a use case diagram such as, use cases, actors, multiplicity, and include / extend relationships. They showed the relationships between the actors and use cases, and demonstrated how these actors and use cases interact.

In Figure 4.4, “ Use case diagram for an ATM system showing include relationships. we see a use case diagram for an ATM system consisting of «include» relationships for the use cases, Maintain ATM and Use ATM. Maintain ATM was further defined by two use cases, "Maintain Equipment" and "Reload ATM". Use ATM was further defined in terms of the behavior of three simpler use cases: "Deposit Cash", "Withdraw Cash" and "Query Account".

More to be written...

4.5.3.  Associations (To be written)

To be written...

4.5.4.  Advanced Diagram Features (To be written)

To be written...

4.5.5.  Use Case Specifications (To be written)

To be written...

4.5.6.  Supplementary Requirements Specification (To be written)

To be written...

Chapter 5.  Analysis

Analysis is the process of taking the “customer” requirements and re-casting them in the language of, and from the perspective of, a putative solution.

We are not actually trying the flesh out the detailed solution at this stage. That occurs in the Design Phase (see Chapter 6, Design ).

Unlike the boundary between Requirements and Analysis Phases, the boundary between Analysis and Design Phases is inherently blurred. The key is that analysis should define the solution no further than is necessary to specify the requirements in the language of the solution. The model elements in Analysis generally represent a high level of abstraction.

Once again the recursive, and iterative nature of our process means we will come back to the Analysis phase many times in the future.

5.1.  The Analysis Process

There are three schools of thought on how Analysis should be approached. The ontologist defines the data (actually the metadata) first and worries about processes later. The true ontologist would prefer not to have to think about processes at all. The phenomenonologist reverses this and favors process over data. The panparadigmist considers both process and data to be equally important and addresses both from the start.

When it comes to being a purist the ontologist has the upper hand. It is possible to define and build a database into which data can be entered and retrieved without concern for what happens to it or is done with it. On the other hand implementing a process without having any data structures for it to operate on is not very meaningful.

5.1.1.  Class, Responsibilities, and Collaborators (CRC) Cards

The CRC methodology favors the phenomenonologists preference for analysis. It is the equivalent of starting with the use cases, the process aspects (operations) of the class diagrams, and scenarios from which sequence diagrams can be initiated.

CRC cards and the associated methodology are described in detail in Appendix G, The CRC Card Methodology . They are used again in the design phase and are further discussed in Chapter 6, Design .

The strength of CRC cards during analysis.

  • Common Project Vocabulary -

  • Spread Domain Knowledge -

  • Making the Paradigm Shift -

  • Live Prototyping -

  • Identifying Holes in Requirements -

In this phase the group should consist of two or three domain experts, one object-oriented technology facilitator, and the rest of the group made up of people who are responsible for delivering the system.

The first time that the Analysis phase occurs a special case of the CRC session happens as there are no classes or scenarios to choose from to define a CRC session. At this point a special type of session known as brainstorming is held. During this session you identify the initial set of classes in the problem domain by using the problem statement or requirements document or whatever you know about the desired result for a starting point. The nouns that are found in whatever you are starting from are a good key to an initial set of classes in the system. In a brainstorming session there should be little or no discussion of the ideas. Record them and filter the results after the brainstorming. At this stage the distinction between class and object is blurred.

Once a reasonable set of classes has been defined by the group, responsibilities can be added. Add responsibilities that are obvious from the requirements or the name of the class. You don't need to find them all (or any for that matter). The scenarios will make them more obvious. The advantage of finding some in the beginning is that it helps provide a starting place.

Select the initial scenarios from the requirements document by examining it's verbs in much the same way that we scanned its nouns earlier. Then as many walk through sessions as necessary to complete the analysis phase are performed.

When is enough of the analysis complete that design can begin? When all the different responsibilities are in place and the system has become stable. After all the normal behavior has been covered, exceptional behavior needs to be simulated. When you notice that the responsibilities are all in place to support the new scenarios, and there is little change to the cards, this is a sign the you are ready to start design.

5.1.2.  Concept Diagram (To be written)

To be written...

5.1.3.  System Sequence Diagram (To be written)

To be written...

5.1.4.  System Statechart Diagram (To be written)

To be written...

5.1.5.  Realization Use Case Diagram (To be written)

To be written...

5.1.6.  Documents (To be written)

Use Case Specifications and Supplementary Requirements Specifications recast in solution language. To be written...

5.2.  Class Diagrams (To be written)

To be written...

5.2.1.  The Class Diagram (To be written)

To be written...

5.2.2.  Advanced Class Diagrams (To be written)

To be written...

5.2.2.1.  Association Classes (To be written)

To be written...

5.3.  Creating Class Diagrams in ArgoUML

5.3.1.  Classes

Identifying class diagrams from existing materials (Vision, Use Cases etc). To be written...

5.3.1.1.  Using the Note Icon in the Tool Bar

Click on your target class. Then click on the note icon. ArgoUML will generate the link automatically.

You can also right click to add a note as well! Be aware that you can add an undefined number of notes to any one class!

[Warning]Warning

Be aware that your note will not appear in the source code documentation tab.

5.3.2.  Associations (To be written)

To be written...

5.3.2.1.  Aggregation (To be written)

To be written...

5.3.3.  Class Attributes and Operations (To be written)

To be written...

5.3.3.1.  Entering Data Into Attributes and Methods Windows

Click directly in the class model element and start typing. Do not use the properties window dialog fields—they are not fully functional and liable to cause you a little frustration.

In fact, it would be interesting to see if you can type stereotypes right into the class attribute box for generating XML diagrams.

5.3.3.2.  Class Attributes (To be written)

To be written...

5.3.3.3.  Class Operations (To be written)

To be written...

5.3.4.  Advanced Class Features (To be written)

5.3.4.1.  Association Classes (To be written)

To be written...

5.3.4.2.  Stereotypes (To be written)

To be written...

5.4.  Sequence Diagrams (To be written)

To be written...

5.4.1.  The Sequence Diagram (To be written)

To be written...

5.4.2.  Identifying Actions (To be written)

To be written...

5.4.3.  Advanced Sequence Diagrams (To be written)

To be written...

5.5.  Creating Sequence Diagrams in ArgoUML

5.5.1.  Sequence Diagrams

5.5.1.1.  Creating a Sequence Diagram

Normally, you can just start a sequence diagram right away. On the Create Diagram menu choose Sequence.

5.5.2.  Actions (To be written)

To be written...

5.5.3.  Advanced Sequence Diagrams (To be written) Erweiterte Sequenzdiagramme (Noch zu beschreiben)

To be written...

5.6.  Statechart Diagrams (To be written)

To be written...

5.6.1.  The Statechart Diagram (To be written)

Types of statechart diagram (Moore, Mealy); Hierarchical diagrams. To be written...

5.6.2.  Advanced Statechart Diagrams (To be written)

To be written...

5.6.2.1.  Hierarchical Statechart Diagrams (To be written)

To be written...

5.7.  Creating Statechart Diagrams in ArgoUML

5.7.1.  Statechart Diagrams (To be written)

To be written...

5.7.1.1.  Creating a Statechart Diagram

Select a class, then you can create a statechart diagram.

5.7.2.  States (To be written)

To be written...

5.7.2.1.  Editing a Composite State

When editing a composite state, how do you provide do and event for a composite state?

The answer is to select a class, then you can create a statechart diagram.

5.7.3.  Transitions (To be written)

To be written...

5.7.4.  Actions (To be written)

To be written...

5.7.5.  Advanced Statechart Diagrams (To be written)

To be written...

5.7.5.1.  Hierarchical Statechart Diagrams (To be written)

To be written...

5.8.  Realization Use Cases (To be written)

To be written...

5.9.  Creating Realization Use Cases in ArgoUML (To be written)

To be written...

5.10.  Case Study (To be written)

Regardless of which methodology you use, at this time you are undoubtedly going to take the problem statement from Section 4.5, “ Case Study and extract the nouns from it. This list should be compacted to contain only those nouns that are expected to result in a class. This effort results in the following.

  • Account

  • Audit trail

  • Bank

  • Cash

  • Customer

5.10.1.  CRC Cards

The project manager convenes a CRC session at which the initial set of classes are to be defined. The facilitator reminds the participants that we are in the analysis phase and are only interested in what needs to be done (at the business level) and are to leave out anything that smacks of how to do it. As a general rule of thumb this means a subset of the nouns from the problem statement (see above). The group starts with a complete list of all of the nouns in the statement, examines each one, and decides which are inappropriate crossing them off the list. Each class is then assigned to one of the participants.

to be continued...

5.10.2.  Concept Class Diagrams (To be written)

To be written...

5.10.2.1.  Identifying classes (To be written)

To be written...

5.10.2.2.  Identifying associations (To be written)

To be written...

5.10.3.  System Sequence Diagrams (To be written)

To be written...

5.10.3.1.  Identifying actions (To be written)

To be written...

5.10.4.  System Statechart Diagrams (To be written)

To be written...

5.10.5.  Realization Use Cases (To be written)

To be written...

Chapter 6.  Design

We now have the problem we are trying to solve specified in the language of a putative solution. In the Design Phase, we construct all the details of that solution.

The blurred boundary between Analysis and Design is reflected in their use of many of the same UML tools. In this chapter we will mostly be reusing UML technology we have already met once. The big step is casting everything into concrete terms. We move from the abstract concepts of analysis to their concrete realization.

Once again the recursive, and iterative nature of our process means we will come back to the Design phase many times in the future.

6.1.  The Design Process (To be written)

The design process extends the modeling effort beyond the business concerns and into the solutions space. It is during this effort that you decide whether you are going to use Java, C++, J2EE, CORBA, SOAP, Dial up line, internet connection dedicated line, XML, etc. Many of these decisions will impact directly the PSM model, others may only be reflected in the documents produced.

...

6.1.1.  Class, Responsibilities, and Collaborators (CRC) Cards

Strength of CRC cards during Design

  • Spreading Objet-Oriented Design Expertise

  • Design Reviews

  • Framework for Implementation

  • Informal Notation

  • Choice of supporting software components

  • Performance Requirements

In this phase developers replace some of the domain experts in the group, but there should always be at least one domain expert in the group.

The focus of the group moves from what is to be done to how to do it. The classes from the solution domain are added to those defined in the analysis phase. Think about what classes are needed to make the system work. Do you need a List class to hold objects? Do you need classes to handle exceptions? Do you need wrapper classes for other subsystems? New classes that are looked for in this part, are classes that support the implementation of the system.

During the design phase the distinction between class and object becomes important. Think about the objects in your scenarios. Who creates the objects? What happens when it is created and destroyed? What is the lifetime of the object vs. the lifetime of the information held be the object?

Now is the time to look at what information the objects hold compared to what is requested from other classes or computed on the fly. Use the back of the card to record the attributes found for the classes. Break you responsibilities into subresponsibilities and list the subresponsibilities indented under the main responsibilities. Move the collaborators next to the subresponsibilities that use them.

After the Collaborator class on your card list the responsibility of the used class that is used in the collaboration. After the collaborating responsibilities on your cards, list the data passed back by the collaborating object in parenthesis.

Redo the scenarios you did in the analysis phase, but know take into consideration all of the design heuristics discussed. Make up your own scenarios and try them.

6.1.2.  Package Diagram (To be written)

To be written...

6.1.3.  Realization Class Diagrams (To be written)

To be written...

6.1.4.  Sequence Diagrams and Collaboration Diagrams (To be written)

To be written...

6.1.5.  Statechart Diagrams and Activity Diagrams (To be written)

To be written...

6.1.6.  Deployment Diagram (To be written)

To be written...

6.1.7.  Documents (To be written)

System Architecture. To be written...

6.2.  Package Diagrams (To be written)

To be written...

6.2.1.  The Package Diagram (To be written)

To be written...

6.2.2.  Advanced Package Diagrams (To be written)

To be written...

6.2.2.1.  Subpackages (To be written)

To be written...

6.2.2.2.  Adding DataTypes (To be written)

To be written...

6.2.2.3.  Adding Stereotypes (To be written)

To be written...

6.3.  Creating Package Diagrams in ArgoUML

6.3.1.  Packages

How to work out what goes in packages. To be written...

6.3.1.1.  Subpackages (To be written)

To be written...

6.3.2.  Relationships between packages (To be written)

To be written...

6.3.2.1.  Dependency (To be written)

To be written...

6.3.2.2.  Generalization (To be written)

To be written...

6.3.2.3.  Realization and Abstraction (To be written)

To be written...

6.3.3.  Advanced Package Features (To be written)

To be written...

6.3.3.1.  Creating New Datatypes (To be written)

To be written...

6.3.3.2.  Creating New Stereotypes (To be written)

To be written...

6.4.  More on Class Diagrams (To be written)

To be written...

6.4.1.  The Class Diagram (To be written)

To be written...

6.4.1.1.  Class Attributes (To be written)

To be written...

6.4.1.2.  Class Operations (To be written)

To be written...

6.4.2.  Advanced Class Diagrams (To be written)

To be written...

6.4.2.1.  Realization and Abstraction (To be written)

To be written...

6.5.  More on Class Diagrams in ArgoUML (To be written)

6.5.1.  Classes (To be written)

More on identifying classes from existing materials and use of stereotypes. To be written...

6.5.2.  Class Attributes and Operations (To be written)

To be written...

6.5.2.1.  Class Attributes (To be written)

To be written...

6.5.2.2.  Class Operations (To be written)

To be written...

6.5.3.  Advanced Class Features

6.5.3.1.  Operations on Interfaces

6.5.3.1.1.  Interfaces that extend interfaces

Add a unnamed interface to the current classdiagram by single-clicking on the interface icon in the tool bar and then clicking at the diagram pane (see Figure 6.1, “ Selecting the Interface tool).

Figure 6.1.  Selecting the Interface tool

Selecting the Interface tool

Then double click on the interfaces name field to change it's name as shown in Figure 6.2, “ Interface model element on the Class Diagram

Figure 6.2.  Interface model element on the Class Diagram

Interface model element on the Class Diagram

and type a name for it (like TestInterface in this case). Press “Enter” when the name is complete. (You could also enter the name by going to the Properties Tab in the Details Pane after adding the interface.)

Add another interface with a different name by repeating the last 2 steps. Then single-click on the Generalization icon in the tool bar as shown in Figure 6.3, “ Generalization on the Class Diagram tool bar .

Figure 6.3.  Generalization on the Class Diagram tool bar

Generalization on the Class Diagram tool bar

Move the mouse pointer to the subinterface, press the left mouse button and drag the generalization to the superinterface, where you release the mouse button. The screenshot of Figure 6.4, “ Generalization between two Interfaces. shows how your diagram should look now.

Figure 6.4.  Generalization between two Interfaces.

Generalization between two Interfaces.

By clicking on the subinterface and the source tab properties pane, and then selecting Java Notation for the source tab, you can see that the interface actually extends it's superinterface.

6.5.3.2.  Stereotypes (To be written)

To be written...

6.6.  Sequence and Collaboration Diagrams (To be written)

[Note]Note

Sequence diagrams does not work in ArgoUML version 0.14.

To be written...

6.6.1.  More on the Sequence Diagram (To be written)

To be written...

6.6.2.  The Collaboration Diagram (To be written)

To be written...

6.6.2.1.  Messages (To be written)

To be written...

6.6.2.2.  Actions (To be written)

To be written...

6.6.3.  Advanced Collaboration Diagrams (To be written)

To be written...

6.7.  Creating Collaboration Diagrams in ArgoUML (To be written)

6.7.1.  Collaboration Diagrams (To be written)

To be written...

6.7.2.  Messages (To be written)

To be written...

6.7.2.1.  Actions (To be written)

To be written...

6.7.3.  Advanced Collaboration Diagrams (To be written)

To be written...

6.8.  Statechart Diagrams (To be written)

To be written...

6.8.1.  The Statechart Diagram (To be written)

More on this. To be written...

6.8.2.  Advanced Statechart Diagrams (To be written)

To be written...

6.8.2.1.  Actions (To be written)

To be written...

6.8.2.2.  Transitions (To be written)

To be written...

6.8.2.2.1.  Triggers (To be written)

To be written...

6.8.2.2.2.  Guards (To be written)

To be written...

6.8.2.2.3.  Effectss (To be written)

To be written...

6.8.2.3.  Pseudo States (To be written)

To be written...

6.8.2.3.1.  Junction and Choice (To be written)

To be written...

6.8.2.3.2.  Fork and Join (To be written)

To be written...

6.8.2.4.  Hierarchical State Machines (To be written)

To be written...

6.8.2.5.  Models for State History (To be written)

Shallow v Deep. To be written...

6.9.  Creating Statechart Diagrams in ArgoUML (To be written)

6.9.1.  Statechart Diagrams (To be written)

To be written...

6.9.2.  States (To be written)

To be written...

6.9.3.  Transitions (To be written)

To be written...

6.9.4.  Actions (To be written)

To be written...

6.9.5.  Advanced Statechart Diagrams (To be written)

To be written...

6.9.5.1.  Transitions (To be written)

To be written...

6.9.5.1.1.  Triggers (To be written)

To be written...

6.9.5.1.2.  Guards (To be written)

To be written...

6.9.5.1.3.  Effectss (To be written)

To be written...

6.9.5.2.  Pseudo States (To be written)

To be written...

6.9.5.2.1.  Junction and Choice (To be written)

To be written...

6.9.5.2.2.  Fork and Join (To be written)

To be written...

6.9.5.3.  Hierarchical State Machines (To be written)

To be written...

6.9.5.4.  History (To be written)

Shallow v Deep. To be written...

6.10.  Activity Diagrams (To be written)

To be written...

6.10.1.  The Activity Diagram (To be written)

More on this. To be written...

6.10.1.1.  Action States (To be written)

To be written...

6.11.  Creating Activity Diagrams in ArgoUML (To be written)

6.11.1.  Activity Diagrams (To be written)

To be written...

6.11.1.1.  Creating an Activity Diagram

Select a use case or class, then you can create an activity diagram.

6.11.2.  Action States (To be written)

To be written...

6.12.  Deployment Diagrams (To be written)

To be written...

6.12.1.  The Deployment Diagram (To be written)

To be written...

6.13.  Creating Deployment Diagrams in ArgoUML (To be written)

6.13.1.  Nodes (To be written)

To be written...

6.13.1.1.  Node Instances (To be written)

To be written...

6.13.2.  Components (To be written)

To be written...

6.13.2.1.  Component Instances (To be written)

To be written...

6.13.3.  Relationships between nodes and components (To be written)

To be written...

6.13.3.1.  Dependency (To be written)

To be written...

6.13.3.2.  Associations (To be written)

To be written...

6.13.3.3.  Links (To be written)

To be written...

6.14.  System Architecture (To be written)

To be written...

6.15.  Case Study (To be written)

6.15.1.  CRC Cards (To be written)

To be written...

6.15.2.  Packages (To be written)

To be written...

6.15.2.1.  Identifying Packages (To be written)

To be written...

6.15.2.2.  Datatypes and Stereotypes (To be written) Datentypen und Stereotypen (Noch zu beschreiben)

To be written...

6.15.3.  Class Diagrams (To be written)

To be written...

6.15.3.1.  Identifying classes (To be written)

To be written...

6.15.3.2.  Identifying associations (To be written)

To be written...

6.15.3.3.  Specifying Attributes and Operations (To be written)

To be written...

6.15.4.  Sequence Diagrams (To be written)

To be written...

6.15.4.1.  Identifying actions (To be written)

To be written...

6.15.5.  Collaboration Diagrams (To be written)

To be written...

6.15.5.1.  Identifying Messages (To be written)

To be written...

6.15.6.  Statechart Diagrams (To be written)

To be written...

6.15.7.  Activity Diagrams (To be written)

To be written...

6.15.8.  The Deployment Diagram (To be written)

To be written...

6.15.9.  The System Architecture (To be written)

To be written...

Chapter 7.  Code Generation, Reverse Engineering, and Round Trip Engineering

7.1.  Introduction

We now have our design fully specified. With the right simulator we could actually execute the design and see if it works. (ArgoUML does not provide such functionality, but this functionality has been provided in alternative tools.)

ArgoUML does allow you to generate code from the design in several different programming languages. We, most likely, already in the design had a programming language in mind because some of the design considerations are to care for a specific language.

The output of this process is the set of files that constitute the program that solves the problem.

Once again the recursive, and iterative nature of our process means we will come back to the Build phase many times in the future.

There is also another side to this and that is the reverse engineering side. If we happen to have an old program that we would like to examine then we could take the files and reverse engineer them to create a design. This can be used when trying to understand some not so well documented program or as a quick start for the design work.

The process of going back and forth between doing changes in the design followed by a code generation and then doing changes in the code followed by a reverse engineering using for every change, the best possible perspective, is called Round-trip Engineering.

7.2.  Code Generation

The output of the Code Generation is the completed program. Depending on the contents of the design, we could also generate Unit test cases.

To do the work we need the design model, containing both static and dynamic descriptions of the program.

7.2.1.  Generating Code from the Static Structure

It is rather straightforward to do this generation, at least as long as we do it for an object-oriented language. This is some of the basic rules:

  • A class will become a class.

    In some target languages (like java, c++) they also become files and compilation units.

  • A generalization will become an inheritance.

    If the target language does not support inheritance and we didn't address this during the design, some special conversions are required to solve this.

  • An attribute will become a member variable.

  • A navigable association will become a member variable.

    Depending on the target language, target platform, and the association multiplicities this will be a pointer, a reference, a collection class, an entry in some table or map.

  • A non-abstract operation in a class will become a method.

  • An abstract operation in a class will become an abstract method.

  • An in parameter in an operation will become a parameter in the method.

    For simple types (int, boolean), this is the normal case. For C++, these will probably const classes. For Java, this cannot be enforced for classes.

  • An out or in/out parameter in an operation will become a referenced parameter in the method.

    For C++, these will be referenced non-const parameters. For Java classes, this is the default. Simple types (int, boolean) must, in java, be converted to an object of a corresponding class (Integer, Boolean).

  • The visibilities of the attributes, associations, and operations will become visibilities on the member variables or methods.

  • Packages will become directories, namespaces, or both.

7.2.2.  Code aus Interaktionen und Zustandsautomaten generieren

This conversion is not as straight-forward as the conversion of the static structure. It is much more depending on the target language and target platform.

In general it is only possible to say the following for interactions:

  • A message is converted into a function call.

    The class of the recipient will have to have a function with the correct name and signature.

    The sender function in the class of the sender will have a call to the function in the recipient.

  • An asynchronous message is converted to either posting a message to be handled by some other thread or a function call to a function that starts a new thread.

The following describes one possible way to generate state machines:

  • A State Machine is generated to a set of member variables that each method in this class refer to when deciding behavior.

  • A State is generated to a closed set of combination of values on these member variables.

  • An Event is generated as a call to a member method that can change the state.

    These methods would then typically have one big switch statement splitting on the current state.

  • A Guard is generated to an if statement in the event member method in the branch for the correct state.

  • A Transition is generated as an assignment of some state variable.

  • An Action is generated as a function call.

7.3.  Code Generation in ArgoUML

7.3.1.  Static Structure

Most of the generation can be done automatically by the provided language modules. Files are generated in a directory hierarchy that need to be filled in by the actual code.

7.3.2.  Interactions and statechart diagrams

There is currently no support for this in ArgoUML, not for any language.

7.4.  Reverse Engineering

Reverse Engineering is used for two main purposes:

  1. To get previously developed classed into the model to build upon.

  2. To get a UML view of previously developed classes to understand how they work.

Essentially this does the opposite of Code Generation.

7.5.  Round-Trip Engineering

Round-Trip Engineering makes it possible to switch perspective while doing the design. Create some classes in a class diagram. Write some code for some of the operations or functions using your favorite editor. Move the operations from one class to another in the class diagram...

ArgoUML currently does not support this for any language.

Part 2.  User Interface Reference

Chapter 8.  Introduction

This chapter describes the overall behavior of the user interface. Description of the various component parts—the menu bar, panes and various diagrams— is in separate chapters.

8.1.  Overview of the Window

Figure 8.1, “ Overview of the ArgoUML window shows the main ArgoUML window.

The titlebar of the window shows the following 4 parts of information, separated from each other by a dash.

  • The current filename. If no filename for the project is set yet, then the titlebar shows "Unititled".

  • The name of the currently active diagram.

  • The name “ArgoUML”.

  • An asterisk (*). This item is only present if the current project file is “dirty”, i.e. it is altered, but not yet saved. In other words, if the asterisk is absent, then the current file has not been altered.

Figure 8.1.  Overview of the ArgoUML window

Overview of the ArgoUML window

At the top of screen is a menu bar, which is described in Chapter 10, The Menu bar . Below that is the toolbar, as described in Chapter 9, The Toolbar .

The bulk of the window comprises four sub-windows or panes. Clockwise from top left these are the explorer (see Chapter 11, The Explorer ), editing pane (see Chapter 12, The Editing Pane ), details pane (see Chapter 13, The Details Pane ) and to-do pane (see Chapter 14, The To-Do Pane ). All 4 panes have a tool bar at the top (in the details pane it is located under the properties tab). An overview of the panes is given in Section 8.3, “ General Information About Panes. Finally at the bottom of the window is a status bar described in Section 8.4, “ The status bar.

8.2.  General Mouse Behavior in ArgoUML

Mouse behavior that is specific to the various panes of ArgoUML (see Section 8.3, “ General Information About Panes) or the menu bar, is discussed in the chapters covering those panes and the menu bar. In this section we cover behavior that is general across all of ArgoUML.

In a number of places in ArgoUML text may be directly edited (for example the constraint editor—see Section 13.7.1, “ The Constraint Editor). The behavior of the mouse when handling text is discussed in the sections that follow.

8.2.1.  Mouse Button Terminology

ArgoUML assumes a two button mouse. We will refer to the buttons as “button 1” and “button 2”. Button 1 is the leftmost button on a right-handed mouse, and sometimes referred to as the select button. Button 2 is the rightmost button on a right-handed mouse, and is sometimes referred to as the adjust button.

A single depress and release of a mouse button with the mouse is referred to as a click. Two clicks in quick succession is referred to as a double click. Moving the mouse while holding a button down is referred to as button motion with the starting point being at button down and the end point at button up.

8.2.2.  Button 1 Click

Clicking on an user-interface object or on a diagram model element may establish many different things. Most of the behaviour is experienced quite intuitive by the user, mainly because the high degree of standardisation, even spanning different computer platforms (Macintosh, PC, UNIX,...). ArgoUML follows the Java Look and Feel Design Guidelines by Sun. See http://java.sun.com/products/jlf/. Hence, behaviour of common user-interface components is generally not discussed in this document.

On the other hand, mouse actions in a diagram may not seem so intuitive to the user, since it is specific for ArgoUML. Hence they are explained here. In short, clicking selects or activates the object beneath the mouse-pointer, and moves the focus (i.e. navigation).

More in detail, the button 1 click may cause the following result:

8.2.2.1.  Selection

Here button 1 is used to choose (select) a model element (in a list or tree or on a diagram) on which subsequent operations will take place. Multiple model elements may be selected by using Shift and/or Ctrl in combination with button 1, see Section 8.2.5, “ Shift and Ctrl modifiers with Button 1. Selection is always clearly indicated by a colored background.

On a diagram, the selected model element is indicated with colored "blocks" at the corners/ends of the object. Model elements can be selected or deselected in different ways:

  • Button 1 click. Deselects all model elements, and selects the one clicked on.

  • Button 1 motion. Button motion (moving the mouse with the button down) in the diagram, not on any model element, allows to draw a rectangle around model elements which will be selected when the button 1 is released.

  • Menu functions and shortcuts. Many menu operations change selection as side-effect, e.g. creating a new diagram. Many keyboard shortcuts for menu operations change the selection, e.g. Ctrl-A, which stands for the Select All function.

8.2.2.2.  Activation

Here button 1 is used to activate the user interface component, e.g. a button. The object is usually highlighted when the mouse button is pressed and then activated when the mouse button is released. Activating an user-interface object means that its function is executed.

8.2.2.3.  Navigation

Here button 1 is used to move the focus from one user interface component or diagram model element to another. It is better known under the term keyboard focus. This because keyboard commands usually work on the model element that has the focus. The focus is indicated by a (hardly visible) box around the model element, or for a text entry box, by a flashing cursor.

8.2.2.4.  General Behavior When Editing Text

Here button 1 is used to select the point within the text at which operations (text entry and deletion) will take place.

8.2.3.  Button 1 Double Click

The behavior of button 1 double click varies betweens panes and is discussed in their chapters.

8.2.3.1.  General Behavior When Editing Text

Here button 1 double click is used to select a complete word, or other syntactic unit within the text. Subsequent operations (text entry and deletion) will replace the selected text.

8.2.4.  Button 1 Motion

8.2.4.1.  General Behavior When Editing Text

Here button 1 motion is used to select a range of text. Subsequent operations (text entry and deletion) will replace the selected text.

8.2.5.  Shift and Ctrl modifiers with Button 1

8.2.5.1.  Within Lists

This behavior applies where there is a list of things that may be selected. This includes various dialog boxes, and the to-do pane, where there is a list of to-do items to be selected.

Where selections are to be made, the SHIFT key is used to with button 1 to extend from the original button 1 selection to the current position.

Similarly the CTRL key with button 1 is used to add individual items to the current selection. Where Ctrl-button 1 is used on an item already selected, that item is removed from the selection.

[Caution]Caution

Users of Microsoft Windows might be familiar with the use of SHIFT-CTRL-Click (i.e. holding both the Shift and Ctrl key down when clicking), to add sub-lists to an existing selection. ArgoUML does not support this. SHIFT-CTRL-Click will behave as CTRL-Click.

8.2.5.2.  General Behavior When Editing Text

In a number of places in ArgoUML text may be directly edited (for example when naming a model—element in the properties pane, or when typing a UML note / comment). Here SHIFT button 1 is used to select a range of text from the previously selected point. Subsequent operations (text entry and deletion) will replace the selected text.

8.2.6.  Alt with Button 1: Panning

When holding down the Alt key during button 1 down on a diagram, movement of the mouse pans the drawing area. The function is indicated by the mousepointer which turns into a crosshair with arrows.

8.2.7.  Ctrl with Button 1: Constrained Drag

When holding down the Ctrl key while dragging with mouse button 1 down on a diagram, the movement of the dragged element element will be constrained to one of eight cardinal directions : North, South, East, West, NE, SE, SW, NW.

8.2.8.  Button 2 Actions

Button 2 actions are all dependent on the pane or menu bar, and discussed in their various chapters.

8.2.9.  Button 2 Double Click

Button 2 actions are all dependent on the pane or menu bar, and discussed in their various chapters.

8.2.10.  Button 2 Motion

Button 2 actions are all dependent on the pane or menu bar, and discussed in their various chapters.

8.3.  General Information About Panes

The four sub-windows of the main ArgoUML window are called panes. Clockwise from top left these are the explorer (see Chapter 11, The Explorer ), editing pane (see Chapter 12, The Editing Pane ), details pane (see Chapter 13, The Details Pane ) and to-do pane (see Chapter 14, The To-Do Pane ). At the top the editing pane is a tool bar.

8.3.1.  Re-sizing Panes

You can re-size panes by dragging on the divider bars between them. To indicate this possibility, the mouse cursor changes shape when hovering over the divider bars.

In addition you will see there are two small left pointing arrows within the vertical divider bars, one at the top of the vertical divider bar between explorer and editing pane and one at the top of the vertical divider bar between to-do pane and details pane. Button 1 click on the first of these will expand the editing pane to the full width of the window, button 1 click on the second will expand the details pane to the full width of the window.

There is also a small downward pointing arrow within the horizontal divider bar at its leftmost end. Clicking on this will expand the explorer and editing panes to the full depth of the window.

By using both the top arrow on the vertical divider and the arrow on the horizontal divider, it is possible to expand the editing pane to use the entire window.

The original configuration can be restored by clicking again on these arrows, which are now located at the edge of the window.

8.4.  The status bar

The status bar is at the very bottom of the ArgoUML window and is used to display short advisory messages. In general such messages are self explanatory. It is e.g. used for displaying parsing error messages in case a text entered on the diagram can not be interpreted.

Chapter 9.  The Toolbar

9.1.  File operations

These buttons have identical functions as their counterparts in the File menu.

9.2.  Edit operations

These buttons have identical functions as their counterparts in the Edit menu.

9.3.  View operations

The Find... button has identical behaviour as its counterpart in the View menu. The Zoom button is a more luxurously version of the function in the View menu.

  • Find... See for a full description Section 10.5.2, “ Find... .

  • Zoom This is a different version of the menu-item for zooming, as described in Section 10.5.3, “ Zoom . Clicking with button 1 on the zoom-icon opens a panel as in the figure below.

    Figure 9.1.  The Zoom slider on the Toolbar

    The Zoom slider on the Toolbar

    Once the panel is open, the following actions are possible:

    • Clicking with button 1 on the "knob" followed by button 1 movement will adjust the zoomfactor.

    • Clicking with button 1 on the shown percentage allows editing the given zoomfactor (in percent) directly with the keyboard. Double clicking on the value shown selects the whole entry for easy overtyping.

    • Clicking with button 1 below or above the knob increases or decreass the zoom factor with 1%. Use this function to easily fine-adjust the percentage.

    • Clicking with button 1 or button 2 on the Zoom tool, or anywhere outside the slider panel closes the panel.

    • The keyboard can be used to operate the Zoom Slider as follows: When the Zoom icon in the toolbar has the focus (indicated by the thin blue box around it), then pressing the spacebar opens the zoon slider panel. Use the arrow keys to increase and decrease the percentage 1 by 1. Use Shift-Tab to set the focus to the percentage box, where you can edit the given value directly. Pressing Enter activates the changed value. When the "knob" has the focus, pressing PageUp/PageDown increases/decreases the percentage by some amount. Pressing Home sets the percentage to 300%, and End sets it to 25%.

9.4.  Create operations

These buttons have identical functions as their counterparts in the Create menu.

Chapter 10.  The Menu bar

10.1.  Introduction

An important principle behind ArgoUML is that actions should be able to be invoked in whatever way the user finds convenient. As a result many (but not all) actions that can be carried out on the menu can be carried out in other ways as well under ArgoUML.

A number of the common menu entries are also available through keyboard shortcuts.

It is also be possible to navigate the menu from the keyboard. Each level of each menu is identified by a letter (shown underlined in the menu or entry name from the moment the ALT key is pressed). This sequence of letters while holding down the ALT key selects the entry.

The following is an explanation of why the menuitems are grouped as they are.

  • The File menu contains operations that affect on the whole project/file. All the items in this menu can be explained as such.

  • The Edit menu is generally intended for editing the model or changing the content of a diagram. It also contains functions to enable editing, like e.g. selecting. This menu is not intended for diagram layout functions. Most functions here do something with the selected modelelement and diagram. The items "Configure Perspectives..." and "Settings..." are a bit different, since they adjust the way ArgoUML works - but they do not belong in the File menu, since their settings are not stored in the project.

  • The View menu is for functions that do never alter the model, nor the diagram layout, only the way you view the diagram. A good example is "zoom". Also navigational functions belong here, e.g. "Find" and "Goto Diagram...". All changes of settings in this menu apply to all diagrams (e.g. zoom).

  • The Create menu contains all possible diagrams that can be created. These functions are context dependend, since they work on the selected modelelement.

  • The Arrange menu allows layout changes in the current diagram, which is not the same as the items in the View menu. Functions here can not alter the UML model.

  • The Generation menu is for Code Generation. The functions here work either on the selected modelelements, or on the whole project.

  • The Critique menu is specific for settings related to critics, which apply for all projects.

  • The Tools menu is currently empty. If plugins are installed, then their functions appear here.

  • The Help menu contains the usual "information" and "about".

10.2.  Mouse Behavior in the Menu Bar

Behavior of the mouse in general, and the naming of the buttons is covered in the chapter on the overall user interface (see Section 8.2, “ General Mouse Behavior in ArgoUML). There is no ArgoUML specific behaviour for the menu.

10.3.  The File Menu

These are actions concerned with input and output and the overall management of projects and the ArgoUML system.

10.3.1.  New

Shortcut Ctrl-N.

This initializes a new project within ArgoUML. The project is created without a the name. It contains a (top-level) Model named untitledModel and two empty diagrams: a class diagram and a use case diagram.

[Caution]Caution

untitledModel is not a conventional model name (most processes suggest models should be build from lower case letters). ArgoUML permits you to use any case letters, but a critic will trigger to warn that this is not conventional. See Section 16.2, “ The Model for a discussion of this.

If the model has been altered (as indicated by the "*" in the titlebar of ArgoUML's window), then activating the "New" function is potentionally not the user's intention, since it will erase the changes. Hence a confirmation dialog appears to allow the user to save his work first, or cancel the operation completely.

Figure 10.1.  The confirmation dialog for New.

The confirmation dialog for New.

10.3.2.  Open Project...

Shortcut Ctrl-O.

This opens an existing project from a file. Selecting this menu option will open a file selection dialog (see Figure 10.2, “ The file selection dialog for Open Project.... ).

Figure 10.2.  The file selection dialog for Open Project....

The file selection dialog for Open Project....

The main body of the dialog is a text area with a listing of all directories and files in the currently selected directory which match the current filter (see below).

Navigating in the directory tree is possible by selecting a directory in the drop down selector at the top of this dialog. Navigating deeper in the tree may be done by double clicking button 1 on the directory shown in the main text area.

In the lower portion of the dialog is a text box labeled File name: for the name of the file to be opened. The file name may be typed directly in here, or selected from the directory listing above using button 1 click.

Beneath this is a drop down selector labeled Files of type: to specify a filter on the files to be shown in the directory listing. Only files that match the filter are listed. The available filters are listed below. The default filter is the first one, which combines all available formats.

  • ArgoUML file (*.zargo, *.uml, *.xmi, *.xml, *.zip)

  • ArgoUML compressed project file (*.zargo)

  • ArgoUML project file (*.uml)

  • XML Metadata Interchange (*.xmi)

  • XML Metadata Interchange (*.xml)

  • XMI compressed project file (*.zip)

10.3.3.  Save Project

Shortcut Ctrl-S.

This saves the project using its current file name. Use Save Project As... to save the project to a different file. If no filename is given yet (e.g. after New), then this function works exactly as Save Project As....

[Note]Note

In certain circumstances, there is nothing to save, and this menuitem is downlighted. E.g. when the user did not yet alter a loaded project. The presence of a “*” in the titlebar of ArgoUML's window indicates that the current project is “dirty” (has been altered), and can be saved.

10.3.4.  Save Project As...

This opens a dialog allowing you to save the project under a different file name (or to specify a file name for the first time if the project is a new project).

The dialog box is almost identical to that for Open Project (see Figure 10.2, “ The file selection dialog for Open Project.... ). The extension of the filename is automatically set.

10.3.5.  Revert to Saved Projekt speichern rückgängig machen

This menu-item allows you to throw away all your recent changes, and reload the last saved version of the current project. It works a bit like an Undo feature, but only restores changes done since the last time the file was saved.

This menu-item is downlighted unless the currentproject has been saved or loaded before (i.e. it has a name), and it has been altered.

When this menu-item is activated, a small confirmation dialog box opens, as shown in the figure below. This warning that all recent changes will be discarded, is needed because the action can not be undone. Selecting No cancels the whole action as if you did not select the menu-item in the first place. Selecting Yes reloads the last saved file.

Figure 10.3.  The warning dialog for Revert to Saved.

The warning dialog for Revert to Saved.

10.3.6.  Import XMI...

This menu-item allows to load an UML 1.3 or 1.4 model which was exported by e.g. another tool, as a XMI file, according the XMI V1.0, V1.1 or V1.2 standard. The extension of such file should be .xmi.

If the model has been altered (as indicated by the "*" in the titlebar of ArgoUML's window), then activating the "Import XMI..." function is potentionally not the user's intention, since it will erase the changes. Hence a confirmation dialog appears to allow the user to save his work first, or cancel the operation completely.

Figure 10.4.  The confirmation dialog for Import XMI....

The confirmation dialog for Import XMI....

When the menu is activated, the standard filechooser appears, see Figure 10.5, “ The dialog for Import XMI.... . Beware the fact that this file will only contain the model, not any diagram layout. Hence, the new project will not contain any diagrams.

Figure 10.5.  The dialog for Import XMI....

The dialog for Import XMI....

10.3.7.  Export XMI...

This menu-item allows to save the complete structure of the UML 1.4 model as a XMI file, according the XMI V1.2 standard. Beware the fact that this file will only contain the model, not any diagram layout. Hence, if the XMI file is reloaded with the File - Open Project... menu, then the diagrams are lost.

When the menu is activated, the standard filechooser appears, see Figure 10.6, “ The dialog for Export XMI.... .

Figure 10.6.  The dialog for Export XMI....

The dialog for Export XMI....

10.3.8.  Import Sources...

A very powerful feature of ArgoUML is that it can “Reverse Engineer” Java code to yield a class diagram. This sub-menu entry specifies Java code to be imported for reverse engineering.

The dialog box is similar to that for Open Project (see Figure 10.2, “ The file selection dialog for Open Project.... ), but with two extra tabs placed alongside the directory listing, as shown in Figure 10.7, “ The file selection dialog for Import Sources.... ).

Figure 10.7.  The file selection dialog for Import Sources....

The file selection dialog for Import Sources....

Those fields that are the same as Open Project behave in the same way (see Section 10.3.2, “ Open Project... ).

Next to the "All Files" file filter, there is the default filter "Java Source File (*.java)".

The first of the two tabs is labeled General and is selected by button 1 click on its tab. It provides a combo box for the language selection (in V0.18 of ArgoUML only Java can be chosen), and the following selections:

  • Descend directories recursively. If enabled (the default), reverse engineering will track through sub-directories for any further Java files. If not it will restrict to the selected directory.

  • Changed/new files only. If enabled (the default), only changed and new files are imported. If not all classes will be replaced.

  • Create diagrams from imported code. If you unselect this, then no diagrams are created, i.e. all data will only be visible in the explorer.

  • Minimise Class icons in diagrams. If enabled, then the attributes and operations compartiments will not be shown in the classes on the generated class diagrams. Note: This item is checked by default, and is overseen by many users, which are then surprised by the result.

  • Perform Automatic Diagram Layout. If selected, then ArgoUML will do its best to layout the generated diagrams automatically. If not, then all items will be placed at the top left corner of the diagram.

  • Level of import detail: Classifiers only / Classifiers plus feature specifications / Full import. The latter is the default.

  • Import source file encoding:. The value Cp1252 is often the default. This string represents the coded character set identifier (CCSID).

The second of the two tabs is labeled Java and is selected by button 1 click on its tab. It provides two pairs of radio boxes.

  • The first radio box allows selection between modeling attributes of Java classes as UML attributes (the default) or as UML associations to the class specified.

  • The second radio box allows selection between modeling arrays as new datatypes in their own right (the default) or as their base datatype with multiplicity.

10.3.9.  Page Setup...

This brings up the standard dialog box provided by the operating system to adjust printer paper size, orientation, and other options.

10.3.10.  Print...

Shortcut Ctrl-P.

This brings up the standard dialog box provided by the operating system allowing the current diagram to be printed out.

In some cases, when the printing is started, the dialog box of Figure 10.8, “ The diagram exceeds page size dialog. appears. Selecting the "Fit to page" button does print the whole diagram fitted on one page by scaling it down. Which might cause all text to be too small to read in case of big diagrams, but it is a quick and easy way to get an usable printout. Selecting the "Multiple pages" option does print unscaled, by dividing the diagram in pieces, on as many pages as needed. Pressing the close button of the dialog does the former option.

Figure 10.8.  The diagram exceeds page size dialog.

The diagram exceeds page size dialog.

[Warning]Warning

If the current diagram contains no selected model elements, then the whole diagram is printed. However, if one or more model elements are selected, then only the area they cover is printed! If scaling is selected (by the "Fit to page" choice in the dialog box descibed above), then the scaling is done on basis of the selected model elements only. If scaling is not chosen (or in case it is not needed), then all pages containing a selected model element are printed.

10.3.11.  Export Graphics... Grafik exportieren...

This menu entry brings up a dialog box allowing the currently selected diagram (in the editing pane) to be saved in one of a number of graphic formats.

The dialog box is identical to that for Open Project (see Figure 10.2, “ The file selection dialog for Open Project.... ), except for the Files of type:. The chosen filetype specifies the graphics format used for saving. The filename is automatically extended with the corresponding ending (if not entered already). A default filename is generated based on the diagram name.

The available graphics types are:

  • GIF image (*.gif)

  • Encapsulated Postscript file (*.eps)

  • PNG image (*.png)

  • Postscript file (*.ps)

  • Scalable Vector Graphics file (*.svg)

The graphics format that is selected by default is set in the dialog under the menu entry Edit - Settings...

10.3.12.  Export All Graphics...

This menu entry brings up a dialog box to select a directory. In this directory, for all diagrams in the current project, a graphics file is generated.

The names of the files are deducted from the diagram names. The graphics format that is produced is set in the dialog under the Edit menu (see Section 10.4.5, “ Einstellungen... ).

10.3.13.  Notation

This sub-menu presents a radio button selection for notation, i.e. the language in which all textual adornments are shown on the diagrams.

This feature defines the project's notation language.

There are 2 ways to set the notation language:

  • In the Edit menu, see Section 10.4.5.7, “ Notation Tab in the notation tab of the settings dialog, which defines the default notation language for new projects. This choice is stored in the argo.user.properties file in the .argouml directory under the user's home directory.

  • In the File menu, item Notation. This determines how all textual adornments of figures on all diagrams of the current project are shown. This choice is stored in the project file.

The following 2 notations are build in ArgoUML:

  • UML 1.4. Uses UML notation as the default notation for every modelelement on any diagram.

  • Java. Uses Java notation as the default notation for every modelelement on any diagram.

The following choices are only available if the corresponding plugin languages are installed.

  • Cpp.

  • CSharp.

  • PHP.

Besides UML, only Java is partly implemented in V0.22 of ArgoUML.

10.3.14.  Project Properties

This menu entry brings up a dialog box, which allows the user to set various options of the currently loaded project.

All settings in this dialog are stored in the project-file together with the model.

Figure 10.9.  The dialog for Project Properties - the User tab.

The dialog for Project Properties - the User tab.

In the User tab, you are able to set the following fields:

  • The first field contains the name of the author or responsible for the current project. By default the name and email of the creator are filled in, so probably you will never need to edit this, but it is possible.

  • The Project Description field may contain any text that you need to describe the project. By default it is empty.

  • The "Last saved with ArgoUML" field indicates the version of ArgoUML that was used to save this project (the last time it was saved). This may be usefull if multiple designers have different versions of ArgoUML, which may not be backwards compatible all the time.

Figure 10.10.  The dialog for Project Properties – the Profiles tab.

The dialog for Project Properties – the Profiles tab.

In the Profiles tab, you are able change the following settings:

  • The type of “Stereotype Visualization” for the project, which may be textual, small or with big icons.

  • The UML profiles which are configured in the project – model elements of these UML profiles may be referenced in the project.

Figure 10.11.  The dialog for Project Properties - the Notations tab.

The dialog for Project Properties - the Notations tab.

In the Notations tab, you are able to set the following fields:

  • The first field is a combobox that allows selection of the project's Notation language. By default, it lists UML and Java, but other languages may be added by plugins. See the chapter on Notation for more explanation: Section 12.11, “ Notation.

  • Use guillemots (« ») for stereotypes (clear by default). By default ArgoUML uses pairs of less than and greater than (<< >>) characters for stereotypes. If this box is checked stereotypes on diagrams are shown between true guillemots (« »).

    This feature is presumably added to ArgoUML because guillemots are poorly supported by various fonts, and if they are present, then they are quite small and poorly visible.

  • Show visibility (clear by default). If this is selected, then ArgoUML will show the visibility indicators in front of e.g. attributes in the diagram. In UML the notation is "+" for public, "-" for private, "#" for protected, and "~" for package. E.g. for an attribute, it may show: +newAttr : int.

  • Show multiplicity (clear by default). If this is selected, then ArgoUML will show the multiplicity of e.g. attributes in the diagram. In UML notation, the multiplicity is shown between [], such as: +newAttr [0..*] : int. This setting has no impact on showing multiplicity near associationends.

  • Show initial value (clear by default). If this is selected, then ArgoUML will show the initial value of e.g. attributes in the diagram. In UML notation, the initial value is shown e.g. like this: +newAttr : int = 1.

  • Show properties (clear by default). If this is selected, then ArgoUML will show various properties between braces {}. E.g. for an attribute, it may show: +newAttr : int { frozen }.

  • Show types and parameters (set by default). When this checkbox is unmarked, attributes in classes are shown without type indication, and operations are shown without parameters. This feature may be usefull during the analysis phase of your project. If all checkmarks in the Notation Tab are unchecked, then e.g. for an attribute, ArgoUML may show: newAttr. And for an operation: newOperation().

  • Show stereotypes in explorer (clear by default). If this is selected, then ArgoUML will show stereotypes next to the icons of the modelelements in the Explorer, i.e. the tree structure at the left hand side.

  • Default shadow width (set to 1 by default). ArgoUML is able to draw all elements on a diagram with a shadow, for esthetical reasons. Use this setting to adjust the size of the shadow, used when the modelelement is created. The details tab "Presentation" allows to set the shadow per modelelement, after they are created, but ArgoUML V0.22 does not retain this latter change after save and load.

Figure 10.12.  The dialog for Project Properties – the Diagram Appearance tab.

The dialog for Project Properties – the Diagram Appearance tab.

In the Diagram Appearance tab, you are able change the diagrams font.

10.3.15.  Most Recent Used Files

ArgoUML remembers a few of the most recently saved files, and lists them here, to enable loading then in the most simple way.

The maximum number of files that is listed here, can be adjusted in the Edit -> Settings... menu. The list of files is stored in the argo.user.properties file in the .argouml directory under the user's home directory.

10.3.16.  Exit

Shortcut Alt-F4.

This closes down ArgoUML. A warning message will pop-up if you have a project open with unsaved changes asking if you wish to save it. See Figure 10.13, “ The save changes dialog.. The options are:

  • Yes (save the project and exit ArgoUML);

  • No (do not save the project, but still exit ArgoUML); and

  • Cancel (do not save the project and do not exit ArgoUML).

  • The dialog box can also be closed by clicking in the close button in the window border. The effect is the same as selecting "Cancel".

Figure 10.13.  The save changes dialog.

The save changes dialog.

10.4.  The Edit Menu

This menu provides support for selecting model elements on the editing pane; removal of model elements from diagrams and the model; and control of user settings.

10.4.1.  Select

This sub-menu provides for selection of items on the editing menu. It has the following entries.

  • Select All (shortcut Ctrl-A). Selects all model elements on the current pane or in the current field. The exact behaviour depends on the current pane (i.e. the last one you clicked in): explorer pane, editing pane, to-do pane, details pane. One rule applies in all cases though: the selection on the diagram (editing pane) and in the explorer are always synchronised.

    If the editing pane is the current pane: First everything in the explorer and on the current diagram is deselected, and then everything that is on the current diagram is selected (and if the same items apear in the explorer, then they are also there indicated as selected, because they are always synchronised).

    If the explorer pane is the current pane: All visible items in the explorer pane are selected, and non-visible items are deselected.

    If the to-do pane is the current pane: All visible items in the to-do pane are selected, and non-visible items are deselected. In fact, this works the same as for the explorer pane, because both are tree structures.

    If the details pane is the current pane: The function only works when the cursor is in certain fields, where selecting is possible, e.g. a Name field. In such a case, the Select All function extends the current selection to the whole field contents.

  • Navigate Back. ArgoUML keeps a record of the model elements that you have been selecting while navigating the model. This button moves you back to the previous one selected. If there are no more previous model elements, the button is grayed out.

  • Navigate Forward. ArgoUML keeps a record of the model elements that you have been selecting while navigating the model. This button moves you forward to the next one selected (after you have used the Navigate Back button to move back). If there are no more next model elements, the button is grayed out.

  • Invert Selection. This inverts the current selection on the current pane. More exact: everything that was selected is de-selected and everything that was not selected within the current pane is selected.

10.4.2.  Remove From Diagram

Shortcut Delete.

This removes the currently selected item(s) from the diagram, but not from the model.

The modelelement can be re-added to the diagram by button 2 click on the modelelement in the explorer, or by dragging it onto the diagram.

10.4.3.  Delete From Model

Shortcut Ctrl-Delete.

This function deletes the selected item(s) from the model completely.

If the item to be deleted is also present on another diagram than the current one, the dialog box from figure x appears.

Figure 10.14.  The dialog for confirmation of Remove from Model.

The dialog for confirmation of Remove from Model.

10.4.4.  Perspektiven konfigurieren...

This menu-item invokes the same dialog as the button at the top of the explorer. See Section 11.5, “ Configuring Perspectives. for a complete description.

10.4.5.  Einstellungen...

This menu entry brings up a dialog box, which allows the user to set various options that control the behavior of ArgoUML (see Figure 10.15, “ The dialog for Settings - Preferences. ).

These settings are saved persistently for use by subsequent ArgoUML sessions.

ArgoUML has various user specific configurations that can be set in this dialog box, or directly on the various panes. Also the main window size and location is such a setting. Activating this menu entry causes the information to be saved in the file argo.user.properties. The location of this file is in the .argouml directory under the "users home directory", which is defined as ${user.home} , and can be determined as described in Section 10.4.5.2, “ Environment Tab .

[Tip]Tip

This is a text file, which you can edit to configure ArgoUML.

Figure 10.15.  The dialog for Settings - Preferences.

The dialog for Settings - Preferences.

The options that can be set up on the various tabs are described in the following sections. For each tab there are three buttons at the bottom of the dialog box.

  • OK. Activating this button (button 1 click) applies the chosen settings and exits the dialog.

  • Cancel. Selecting this button (button 1 click) exits the dialog without applying any settings changed since the last Apply (or since the dialog started if Apply has not been used).

  • Apply. Selecting this button (button 1 click) applies the chosen settings and remains in the dialog.

Closing the dialog (with the close button in the top corner in the border of the window) causes the same effect as Cancel.

10.4.5.1.  Preferences Tab

Selecting the Preferences tab (button 1 click on the tab) gives the following options as check boxes.

  • Show Splash Panel (set by default). If enabled ArgoUML will show a small panel with a picture while starting up.

    [Tip]Tip

    The splash panel can be seen by using the Help menu (see Section 10.11.2, “ About ArgoUML ).

  • Reload last saved project on startup (clear by default). Check this item if you always work on the same project, and wish to load it automatically when you start up ArgoUML.

  • Strip (non-standard) diagrams from XMI file during import (clear by default). Checking this item will tell ArgoUML to ignore the "Diagram" elements when importing XMI files.

    You only need to use this setting, if ArgoUML gives an error while importing your XMI file saying that it encountered unrecognized elements named "Diagram." Some versions of Poseidon are known to create this type of file by default although there's usually an export option to force them to create standard XMI files.

10.4.5.2.  Environment Tab

Selecting the Environment tab (button 1 click on the tab) lists several environmental items. Note that none of the paths can be altered — these are just a matter of record.

Figure 10.16.  The dialog for Settings - Environment.

The dialog for Settings - Environment.

  • Default Graphics Format. Here you can select the same graphics formats as in the menu Section 10.3.11, “ Export Graphics... Grafik exportieren.... The chosen format is selected by default in the Export Graphics and Export All Graphics menu-items.

  • Graphics Export Resolution. This allows you to artificially increase the resolution of produced graphics. The advised setting is "Standard". To be able to use "High" or "Extra High", you usually need to start the Java virtual machine with extra memory.

  • ${argo.ext.dir}. The directory holding ArgoUML extensions—by default the ext sub-directory of the ArgoUML build directory.

  • ${java.home}. The home directory of the Java Runtime Environment (JRE).

  • ${user.home}. The user's home directory. Used for storing the argo.user.properties file, under the .argouml directory.

  • ${user.dir}. The directory from which ArgoUML was started.

  • Startup Directory. The directory in which ArgoUML starts file searches etc.

10.4.5.3.  User Tab

This tab allows the user to record additional information of use to the system. There are two text boxes provided.

Figure 10.17.  The dialog for Settings - User.

The dialog for Settings - User.

  • Full Name. Allows the user to record her full name.

  • Email Address. Allows the user to record his Email address.

This information is used when requesting to-do help by Email.

10.4.5.4.  Appearance Tab

This tab allows the user to specify the LAF (Look And Feel) and theme, i.e. what the complete ArgoUML UI looks like. It comprises the following settings.

Figure 10.18.  The dialog for Settings - Appearance.

The dialog for Settings - Appearance.

  • Look and Feel. The choice made here influences the complete User Interface. It only becomes effective when ArgoUML is exited and restarted.

  • Metal Theme. This item is downlighted if the Metal LAF is not chosen. The choice made here influences the complete User Interface. It only becomes effective when ArgoUML is exited and restarted.

  • Smooth edges of diagram lines and text. This feature is known as “anti-aliasing” on certain platforms. It causes diagonal lines to look much less jagged, by making use of several shades of gray. This feature only works if the operating system supports it.

10.4.5.5.  Profiles Tab

In this tab the user may configure the ArgoUML application settings related to the profiles.

Figure 10.19.  The dialog for Settings - Profiles.

The dialog for Settings - Profiles.

  • Stereotype Visualization – select to view stereotypes as text, small or big icons.

  • Default XMI directories – allows the user to configure directories where ArgoUML can find the user defined profiles.

  • Default Profiles – select which profiles from the available profiles are configured in new projects by default.

10.4.5.6.  Configure Shortcuts Tab

(To Be Written)

Figure 10.20.  The dialog for Settings - Configure Shortcuts.

The dialog for Settings - Configure Shortcuts.

10.4.5.7.  Notation Tab

This tab allows the user to specify certain notation settings, i.e. how things are shown on diagrams. It comprises the following check boxes.

All settings here, only define the defaults used for new projects. If you want to change the way the diagrams in your current project look, then see the File - Properties menu.

Figure 10.21.  The dialog for Settings - Notations.

The dialog for Settings - Notations.

  • Notation Language ( UML 1.4 by default). This feature allows changing the default notation (i.e. language: UML, Java,...) used on the diagrams for new projects. Suppose that a designer indicates that the default notation of a project is Java. When he saves the project, the choice for Java is stored inside the project file. If someone else is viewing the diagram, he will see the Java notation, too. This person can select the UML notation in the File - Notation menu, and see all diagrams in UML language. See Section 10.3.13, “ Notation).

  • Show name of nodes in bold font.

    This feature causes the names of every node (i.e. something drawn with a closed polygon) to be bold.

    There is no semantics to showing bold names, but it may make your diagrams look nicer.

  • Use guillemots (« ») for stereotypes (clear by default). By default ArgoUML uses pairs of less than and greater than (<< >>) characters for stereotypes. If this box is checked stereotypes on diagrams are shown between true guillemots (« »).

    This feature is presumably added to ArgoUML because guillemots are poorly supported by various fonts, and if they are present, then they are quite small and poorly visible.

    Independent of the way they are shown, when entering stereotypes, you can always type real guillemots (if your keyboard supports it) or their << >> equivalents.

  • Show association names.

    This feature causes the names of every association to be hidden if not marked.

  • Show visibility (clear by default). If this is selected, then ArgoUML will show the visibility indicators in front of e.g. attributes in the diagram. In UML the notation is "+" for public, "-" for private, "#" for protected, and "~" for package. E.g. for an attribute, it may show: +newAttr : int.

  • Show multiplicity (clear by default). If this is selected, then ArgoUML will show the multiplicity of e.g. attributes in the diagram. In UML notation, the multiplicity is shown between [], such as: +newAttr [0..*] : int. This setting has no impact on showing multiplicity near associationends.

  • Show initial value (clear by default). If this is selected, then ArgoUML will show the initial value of e.g. attributes in the diagram. In UML notation, the initial value is shown e.g. like this: +newAttr : int = 1.

  • Show properties (clear by default). If this is selected, then ArgoUML will show various properties between braces {}. E.g. for an attribute, it may show: +newAttr : int { frozen }.

  • Show types and parameters (set by default). When this checkbox is unmarked, attributes in classes are shown without type indication, and operations are shown without parameters. This feature may be usefull during the analysis phase of your project. If all checkmarks in the Notation Tab are unchecked, then e.g. for an attribute, ArgoUML may show: newAttr. And for an operation: newOperation().

  • Show stereotypes in explorer (clear by default). If this is selected, then ArgoUML will show stereotypes next to the icons of the modelelements in the Explorer, i.e. the tree structure at the left hand side.

  • Show "1" multiplicities.

    This feature allows the user to select if he wants to show all multiplicities that are "1" to be shown or not..

    Some people consider not showing the multiplicity as "undefined". So, the only way to differentiate between a multiplicity of 1 and an undefined multiplicity, is to mark this checkbox.

  • Hide arrowheads for bi-directional associations..

    The UML standard defines different ways to indicate navigability of an association on diagrams. Presentation option 1 is to show all arrows (i.e. you can only navigate in a certain direction if an arrow is drawn), presentation option 2 is to show no arrows at all, and presentation option 3 is to show only an arrow if the association is unidirectional.

    Before the release of version 0.26, ArgoUML could only use presentation option 3. Currently, the user can choose between option 1 and 3. Option 2 is not supported.

    In the past, option 3 was used most often in other UML tools, but nowadays option 1 is more common.

  • Default shadow width (set to 1 by default). ArgoUML is able to draw all elements on a diagram with a shadow. Use this setting to adjust the size of the shadow, used when the modelelement is created. The details tab "Presentation" allows to set the shadow per modelelement, after they are created.

10.4.5.8.  Diagram Appearance Tab

(To Be Written)

Figure 10.22.  The dialog for Settings - Diagram Appearance.

The dialog for Settings - Diagram Appearance.

10.4.5.9.  Modules Tab

This tab shows a list of modules that are installed, which may be enabled or disabled. Since this is a new concept for ArgoUML, it currently contains a list of modules that can not be removed, and a button to test the concept. Pressing this button adds a useless menu-item on the Tools menu, nothing else.

Notice also that this is a "new" modules concept so the old Pluggable modules do not work this way, and are not listed.

10.4.5.10.  Extra Tabs added by Plugins

A plug-in module has the possibility to add extra tabs. One example is C++; it adds the following tab.

Figure 10.23.  The dialog for Settings - C++.

The dialog for Settings - C++.

10.5.  The View Menu

This menu is used for actions that affect how the various panes are viewed.

10.5.1.  Goto Diagram...

This menu entry brings up a dialog box, describing all the diagrams in the current project under ArgoUML.

Figure 10.24.  The dialog for Goto Diagram....

The dialog for Goto Diagram....

The dialog box contains a table with three columns and one row for each diagram in the current project. A scroll bar gives access if the table is too long for the box. Double button 1 click on any row will select that diagram in the editing pane. The three columns are as follows.

  • Type. Lists the type of diagram.

  • Name. Lists the name given to the diagram.

  • Description. Shows how many nodes and edges there are on the diagrams. A node is a “2-D” model element and an edge is a connector model element.

This dialog box is not modal, which allows it to remain open while editing the model for easy navigation.

[Warning]Warning

The V0.26 implementation of ArgoUML does not immediately update the dialog box with changes made to diagrams: change of name, addition of diagrams, deletion of diagrams.

10.5.2.  Find...

This menu entry brings up a non-modal dialog box for the ArgoUML search engine.

Figure 10.25.  The dialog for Find....

The dialog for Find....

The Name and Location specifies the search to be made. It contains the following:

  • A text box labeled Element Name: specifies the name of the model element to search for. Wild cards (*, ?) may be used here. A drop down gives access to find expressions previously used.

  • A text box labeled In Diagram: specifies which diagrams are to be searched. Again wild cards may be used. Both these two text boxes have a default entry of *, i.e. match anything.

  • To the right of these two text boxes, a selector labeled Element Type: allows you to specify the UML metaclass for which you are searching.

  • A selector labeled Find in: allows the search to be made over the entire project (the default) or as a sub-search over the results of a previous search. When opened, a list of all the search result tabs appears.

  • Beneath these boxes is the button Clear Tabs. This clears the display of tabs with the results from previous searches (see below). This button is downlighted if there are no tabs but the Help tab.

  • And finally, there is the button Find. This causes the search specified in the text boxes and selectors above to be executed. The results are displayed in a tab taking up the lower two thirds of the page.

The lower two thirds of the dialog comprises an initial tab (labeled Help) giving summary help, and further tabs displaying the results of searches. These search tabs are labeled with a summary of the search element in diagram and are divided horizontally in two halves.

Button 1 double clicking on these tabs removes the tab, and spawns a new window that contains the tab contents, i.e. the search results. This window can be moved and sized at will. This does not work for the help tab.

The top half is labeled Search Results: followed by a count of the number of items found. It comprises a table with one row for each model element and four columns. The width of the columns can be adjusted.

  • Type. Lists the type of model element.

  • Name. Lists the name given to the model element.

  • In Diagram. Where the model element is visible on a diagram, this lists the name of the diagram, otherwise it shows N/A.

  • Description. Contains a description of the model element. In ArgoUML V0.18 this seems to be restricted to the single entry docs.

Button 1 click on any row will give more information on that model element by showing related model elements in the bottom half (see below). Double click on any row describing a model element on a diagram and that item and diagram will be selected.

The bottom half of the tab is a table labeled Related Elements: and is a table with the same columns as the top half. When a model element has been selected in the top half, this table shows the details of any related elements.

[Tip]Tip

Enlarging the dialog vertically shows that the "Related Items" part changes in size, but not the Search results part. However, between them is a divider line and when hovering over this line, the mouse pointer changes into a sizing icon, and the border between these 2 areas can be moved up or down to redistribute the space in the window.

[Warning]Warning

This dialog box is not modal, which allows it to remain open while editing the model for easy navigation. But the V0.26 implementation of ArgoUML does not immediately update the dialog box with changes made to the found model elements: change of model element name, change of diagram name. Deletion of a diagram does not stop the possibility to navigate to it.

10.5.3.  Zoom

This entry brings up a sub-entry, which allows scaling the view of all diagrams to a factor of its normal size. This setting is not saved persistently.

The sub-menu items that can be selected are:

  • Zoom Out. Shortcut (Ctrl-Minus). Gives more overview over the drawing.

  • Zoom Reset. Returns to the default zoom ratio (i.e. 100%).

  • Zoom In. Shortcut (Ctrl-Plus). Makes the items on the drawings bigger.

10.5.4.  Adjust Grid

This allows selection of the grid representation on the screen between the following:

  • Lines 16: full grid at 16 pixel spacing.

  • Lines 8: full grid at 8 pixel spacing.

  • Dots 16: dots at 16 pixel spacing (the default).

  • Dots 32: dots at 32 pixel spacing.

  • None: no grid of any form.

10.5.5.  Adjust Snap

This allows selection of the spacing of grid snapping between the following:

  • Snap 4: snap at 4 pixel spacing.

  • Snap 8: snap at 8 pixel spacing (the default).

  • Snap 16: snap at 16 pixel spacing.

  • Snap 32: snap at 32 pixel spacing.

[Note]Note

There is no option to turn off snap to grid altogether

[Note]Note

If you wish to align existing elements to changed snap boundaries, you can use the Arrange > Align To Grid Snap menu (see Section 10.7.1, “ Align).

10.5.6.  Page Breaks

This toggles whether page breaks are shown on the diagram (as white dotted lines).

[Warning]Warning

This menu item does not work in ArgoUML V0.26.

10.5.7.  Toolbars

This menu allows the user to hide or show any of the bars at will. By default, all of them are shown.

10.5.8.  XML Dump

This activates a window that shows the complete contents of the current project in XML format.

Although very useful for debugging ArgoUML, this menu function is hardly interesting to the common user.

10.6.  The Create Menu

This menu provides for creating the various types of UML diagrams supported by ArgoUML.

10.6.1.  New Use Case Diagram

This menu entry creates a blank use case diagram, and selects that diagram in the editing pane. If a package is currently selected, then the use case diagram will be created within that package. This means that it will be shown within the package on the explorer hierarchy (under Package-centric view) and model elements created on the diagram will be created within the namespace of the package. This does not only apply to a package, but also to a class, interface, use case, etc.

[Tip]Tip

This does not prevent model elements from other namespaces/packages appearing on the diagram. They can be added from the explorer using Add to Diagram from the button 2 pop-up menu.

10.6.2.  New Class Diagram Klassendiagramm

This menu entry creates a blank class diagram, and selects that diagram in the editing pane. If a package is currently selected, the class diagram will be created within that package. This means that it will be shown within the package on the explorer hierarchy (under Package-centric view) and model elements created on the diagram will be created within the namespace of the package. This does not only apply to a package, but also to a class, interface, use case, etc.

[Tip]Tip

This does not prevent model elements from other namespaces/packages appearing on the diagram. They can be added from the explorer using Add to Diagram from the button 2 pop-up menu.

10.6.3.  New Sequence Diagram

This menu entry creates a blank sequence diagram, and selects that diagram in the editing pane. It also creates a Collaboration UML element, which is a container for the elements shown on the new diagram. If a class is currently selected, a sequence diagram and a collaboration will be created that represent the behaviour of that class. This means that the created elements will be shown within the class in the explorer hierarchy (under Package-centric view) and model elements created on the diagram will be created within the namespace of the collaboration. A sequence diagram may not only represent the behavior of a class, but also of any other classifier, such as interface, use case, etc. It is also possible to make sequence diagrams for an operation.

10.6.4.  New Collaboration Diagram

This menu entry creates a blank collaboration diagram, and selects that diagram. It also creates a Collaboration UML element, which is a container for the elements shown on the new diagram. If a package is selected when this menu item is activated, the collaboration diagram will be created within a collaboration within that package. This means that it will be shown within the collaboration within the package on the explorer hierarchy (under Package-centric view) and model elements created on the diagram will be created within the namespace of the collaboration within the package.

[Tip]Tip

This does not prevent model elements from other namespaces/packages appearing on the diagram. They can added from the explorer by dragging or by using Add to Diagram from the button 2 pop-up menu.

10.6.5.  New Statechart Diagram

This menu entry creates a blank statechart diagram associated with the currently selected class, and selects that diagram in the editing pane. It also creates a Statemachine UML element, which is a container for the elements shown on the new diagram.

Statechart diagrams are associated with a model element capable of dynamic behavior, such as a classifier or a behavioral feature, which provides the context for the state machine it represents. Suitable model elements are e.g. a class, an operation, and a use case. If such element is not selected at the time the New Statechart Diagram menu is activated, then an unattached statemachine is created. To obtain a well-formed UML model, you have to set the context of the statemachine on its details pane.

10.6.6.  New Activity Diagram

This menu entry creates a blank activity diagram associated with the currently selected class, and selects that diagram in the editing pane. It also creates a ActivityGraph UML element, which is a container for the elements shown on the new diagram.

Activity diagrams are associated with a model element capable of dynamic behavior, such as packages, classifiers (including use cases) and behavioral features. Suitable model elements are e.g. a class, a use case, an operation, and a package. If such element is not selected at the time the New Activity Diagram menu is activated, then an unattached ActivityGraph is created. To obtain a well-formed UML model, you have to set the context of the ActivityGraph on its details pane.

10.6.7.  New Deployment Diagram

This menu entry creates a blank deployment diagram, and selects that diagram in the editing pane.

[Tip]Tip

Model elements from other namespaces/packages can be added from the explorer by dragging or by using Add to Diagram from the button 2 pop-up menu.

10.7.  The Arrange Menu

This menu provides a range of functions to help in the alignment of model elements on diagrams within the editing pane. In general the menu function invoked is applied to any model element or model elements currently selected in the editing pane.

10.7.1.  Align

This sub-menu aligns the selected items. There are seven alignment options provided.

  • Align Tops. Aligns the selected model elements by their top edges.

  • Align Bottoms. Aligns the selected model elements by their bottom edges.

  • Align Rights (Shortcut Ctrl-R). Aligns the selected model elements by their right edges.

  • Align Lefts (Shortcut Ctrl-L). Aligns the selected model elements by their left edges.

  • Align Horizontal Centers. Aligns the selected model elements so their horizontal centers are in a vertical line.

  • Align Vertical Centers. Aligns the selected model elements so their vertical centers are in a horizontal line.

  • Align To Grid. Aligns the selected model elements so their top and right edges are on the grid snap boundary (see Section 10.5.5, “ Adjust Snap ) edge.

    [Tip]Tip

    The alignment is to the current grid snap setting, which may be smaller, larger or the same as the displayed grid. Since items are aligned to the grid snap boundary any way when you place them, this menu entry has no effect unless you have either changed the grid snap to a larger value or used one of the other Arrange menu entries to push items off their initial positions.

10.7.2.  Distribute

This sub-menu distributes the selected items. There are four distribution options provided.

  • Distribute Horizontal Spacing. The leftmost and rightmost selected model elements are not moved. The others are adjusted horizontally until the horizontal space (i.e. from the right edge of the left model element to the left edge of the right model element) is the same for all of the selected items

  • Distribute Horizontal Centers. The leftmost and rightmost selected model elements are not moved. The others are adjusted horizontally until the distance between the horizontal centers of all the selected items is the same.

  • Distribute Vertical Spacing. The top and bottom selected model elements are not moved. The others are adjusted vertically until the vertical space (i.e. from the bottom edge of the top model element to the top edge of the bottom model element) is the same for all of the selected items

  • Distribute Vertical Centers. The top and bottom selected model elements are not moved. The others are adjusted vertically until the distance between the vertical centers of all the selected items is the same.

10.7.3.  Reorder

This sub-menu adjusts the ordering of overlapping items. There are four reorder options provided.

  • Forward. The selected model elements are moved one step forward in the ordering hierarchy with respect to other model elements they overlap.

  • Backward. The selected model elements are moved one step back in the ordering hierarchy with respect to other model elements they overlap.

  • To Front. The selected model elements are moved to the front of any other model elements they overlap.

  • To Back. The selected model elements are moved to the back of any other model elements they overlap.

10.7.4.  Size To Fit Contents

This menu-item acts on all selected items on the current diagram. It resets all sizes of all model elements to its minimum size for which all text fits inside.

10.7.5.  Layout

This menu-item provides an automatic diagram layout function, i.e. when activating this menu-item, all items on the current class diagram are rearranged according a certain layout algorithm.

This function currently only works for classdiagrams. For all other types of diagrams, the menu-item does not do anything .

10.8.  The Generation Menu

This menu provides support for code generation from UML diagrams. The functionality is built around the structural information of class diagrams.

[Note]Note

Without any plugin modules installed, ArgoUML supports only code generation of Java. ArgoUML V0.20 supports the following languages by plugin: C#, C++, php4, php5.

[Warning]Warning

Code generation is still very much a work in progress. The current version of ArgoUML will generate a structural template for your code, but is not able to handle behavioral specifications to generate code for the dynamic behavior of the model.

10.8.1.  Generate Selected Classes ...

This menu entry brings up a dialog box for the ArgoUML code generator (see Figure 10.26, “ The dialog for Generate Selected Classes.... ).

Figure 10.26.  The dialog for Generate Selected Classes....

The dialog for Generate Selected Classes....

Below a label Available Classes the dialog box lists each of the selected classes by name with a check box to the left, for each language installed. All the checkboxes are initially unchecked. Checking any of these boxes will cause code generation for that class. Checking multiple languages for a class causes it to be generated in all these languages.

The buttons Select All and Select None may help when a lot of items have to be selected or deselected.

In the lower portion of the dialog box is an editable combo box labeled Output Directory to specify the directory in which code is generated. Within this directory, a top level directory will be created with the name of the model. Further sub-directories will be created to reflect the package/namespace hierarchy of the model. A drop down selector gives access to previously selected output directories.

Finally, at the bottom of the dialog box are two buttons, labeled Generate and Cancel. Button 1 click on the former will cause the code to be generated, button 1 click on the latter will cancel code generation.

10.8.2.  Generate All Classes...

Shortcut F7.

This function behaves as Generate Selected Classes... (see Section 10.8.1, “ Generate Selected Classes ...) would with all classes in the current diagram selected.

10.8.3.  Generate Code for Project... (To be Written)

10.8.4.  Settings for Generate for Project... (To be Written)

10.9.  The Critique Menu

This menu controls one of ArgoUML's unique features—the use of critics to guide the designer. The theory behind this is well described in Jason Robbins' PhD dissertation http://argouml.tigris.org/docs/robbins_dissertation/.

[Note]Note

A word about terminology: The critics are background processes, which evaluate the current model according to various “good” design criteria. There is one critic for every design criterion.

The output of a critic is a critique—a statement about some aspect of the model that does not appear to follow good design practice.

Finally a critique will generally suggest how the bad design issue it has identified can be rectified, by raising a to-do item.

[Note]Note

The critics run as asynchronous processes in parallel with the main ArgoUML tool. Changes typically take a second or two to propagate as the critics wake up.

10.9.1.  Toggle Auto-Critique

This is a check box, controlling whether the critics are running. By default it is checked. If unchecked, then all critics are disabled, and any to-do items generated by critics (the only others being those the designer has added by hand) are hidden in the to-do pane.

10.9.2.  Design Issues...

This menu entry brings up a dialog box controlling how critics associated with a particular design area are to be handled (see Figure 10.27, “ The dialog for Design Issues.... ).

Figure 10.27.  The dialog for Design Issues....

The dialog for Design Issues....

ArgoUML categorizes critics according the the design issue they address. There are 16 such categories. The critics in each category are discussed in detail in the chapter on critics (Chapter 15, The Critics ).

The sliders may be set for each category to control the critics that trigger for that category. Setting a slider to Off will disable all critics in that category, and remove all associated to-do items from the to-do pane.

Setting a slider to a higher priority value will enable all critics at or above that priority level within the design issue category (Off being the lowest priority).

[Note]Note

The sliders are set by default to High for all design categories.

10.9.3.  Design Goals...

This menu entry brings up a dialog box controlling how design goals are to be handled (see Figure 10.28, “ The dialog for Design Goals.... ).

Figure 10.28.  The dialog for Design Goals....

The dialog for Design Goals....

ArgoUML has the concept that the designer will have a number of design goals to be achieved (for example good structural representation, detailed behavioral representation etc). Critics are associated with one or more goals.

This dialog allows the user to specify the priority of each design goal.

The sliders may be set for each design goal to control the critics that trigger for that goal. Setting a slider to zero will disable all critics in that goal, and remove all associated to-do items from the to-do pane.

Setting a slider to a higher value will enable all critics at or above that priority level within the design issue category (1 being the highest priority and 5 the lowest).

[Tip]Tip

It may be useful to think of this function as very similar to Design Issues... (see Section 10.9.2, “ Design Issues...), but with grouping of critics according to the outcomes of OOA&D rather than grouping according to the structure of UML.

[Warning]Warning

The V0.20 version of ArgoUML provides a single design goal, Unspecified, with its slider set by default to priority 1. However it contains no critics and so has no effect.

10.9.4.  Browse Critics...

This menu entry brings up a dialog box controlling the individual critics (see Figure 10.29, “ The dialog for Browse Critics.... ).

Figure 10.29.  The dialog for Browse Critics....

The dialog for Browse Critics....

This dialog controls the behavior of individual critics. To the left is a list of all the critics, to enable them to be switched on or off individually. For each critic there are three columns, labeled Active, Headline and Snoozed. The first of these is a check box, which may be toggled with button 1 click. The second is the headline name of the critic, the third indicates if the critic has been snoozed from the to-do pane (see Chapter 14, The To-Do Pane . A critic is only really active if the box in the first column is checked and the critic has not been snoozed.

Any critic for which the box in the first column is unchecked is inactive and will not trigger. In addition any to-do items associated with that critic will be removed from the to-do pane.

The V0.26 version of ArgoUML has a total of 90 critics, a few of which are incompletely implemented. They are described in detail by design issue category in the chapter on critics (see Chapter 15, The Critics ).

To the right of the list are a series of fields, titled Critic Details, giving detailed control over individual critics. Selecting a critic in the list on the left will populate the fields for that critic.

The first field on the right is titled Critic Class: and then the full name of the class in ArgoUML that implements the critic. This name can be used as unique identifier of the critique, e.g. in conversations about the critic.

The first field below this title is a text box labeled Headline: giving the complete headline of the critic (which may be truncated in the list on the left).

[Note]Note

In the headline you may see the text <ocl>self</ocl>, which will be replaced by the name of the model element in question when the critic is triggered.

The next field is a drop-down selector, labeled Priority:. The three options available are High, Medium and Low and specify the priority category of any to-do item generated by this critic. This does not alter the priority of the already existing todo items, only the newly generated ones. Changing the priority of a critic is not saved persistently.

The next field is labeled MoreInfo: and contains a URL pointing to further information with a button to the right labeled Go to navigate to that URL.

[Warning]Warning

In the V0.26 release of ArgoUML there is no further information available, and the Go button is always grayed out and disabled.

The next field is labeled Description: and is a text area with a detailed explanation of what this critic means. If the text is too large for the area a scroll bar is provided to the right.

[Note]Note

In this text area you may see the text <ocl>self</ocl>, which will be replaced by the name of the model element in question when the critic is triggered.

The last field is a drop-down selector labeled Use Clarifier, with three options, Always, If Only One and Never.

Clarifiers are the icons and wavy red underlines drawn on the actual diagrams to indicate the artefact to which the critic refers. The original intention was to make the mapping from critics to clarifiers somewhat customizable.

For example one user might make a Missing Name critic show a red underline, another user might turn off the clarifier, or have it draw a wavy green underline or a blue questionmark. Critics with their clarifier's disabled would still produce feedback that is listed in the to-do pane.

[Caution]Caution

In the V0.26 release of ArgoUML this selector has no function whatsoever. It is for future development.

Underneath the fields are two buttons in a horizontal row.

  • Wake. It is possible to snooze a critic from the to-do pane (see Chapter 14, The To-Do Pane ), which makes the critic inactive for a period. If the critic has been snoozed, this button is enabled and will wake the critic back up again. Otherwise it is grayed out.

    [Tip]Tip

    You can tell a snoozed critic, because in the list on the left it will be indicated in the third column.

  • Advanced. This button causes ArgoUML to show a few columns extra in the table of critics. It allows more detailed inspection of critic properties.

Finally the bottom right of the dialog contains a button labeled OK. Button 1 click here dismisses the dialog.

10.10.  The Tools Menu

This menu provides a generic menu attachment point for any plug-ins provided with ArgoUML. The standard system has no plug-in, and this menu entry is empty by default.

10.11.  The Help Menu

This menu provides help on the use of ArgoUML. It has two entries.

10.11.1.  System Information

This menu entry brings up the system information dialog, see Figure 10.30, “ The dialog for System Information.

Figure 10.30.  The dialog for System Information.

The dialog for System Information.

Use this menu to describe the system that runs ArgoUML to the system manager or developer. Pressing the button Run Garbage Collector not only runs the Java gargage collector, but also refreshes the information shown. To facilitate copy and paste into (e.g.) an email, the button Copy Information to System Clipboard is foreseen. The Cancel button dismisses the dialog box.

10.11.2.  About ArgoUML

This menu entry brings up the help window for ArgoUML (see Figure 10.31, “ The help window for ArgoUML).

Figure 10.31.  The help window for ArgoUML

The help window for ArgoUML

The window has six tabs, which are selected by button 1 click. By default the first tab (Splash) is shown.

  • Splash. This displays the picture shown when ArgoUML starts up, and the current version number.

  • Version. This provides version information on the various packages that make up ArgoUML, and some operating system and environment information.

  • Credits. This details all those who have created ArgoUML, including contact details for the various module owners.

  • Contact Info. This gives the major contact points for the ArgoUML project—the web site, and the developers mailing list.

  • Report bugs. This gives information about how to deal with bugs in ArgoUML. It is important that all bugs are reported, and all cooperation is appreciated.

  • Legal. A statement of the FreeBSD license which covers all the ArgoUML software.

    [Caution]Caution

    The various documentation of the project are not all covered by FreeBSD (which is really meant for software). In particular this manual is covered by the OpenPub license (see Appendix F, Open Publication License ).

Chapter 11.  The Explorer

The Explorer was previously called Navigation Pane/Tree or sometimes Navigator Pane/Tree.

11.1.  Introduction

Figure 11.1, “ Overview of the explorer shows the ArgoUML window, with the explorer highlighted.

Figure 11.1.  Overview of the explorer

Overview of the explorer

The explorer allows the user to view the structure of the model from a number of predefined perspectives. It also allows the user to define their own perspetives for custom exploring of the model.

An important feature, related to the cognitive psychology ideas behind ArgoUML is that not all model elements are necessarily shown in all perspectives. Rather, the perspectives are used to implement hiding of uninteresting parts of the model.

11.2.  Mouse Behavior in the Explorer

Behavior of the mouse in general, and the naming of the buttons is covered in the chapter on the overall user interface (see Chapter 8, Introduction ).

11.2.1.  Button 1 Click

Within the hierarchical display, elements which have sub-hierarchies are indicated by when the hierarchy is hidden and when the hierarchy is open.

Button 1 click over the name of any diagram model element will cause the diagram to be selected and displayed in the editing pane. Its details will also be displayed in the details pane.

Button 1 click over the name of any model element other than a diagram in the main area of the explorer will cause it to be selected, and its details shown in the details pane. If the model element is part of a diagram currently displayed in the editing pane, it will be highlighted there.

[Note]Note

If the model element is part of a diagram other than that currently displayed in the Editing Pane, there will be no change of diagram in the Editing Pane.

Where button 2 click has been used to bring up a context sensitive pop-up menu (see below), button 1 click is used to select the menu entry required. button 1 click outside the menu area will remove it.

11.2.2.  Button 1 Double Click

This has the effect of a button 1 single click, and if the tree item was not a leaf, it will toggle the hierarchy open or close.

11.2.3.  Button 1 Motion

Button 1 motion means that you pick up one or more modelelements, and drag them to a new location. Dropping the modelelement somewhere causes ArgoUML to execute some function that depends on where you drop the modelelements.

11.2.3.1.  From Explorer to Explorer

Releasing the mouse button above a namespace, makes the modelelement owned by the namespace. In the Package-centric explorer perspective, this means a straigh-forward drag-and-drop function.

Use this drap and drop feature to easily move e.g. classes from one package into another.

11.2.3.2.  From Explorer to Diagram

Dropping a modelelement on the diagram is the equivalent of the "Add to Diagram" function. Hence, if the diagram did not yet show this modelelement, it is added.

Use this drap and drop feature e.g. to easily create a diagram from imported XMI files. This because XMI files contain all the modelelements, but not any diagram information.

11.2.4.  Button 2 Actions

When used in the the explorer, this will display a selection dependent pop-up menu. Menu entries are highlighted (but not selected) and sub-menus exposed by subsequent mouse motion (without any buttons). Menu entry selection is with button 1 or button 2.

11.2.5.  Button 2 Double Click

This has no effect other than that of button 2 single click.

11.3.  Keyboard Behavior in the Explorer

All keys active in a tree widget have their normal behaviour.

When a diagram is selected, pressing Ctrl-C will copy the diagram in GIF format to the system clipboard.

11.4.  Perspective Selection

The model elements in the ArgoUML model may be configured for displaying in the tree by a number of perspectives. To this end, a drop-down at the top allows selection of the explorer perspective.

Below that, there is a drop-down to select the ordering of the atifacts within the hierarchy. The two possibilities are "Order by Type, Name" and "Order by Name". The former groups all items per type, and sorts them per group alphabetically on the name. The lattter simply sorts on name only.

The following explorer perspectives may be selected in the drop-down at the top:

  • Package-centric (the default). The exploring hierarchy is organized by package hierarchy. The top level shows the model. Under this are all the top level packages in the model and all the model elements that are directly in the namespace of the model.

    Beneath each package are all the model elements that sit within the namespace of that package, including any further sub-packages (which in turn have their own sub-hierarchies).

  • Class-centric. Shows classes in their package hierarchy as well as datatypes and use case diagram elements. Similar to the Package-centric view but it doesn't show connecting or associating elements.

  • Diagram-centric. In this view the top level comprises all the diagrams in the model. Beneath each diagram is a flat listing of all the model elements on the diagram. Model elements that have sub-model elements that do not appear on the diagram have their own hierarchy (for example attributes and operations of classes).

  • Inheritance-centric. In this view the top level shows the model. Beneath this are all model elements that have no generalization in the model. Those model elements that have specializations have a sub-hierarchy showing the specializations.

  • Class Associations. In this view the top level shows the model. Beneath this are all diagrams and all classes. All classes that have associations have a hierarchy tracking through the associated classes.

  • Residence-centric. In this view the model is shown at the top-level, with below it only Nodes, and below these only components that reside on the nodes, and below these components all elements that reside on the components.

  • State-centric. In this view the top level shows all the state machines and all activity graphics associated with classes.

    Beneath each state machine is a hierarchy showing the statechart diagram and all of the states. Beneath each state is a list of the transitions in and out of the state.

    Beneath each activity graph is a hierarchy showing the activity diagram and all of the action states. Beneath each action state is a list of the transitions in and out of the action state.

  • Transitions-centric. This is very similar to State-centric view, but under each state machine is listed the diagrams and all transitions on the diagram, with states being shown as sub-hierarchies under their connected transitions.

    Similarly under each activity graph is listed the diagrams and all transitions on the diagram, with action states being shown as sub-hierarchies under their connected transitions.

  • Composite-centric. In this view, all modelelements are shown according their composition in the UML metamodel.

    This perspective shows far more modelelements then all others - it does not hide anything. Hence, this view is not so user-friendly, but very suited for the UML specialist.

11.5.  Configuring Perspectives

The explorer is designed to be user configurable, to allow the designer to view in his or her preferred way.

11.5.1.  The Configure Perspectives dialog

button 1 click on the "Configure Perspectives" icon ( ) at the top left of the explorer brings up the explorer perspectives dialog (see Figure 11.2, “ The Configure Perspectives dialog box).

Figure 11.2.  The Configure Perspectives dialog box

The Configure Perspectives dialog box

The top half of the dialog contains a list of all the currently defined perspectives and to the right a series of buttons stacked vertically. Button 1 click can be used to select a perspective. You can select only one perspective at a time.

Selecting a perspective reveals a text field above the list, where the name of the perspective can be edited.

The lower half of the dialog contains two list areas. The one on the left, labeled Rules Library, contains the list of available rules that may be used to create the perspective. The one on the right, labeled Selected Rules contains the actual rules chosen for the perspective that has been selected in the list of perspectives at the top. In both lists, you can select only one rule at a time.

Separating the two areas in the lower half of the dialog are buttons labeled >> and <<. The first of these transfers the rule selected in the library on the left to the list of rules on the right—i.e. it adds a rule to the perspective. The second one transfers the rule selected on the right to the library list on the left—i.e. it removes a rule from the perspective.

If you hover the mouse over the horizontal line that separates the two halves of the dialog, then you see it change shape, to indicate that you can grab this line and drag it up or down.

All three titles of the lists show the number of items in the list. ArgoUML V0.26 has 9 default perspectives, and 72 rules in the library to build perspectives from.

The buttons at the top right are explained as follows:

  • New. This creates a new perspective from scratch with no rules selected, with an automatically generated name.

  • Remove. This removes the selected perspective.

  • Duplicate. This creates a copy the selected perspective so it can be used as the basis of a new perspective. The new one is named "Copy of" followed by the original name.

  • Move Up. This moves the selected perspective one place up in the list. This button is downlighted for the topmost perspective.

  • Move Down. This moves the selected perspective one place down in the list. This button is downlighted for the last perspective.

  • Restore Defaults. This restores all perspectives and their selected rules to the build-in defaults of ArgoUML.

At the very bottom right is a button labeled OK to be used when all changes are complete. button 1 click on this button will close the dialog window. The changes are saved when you exit ArgoUML in the argo.user.properties file.

Then there is the Cancel button, which cancels all changes made in the dialog. Pressing the dialog close icon (usually at the top right corner) has the same effect as pressing the cancel button.

11.6.  Context Sensitive Menu

Button 2 Click over any selected model element in the main area of the explorer will cause a pop-up menu to appear.

The presence of all these menu entries depends on the selected element(s), and other circumstances. E.g. the "Create ModelElement" submenu is not present on modelelements that are part of a non-editable profile element.

11.6.1.  Create Diagram

This entry on the pop-up menu opens a choice of submenus, one for each diagram type.

The namespace of the new diagram will be based on the selected modelelement.

11.6.2.  Create ModelElement

This entry on the pop-up menu opens a choice of submenus, one for each model element.

The new model element will be composed by the selected modelelement.

Selecting multiple elements leads to different menu items, e.g. selecting 2 classes allows the creation of several kinds of relationships.

11.6.3.  Copy Diagram to Clipboard as Image

This entry on the pop-up menu creates a graphical file, in the default graphical format, and puts it on the clipboard of your PC. The graphics can inmediately be pasted into e.g. a requirements document in OpenOffice.Org.

The graphics format and its resolution are determined by ArgoUML's default setting: Select in the menu Edit, then Settings..., then the tab Environment. The PNG and GIF formats are advised, and the resolution Standard.

[Tip]Tip

Some applications (such as Doors from Telelogic) require the background color of the generated graphics to be adapted (else the image is empty). This can be done with a tool like IrfanView; it is as easy as clicking its paste button, and then its copy button.

11.6.4.  Add to Diagram

This entry on the pop-up menu appears for any model element that could be added to the diagram in the editing pane.

The item can be placed in a diagram by moving the cursor to the editing pane or a spawned editing pane window (where it will appear as a cross) and clicking button 1.

[Caution]Caution

This menu entry only appears as not grayed out, if the diagram in the editor pane allows to contain the model element, and the model element is not present yet in the diagram. ArgoUML will not let you place more than one copy of any particular model element on a diagram.

11.6.5.  Delete From Model

This entry on the pop-up menu appears for any model element that could be deleted from the model.

[Warning]Warning

This deletes the model element from the model completely, not just from the diagram. To remove the model element just from the diagram, use the edit menu (see Section 10.4.2, “ Remove From Diagram ).

[Caution]Caution

You can delete a diagram from the model. Depending on the type of diagram, that might delete all model elements shown on the diagram. To illustrate the differences, consider the following examples:

  • Deleting a class diagram does not delete any model element drawn on it. All model elements that were shown on the diagram remain present in the model. This because a class diagram does not "map" on any model element according the UML standard V1.4.

  • Deleting a statechart diagram also deletes the statemachine it represents, and hence also all the model elements owned by the statemachine. This because a statechart diagram does "map" into a StateMachine according the UML standard V1.4.

11.6.6.  Set Source Path... (To be written)

This entry on the pop-up menu ...

11.6.7.  Add Package

This entry on the pop-up menu is available whenever a model element is selected that may contain a package, e.g. a package. After activating this menu the model element will own a new package.

11.6.8.  New Stereotype

This entry on the pop-up menu is available ... (to be written)

11.6.9.  Add All Classes in Namespace Alle Klassen im Namensraum hinzufügen

This entry on the pop-up menu is available for Class Diagrams only. Activating this menu-item will add all classes in the current namespace to the diagram. They will be located at the top left corner—obviously a perfect occasion to use the “Arrange->Layout” function in the menu.

Chapter 12.  The Editing Pane

12.1.  Introduction

Figure 12.1, “ Overview of the editing pane shows the ArgoUML window with the editing pane highlighted.

Figure 12.1.  Overview of the editing pane

Overview of the editing pane

This is where all the diagrams are drawn. In earlier versions of ArgoUML this pane went under a variety of names. You may encounter “drawing pane”, “diagram pane” or “multi-editor pane” in other documentation that is still being updated.

The pane has a tool bar at the top, and a single tab labeled As Diagram at the bottom, which has no function in the 0.20 version of ArgoUML. The main area shows the currently selected diagram, of which the name is shown in the window title bar.

12.2.  Mouse Behavior in the Editing Pane

Behavior of the mouse in general, and the naming of the buttons is covered in the chapter on the overall user interface (see Chapter 8, Introduction ).

12.2.1.  Button 1 Click

In the tool bar of the editing pane, button 1 click is used to select a tool for creating a new model element and adding it to the diagram (see double clicking for creating multiple model elements). For most tools, adding a new model element to the diagram is achieved by moving the mouse into the editing area and clicking again.

In the main editing area button 1 click is used to select an individual model element.

Many model elements (e.g. actor, class) show special handles when selected and the mouse hovers over them. These are called “Selection Action Buttons”, see Section 12.6, “ Selection Action Buttons. They appear at the sides, top and bottom, and indicate a relationship type. Clicking on a Selection Action Button creates a new related model element, with the relation of the type that was indicated. If the shift key is pressed when hovering the mouse over a selected model element, sometimes different handles are shown, which stand for different relation types.

Where button 2 click has been used to bring up a context sensitive pop-up menu (see below), button 1 click is used to select the menu entry required. The pop-up menu will be removed by any button 1 click outside of the menu area.

There are various more detailed effects, which are discussed under the descriptions of the various tools (see Section 12.4, “ The tool bar).

12.2.2.  Button 1 Double Click

When used on the tool bar with a tool to add a model element, the selected model element will be added multiple times to the drawing area, once for each further button click, until the tool is again selected or another tool chosen.

When used within the drawing area on a model element that has sub-components, double click will select the sub-component for editing (creating it if necessary).

For example double clicking over an operation compartment of a class will select the operation. Or create one if there is none yet.

A special use is with package model elements on the class diagram. A double click on a package will navigate to the class diagram associated with a package (the first created if there is more than one), or will offer to create one for you if there is none. See Figure 12.2, “ The dialog for adding a new class diagram

Figure 12.2.  The dialog for adding a new class diagram

The dialog for adding a new class diagram

12.2.3.  Button 1 Motion

Where the model element being added is some form of connector its termination point is shown with button 1 up over the terminating model element. button 1 click may be used in the space between model elements to create articulation points in the connector. This is particularly useful where connectors must loopback on themselves.

Over graphical model elements button 1 motion will move the model element to a new position.

Graphical model elements that are selected show handles at the corners or ends, and these can be used for re-sizing.

Some model elements (e.g. actor, class) show special handles (called “Selection Action Buttons”, see Section 12.6, “ Selection Action Buttons) at the sides, top and bottom, which can be dragged to form types of relationship with other model elements.

Where the model element is some form of connector between other items, button 1 motion other than at a handle will cause a new handle to be created, allowing the connector to be articulated at that point. This only works when the connecting line is not straight angled. Such new handles can be removed by moving them to the end of the connector.

There are various more detailed effects, which are discussed under the descriptions of the various tools (see Section 12.4, “ The tool bar).

12.2.4.  Shift and Ctrl modifiers with Button 1

Where multiple selections are to be made, the CTRL key is used with button 1 to add unselected model elements to the current selection. Where a model element is already selected, it is removed from the current selection.

Clicking Button 1 while the SHIFT key is pressed, invokes the broom tool, which causes the selected model elements (and any others swept up with them) to be moved with the broom tool (see Section 12.4.1, “ Layout Tools).

12.2.5.  Alt with Button 1 motion

Button 1 down anywhere in the diagram while the ALT key is pressed, allows to scroll the canvas in all directions with button 1 motion.

12.2.6.  Button 2 Actions

When used over model elements in the the editing pane, this will display a context dependent pop-up menu. Menu entries are highlighted (but not selected) and sub-menus exposed by subsequent mouse motion (without any buttons). Menu entry selection is with button 1 or button 2. See Section 12.10, “ Pop-Up Menus for details of the specific pop-up menus.

In case multiple elements are selected, the pop-up menu only appears if all the items are of the same kind. In this case, the functions apply to all selected elements.

12.2.7.  Button 2 Double Click

This has no effect other than that of button 2 single click.

12.2.8.  Button 2 Motion

This is used to select items in a context sensitive menu popped up by use of button 2 click.

12.3.  Keyboard Behavior in the Editing Pane

Many keyboard shortcuts can be used when the editing pane is active, mainly to change the selection or to move across the model elements.

12.3.1.  Nudging a model element

You can nudge a fig by selecting an element and using arrow keys. Keeping the shift or the alt key pressed will produce a wider movement.

12.3.2.  Moving across the model elements

You can select the model element nearer to the selected one by using the arrow keys while clicking the right button of the mouse. You can also select the next model element using the tab key, or the previous using the ctrl+tab keys.

12.4.  The tool bar

The toolbar at the top of the editing pane provides the main functions of the pane. The default tool is the Select tool (). In general button 1 click on any tool selects a tool for one use, before reverting to the default tool, and button 1 double click selects a tool for repeated use.

The tools fall into four categories.

  • Layout tools. Provide assistance in laying out model elements on the diagram.

  • Annotation tools. Used to annotate model elements on the diagram.

  • Drawing tools. Used to add general graphic objects to diagrams.

  • Diagram specific tools. Used to add UML model elements specific to a particular diagram type to the diagram.

Some of the tools that are generally not all used so often, are combined in a dropdown, to take less space on the toolbar. See e.g. Figure 12.3, “ The drawing tools selector.. Press the symbol at the right of the tool to pop it open. These drop-down tools remember their last used tool persistently. This means that when ArgoUML starts, they show the last tool that was activated the previous time ArgoUML was run.

12.4.1.  Layout Tools

The following two tools are provided in all diagrams in this category.

  • Select. This tool provides for general selection of model elements on the diagram. Button 1 click will select a model element. CTRL with button 1 can be used to select (or deselect) multiple model elements. Button 1 motion will move selected 2D items or add and move a new control point on a link. Button 1 motion on a selected component's control point will stretch that component's shape.

  • Broom. Button 1 motion with this tool provide a “broom” which will sweep all model elements along. This is a very shortcut way of lining things up.

    The Broom can also be invoked by using SHIFT with button 1 motion when the Select tool is in use.

    The Broom is discussed at length in its own chapter, see Section 12.5, “ The Broom

[Tip]Tip

Additional control of model element layout is provided through the Arrange menu (see Section 10.7, “ The Arrange Menu).

12.4.2.  Annotation Tools

The annotation tool Comment ( ) is used to add a comment to a selected UML model element.

[Caution]Caution

Unlike most other tools you use the Select tool to select a model element, and then button 1 click on Comment to create the comment. If no element is selected when the comment tool is clicked, then the comment is created and put at the left top corner.

The comment is created alongside the selected model element, empty by default. The text can be selected with button 1 double-click and edited from the keyboard.

The UML standard allows comments to be attached to any model element.

You can link any comment to aditional elements using the CommentLink ( ) tool.

12.4.3.  Drawing Tools

These are a series of tools for providing graphical additions to diagrams. Although they are not UML model elements, the UML standard provides for such decoration to improve the readability of diagrams.

[Tip]Tip

These drawing tools provide a useful way to partially support some of the UML features (such as general purpose notes) that are missing from the current release of ArgoUML.

Eight tools are provided, all grouped into one drop-down widget. See Figure 12.3, “ The drawing tools selector.. Button 1 click on the diagram will place an instance of the graphical item of the same size as the last one placed. The size can be controlled by button 1 motion during placement. One side or end of the element will be at button 1 down, the other side or end at button 1 up. In general after they are placed on the diagram, graphical elements can be dragged with the Select tool and button 1 and re-sized by button 1 motion on the handles after they have been selected.

Figure 12.3.  The drawing tools selector.

The drawing tools selector.

  • Rectangle. Provides a rectangle.

  • Rounded Rectangle. Provides a rectangle with rounded corners. There is no control over the degree of rounding.

  • Circle. Provides a circle.

  • Line. Provides a line.

  • Text. Provides a text box. The text is entered by selecting the box and typing. Text is centered horizontally and after typing, the box will shrink to the size of the text. However it can be re-sized by dragging on the corners.

  • Polygon. Provides a polygon. The points of the polygon are selected by button 1 click and the polygon closed with button 1 double click (which will link the final point to the first point).

  • Spline. Provide an open spline. The control points of the spline are selected with button 1 and the last point selected with button 1 double click.

  • Ink. Provide a polyline. The points are provided by button 1 motion.

12.4.4.  Use Case Diagram Specific Tools

Several tools are provided specific to UML model elements on use case diagrams. The detailed properties of these model elements are described in the section on use case diagram model elements (see Chapter 17, Use Case Diagram Model Element Reference ).

  • Actor. Add an actor to the diagram. For convenience, when the mouse is over a selected actor it displays two handles to left and right which may be dragged to form association relationships.

  • Use Case. Add a use case to the diagram. For convenience, when the mouse is over a selected use case it displays two handles to left and right which may be dragged to form association relationships and two handles top and bottom which may be dragged to form generalization and specialization relationships respectively.

  • Association. Add an association between two model elements selected using button 1 motion (from the first model element to the second). There are 6 types of association offered here, see Figure 12.4, “ The association tool selector.: association, aggregation and composition, and all these three can be bidirectional or unidirectional.

    Figure 12.4.  The association tool selector.

    The association tool selector.

  • Dependency. Add a dependency between two model elements selected using button 1 motion (from the dependent model element).

  • Generalization. Add a generalization between two model elements selected using button 1 motion (from the child to the parent).

  • Extend. Add an extend relationship between two model elements selected using button 1 motion (from the extended to the extending use case).

  • Include. Add an include relationship between two model elements selected using button 1 motion (from the including to the included use case).

  • Add Extension Point. Add an extension point to a selected use case. The extension point is given the default name newEP and location loc. Where the extension point compartment is displayed, the extension point may be edited by button 1 double click and using the keyboard, or by selecting with button 1 click (after the use case has been selected) and using the property tab. Otherwise it may be edited through its property tab, selected through the property tab of the owning use case.

    [Note]Note

    This tool is grayed out except when a use case is selected.

12.4.5.  Class Diagram Specific Tools

Several tools are provided specific to UML model elements on class diagrams. The detailed properties of these model elements are described in the section on class diagram model elements (see Chapter 18, Class Diagram Model Element Reference ).

  • Package. Add a package to the diagram.

  • Class. Add a class to the diagram. For convenience, when the mouse is over a selected class it displays two handles to left and right which may be clicked or dragged to form association relationships (or composition in case SHIFT has been pressed) and two handles top and bottom which may be dragged or clicked to form generalization and specialization relationships respectively.

  • Association. Add an association between two model elements selected using button 1 motion (from the first model element to the second). There are 2 types of association offered here, bidirectional or unidirectional.

  • Aggregation. Add an aggregation between two model elements selected using button 1 motion (from the first model element to the second). There are 2 types of aggregation offered here, bidirectional or unidirectional.

  • Composition. Add an composition between two model elements selected using button 1 motion (from the first model element to the second). There are 2 types of composition offered here, bidirectional or unidirectional.

  • Association-end. Add another end to an already existing association using button 1 (from the association middle to a class, or vice versa). This is the way to create so-calld N-ary associations.

  • Generalization. Add a generalization between two model elements selected using button 1 (from the child to the parent).

  • Interface. Add an interface to the diagram. For convenience, when the mouse is over a selected interface it displays a handle at the bottom which may be dragged to form a realization relationship (the target being the realizing class).

  • Realization. Add a realization between a class and an interface selected using button 1 motion (from the realizing class to the realized interface).

  • Dependency. Add a dependency between two model elements selected using button 1 motion (from the dependent model element). There are also 2 special types of dependency offered here, Permission ( ) and Usage ( ). A Permission is created by default with stereotype Import, and is used to import elements from one package into another.

  • Attribute. Add a new attribute to the currently selected class. The attribute is given the default name newAttr of type int and may be edited by button 1 double click and using the keyboard, or by selecting with button 1 click (after the class has been selected) and using the property tab.

    [Note]Note

    This tool is grayed out except when a class is selected.

  • Operation. Add a new operation to the currently selected class or interface. The operation is given the default name newOperation with no arguments and return type void and may be edited by button 1 double click and using the keyboard, or by selecting with button 1 click (after the class has been selected) and using the property tab.

    [Note]Note

    This tool is grayed out except when a class or interface is selected.

  • Association Class. Add a new association class between two model elements selected using button 1 motion (from the first model element to the second).

  • Datatype. Add a datatype to the diagram. For convenience, when the mouse is over a selected datatype it displays handles at the top and at the bottom which may be clicked or dragged to form a generalization relationship (the target being another datatype). There are 2 other elements available here, Enumeration and Stereotype. These two have similar handles, except the one at the top of a stereotype: when clicked, it creates a metaclass, connected by a dependency marked with «stereotype». This eases the creation of "stereotype declaration" diagrams - see the literature on the subject.

12.4.6.  Sequence Diagram Specific Tools

Seven tools are provided specific to UML model elements on sequence diagrams. The detailed properties of these model elements are described in the section on sequence diagram model elements (see Chapter 19, Sequence Diagram Model Element Reference ).

  • ClassifierRole. Add a classifierrole to the diagram.

  • Message with Call Action. Add a call message between two classifierroles selected using button 1 motion (from the originating classifierrole to the receiving classifierrole).

  • Message with Return Action. Add a return message between two classifierroles selected using button 1 motion (from the originating classifierrole to the receiving classifierrole).

  • Message with Create Action. Add a create message between two classifierroles selected using button 1 motion (from the originating classifierrole to the receiving classifierrole).

  • Message with Destroy Action. Add a destroy message between two classifierroles selected using button 1 motion (from the originating classifierrole to the receiving classifierrole).

  • Add Vertical Space to Diagram. Add vertical space to a diagram by moving all messages below this down. Click the mouse at the point where you want the space to be added and drag down the screen vertically the distance which matches the height of the space you'd like to have added.

  • Remove Vertical Space in Diagram. Remove vertical space from diagram and move all elements below up vertically. Click and drag the mouse vertically over the space that you want deleted.

12.4.7.  Collaboration Diagram Specific Tools

Three tools are provided specific to UML model elements on collaboration diagrams. The detailed properties of these model elements are described in the section on collaboration diagram model elements (see Chapter 21, Collaboration Diagram Model Element Reference ).

  • Classifier Role. Add a classifier role to the diagram.

  • Association Role. Add an association role between two classifier roles selected using button 1 motion (from the originating classifier role to the receiving classifier role). There are 6 types of association roles offered here, see Figure 12.4, “ The association tool selector.: association, aggregation and composition, and all these three can be bidirectional or unidirectional.

  • Generalization. Add a generalization between two model elements selected using button 1 (from the child to the parent).

  • Dependency. Add a dependency between two model elements selected using button 1 motion (from the dependent model element).

  • Add Message. Add a message to the selected association role.

    [Note]Note

    This tool is grayed out except when an association role is selected.

12.4.8.  Statechart Diagram Specific Tools

Eleven tools are provided specific to UML model elements on statechart diagrams. The detailed properties of these model elements are described in the section on statechart diagram model elements (see Chapter 20, Statechart Diagram Model Element Reference ).

  • Simple State. Add a simple state to the diagram.

  • Composite State. Add a composite state to the diagram. All model elements that are subsequently placed on the diagram on top of the composite state will form part of that composite state.

  • Transition. Add a transition between two states selected using button 1 motion (from the originating state to the receiving state).

  • Synch State. Add a synchstate to the diagram.

  • Submachine State. Add a submachinestate to the diagram.

  • Stub State. Add a stubstate to the diagram.

  • Initial. Add an initial pseudostate to the diagram.

    [Caution]Caution

    There is nothing to stop you adding more than one initial state to a diagram or composite state. However to do so is meaningless, and one of the critics will complain.

  • Final State. Add a final state to the diagram.

  • Junction. Add a junction pseudostate to the diagram.

    [Caution]Caution

    A well formed junction should have at least one incoming transition and at least one outgoing. ArgoUML does not enforce this, but an ArgoUML critic will complain about any junction that does not follow this rule.

  • Choice. Add a choice pseudostate to the diagram.

    [Caution]Caution

    A well formed choice should have at least one incoming transition and at least one outgoing. ArgoUML does not enforce this, but an ArgoUML critic will complain about any choice that does not follow this rule.

  • Fork. Add a fork pseudostate to the diagram.

    [Caution]Caution

    A well formed fork should have exactly one incoming transition and two or more outgoing. ArgoUML does not enforce this, but an ArgoUML critic will complain about any fork that does not follow this rule.

  • Join. Add a join pseudostate to the diagram.

    [Caution]Caution

    A well formed join should have exactly one outgoing transition and two or more incoming. ArgoUML does not enforce this, but an ArgoUML critic will complain about any join that does not follow this rule.

  • Shallow History. Add a shallow history pseudostate to the diagram.

  • Deep History. Add a deep history pseudostate to the diagram.

  • Call Event. Add a Call Event as trigger to a transition. There are 4 types of events offered here: Call Event, Change Event, Signal Event and Time Event.

  • Guard. Add a guard to a transition.

  • Call Action. Add a call action (i.e. the effect) to a transition. There are 7 types of actions offered here: Call Action, Create Action, Destroy Action, Return Action, Send Action, Terminate Action, Uninterpreted Action and Action Sequence.

12.4.9.  Activity Diagram Specific Tools

Seven tools are provided specific to UML model elements on activity diagrams. The detailed properties of these model elements are described in the section on activity diagram model elements (see Chapter 22, Activity Diagram Model Element Reference ).

  • Action State. Add an action state to the diagram.

  • Transition. Add a transition between two action states selected using button 1 motion (from the originating action state to the receiving action state).

  • Initial. Add an initial pseudostate to the diagram.

    [Caution]Caution

    There is nothing to stop you adding more than one initial state to a diagram. However to do so is meaningless, and one of the critics will complain.

  • Final State. Add a final state to the diagram.

  • Junction. Add a junction (decision) pseudostate to the diagram.

    [Caution]Caution

    A well formed junction should have one incoming transition and two or more outgoing. ArgoUML does not enforce this, but an ArgoUML critic will complain about any junction that does not follow this rule.

  • Fork. Add a fork pseudostate to the diagram.

    [Caution]Caution

    A well formed fork should have one incoming transition and two or more outgoing. ArgoUML does not enforce this, but an ArgoUML critic will complain about any fork that does not follow this rule.

  • Join. Add a join pseudostate to the diagram.

    [Caution]Caution

    A well formed join should have one outgoing transition and two or more incoming. ArgoUML does not enforce this, but an ArgoUML critic will complain about any join that does not follow this rule.

  • CallState. Add a callstate to the diagram. A call state is an action state that calls a single operation. Hence, the name of the operation being called is put in the symbol, along with the name of the classifier that hosts the operation in parentheses under it.

  • ObjectFlowState. Add a objectflowstate to the diagram. An objectflowstate is an object that is input to or output from an action.

12.4.10.  Deployment Diagram Specific Tools

Ten tools are provided specific to UML model elements on deployment diagrams. The detailed properties of these model elements are described in the section on deployment diagram model elements (see Chapter 23, Deployment Diagram Model Element Reference ).

[Note]Note

Remember that ArgoUML's deployment diagrams are also used for component diagrams.

  • Node. Add a node to the diagram. For convenience, when the mouse is over a selected node it displays four handles to left, right, top and bottom which may be dragged to form association relationships.

  • Node Instance. Add a node instance to the diagram. For convenience, when the mouse is over a selected node instance it displays four handles to left, right, top and bottom which may be dragged to form link relationships.

  • Component. Add a component to the diagram. For convenience, when the mouse is over a selected component it displays four handles to left, right, top and bottom which may be dragged to form dependency relationships.

  • Component Instance. Add a component instance to the diagram. For convenience, when the mouse is over a selected component instance it displays four handles to left, right, top and bottom which may be dragged to form dependency relationships.

  • Generalization. Add a generalization between two model elements selected using button 1 (from the child to the parent).

  • Realization. Add a realization between a class and an interface selected using button 1 motion (from the realizing class to the realized interface).

  • Dependency. Add a dependency between two model elements selected using button 1 motion (from the dependent model element).

  • Association. Add an association between two model elements (node, component, class or interface) selected using button 1 motion (from the first model element to the second model element). There are 6 types of association offered here, see Figure 12.4, “ The association tool selector.: association, aggregation and composition, and all these three can be bidirectional or unidirectional.

    [Caution]Caution

    The constraint that associations between classes and interfaces must not be navigable from the interface still applies on deployment diagrams.

  • Object. Add an object to the diagram. For convenience, when the mouse is over a selected object it displays four handles to left, right, top and bottom, which may be dragged to form link relationships.

  • Link. Add a link between two model elements (node instance, component instance or object) selected using button 1 motion.

12.5.  The Broom

ArgoUML's broom alignment tool is specialized to support the needs of designers in achieving the kind of alignment used in UML diagrams. It is common for designers to roughly align objects as they are created or by using simple movement commands. The broom is an easy way to precisely align objects that are already roughly aligned. Furthermore, the broom's distribution options are suited to the needs of UML designers: making related objects appear evenly spaced, packing objects to save diagram space, and spreading objects out to make room for new objects. The broom also makes it easy to change from horizontal to vertical alignment or from left-alignment to right-alignment.

The T-shaped icon in ArgoUML's diagram toolbar invokes the broom alignment tool. When the mouse button 1 is pressed while in broom-mode, the designer's initial mouse movement orients the broom to face in one of four directions: north, south, east, or west. After that, mouse drag events cause the broom to advance in the chosen direction, withdraw, or grow in a lateral direction. Like a real-world push broom, the broom tool pushes diagram elements that come in contact with it. This has the effect of aligning objects along the face of the broom and provides immediate visual feedback (see the figure below). Unlike a real-world broom, moving backwards allows diagram elements to return to their original position. Growing the broom makes it possible to align objects that are not near each other. When the mouse button is released, the broom disappears and the moved objects are selected to make it easy to manipulate them further.

Figure 12.5.  The Broom.

The Broom.
The Broom.
The Broom.
The Broom.

If the designer presses the space bar while using the broom, objects on the face of the broom are distributed (i.e., spaced evenly). ArgoUML's broom supports three distribution modes: objects can be spaced evenly across the space that they use, objects can be packed together with only a small gap between them, or objects can be distributed evenly over the entire length of the broom's face. Repeatedly pressing the space bar cycles among these three distribution modes and displays a brief message indicating the operation just performed: Space evenly, Pack tightly, Spread out and Original.

If the designer presses the Enter key while using the broom, the broom turns red (instead of the normal blue), and objects are not picked up by the broom when moving forward. It works like lifting up the broom. Pressing Enter again returns to the normal mode.

Pressing the Tab key works exactly like the Enter key.

12.6.  Selection Action Buttons

When the user selects a model element in a UML diagram, several handles are drawn on it to indicate that it is selected and to provide user interface affordances to resize the node. ArgoUML also displays some “selection-action buttons” around the selected model element. See the figure below for some examples of the handles and “selection-action buttons”. The two figures for a class differ because for creating the second one, the shift key has been depressed.

Figure 12.6.  Some examples of “Selection Action Buttons”.

Some examples of Selection Action Buttons.
Some examples of Selection Action Buttons.
Some examples of Selection Action Buttons.
Some examples of Selection Action Buttons.
Some examples of Selection Action Buttons.
Some examples of Selection Action Buttons.

Selection-action buttons offer common operations on the selected object. For example, a class node has a button at 12-o'clock for adding a superclass, one at 6-o'clock for adding a subclass, and buttons at 3-o'clock and 9-o'clock for adding associations. These buttons support a "click or drag" interaction: a single click creates a new related class at a default position relative to the original class and creates a generalization or association; a drag from the button to an existing class creates only the generalization or association; and, a drag to an empty space in the diagram creates a new class at the mouse position and the generalization or association. ArgoUML provides some automated layout support so that clicking the subclass button will position the new classes so that they do not overlap.

Selection-action buttons are transparent. They have a visibly recognizable rectangular shape and size and they contain an icon that is the same as the icon used for the corresponding type of design element on the standard toolbar. However, these icons are unfilled line drawings with many transparent pixels. This allows selection-action buttons to be overlaid onto the drawing area without overly obscuring the diagram itself. Also, the buttons are only drawn when the mouse is over the selected model element; if any part of the diagram is obscured, the mouse can simply be moved away to get a clearer view of the diagram.

12.7.  Clarifiers

A key feature of ArgoUML are the critics, which run in parallel with the main ArgoUML tool. When they find a problem, they typically raise a to-do item, and also highlight the problem on the editing pane. The graphical techniques used for highlighting are called Clarifiers

  • Note icon (). Displayed at the top left of a model element indicates a critic of that model element. Moving the mouse over the icon will pop up the critic headline.

  • Colored wavy line ( ). Used for critics specific to sub-components of graphical model elements. For example to underline attributes with a problem within a class.

  • Solid colored line ( ). Not seen in ordinary editing, but used when a to-do item is highlighted from the to-do pane (see Chapter 14, The To-Do Pane ) by button 1 double click. The solid line is used to show all the model elements affected by the critic, for example all stimuli that are out of order.

12.8.  The Drawing Grid

The editing pane is provided with a background grid which can be set in various styles or turned off altogether through the menu (see Section 10.5.4, “ Adjust Grid).

Whatever grid is actually displayed, placement of items on the diagram is always controlled by the setting for grid snap, which ranges from 4 to 32 pixels (see Section 10.5.5, “ Adjust Snap).

12.9.  The Diagram Tab

At the bottom of the editing pane is a small tab labeled as As Diagram. The concept is that a UML diagram can be displayed in a number of ways, for example as a graphical diagram or as a table. Each representation would have its own tab and be selected by button 1 click on the tab.

Earlier versions of ArgoUML did implement a tabular representation, but the current release only supports a diagram representation, so this tab does not have any function.

12.10.  Pop-Up Menus

Within the editing pane, button 2 click over a model element will bring up a pop-up menu with a variable number of main entries, many with a sub-menu.

12.10.1.  Critiques

This sub-menu gives list of all the critics that have triggered for this model element. Selection of a menu entry causes that entry to be highlighted in the to-do pane and its detailed explanation to be placed in the ToDoItem tab of the details pane. A solid colored line indicates the offending element.

12.10.2.  Ordering

This menu controls the ordering of overlapping model elements on the diagram. It is equivalent to the Reorder sub-menu of the Arrange menu (see Section 10.7.3, “ Reorder). There are four entries.

  • Forward. The selected model elements are moved one step forward in the ordering hierarchy with respect to other model elements they overlap.

  • Backward. The selected model elements are moved one step back in the ordering hierarchy with respect to other model elements they overlap.

  • To Front. The selected model elements are moved to the front of any other model elements they overlap.

  • To Back. The selected model elements are moved to the back of any other model elements they overlap.

12.10.3.  Add

This sub-menu only appears for model elements that can have notes attached (class, interface, object, state, pseudostate) or have operations or attributes added (class, interface). There are at most three entries.

  • New Attribute. Only appears where the selected model element is a class. Creates a new attribute on the model element.

  • New Operation. Only appears where the selected model element is a class or interface. Creates a new operation on the model element.

  • New Comment. Attaches a new comment to the selected model element.

  • Add All Relations. Only appears where the selected model element is a class or interface. Makes all relations visible that exist in the model and that are connected to the selected model element.

  • Remove all Relations. Only appears where the selected model element is a class or interface. Removes all connected relations from the diagram (without removing them from the model).

12.10.4.  Show

This sub-menu only appears with certain model elements. It is completely context dependent. There are many possible entries, depending on the selected model element and its state.

  • Hide Extension Point Compartment. Only appears when the extension point compartment of a use case is displayed. Hides the compartment.

  • Show Extension Point Compartment. Only appears when the extension point compartment of a use case is hidden. Displays the compartment.

  • Hide All Compartments. Only appears when both attribute and operation compartments are displayed on a class or object. Hides both compartments.

  • Show All Compartments. Only appears when both attribute and operation compartments are hidden on a class or object. Displays both compartments.

  • Hide Attribute Compartment. Only appears when the attribute compartment of a class or object is displayed. Hides the compartment.

  • Show Attribute Compartment. Only appears when the attribute compartment of a class or object is hidden. Displays the compartment.

  • Hide Operation Compartment. Only appears when the operation compartment of a class or object is displayed. Hides the compartment.

  • Show Operation Compartment. Only appears when the operation compartment of a class or object is hidden. Displays the compartment.

  • Hide Enumeration Literal Compartment. Only appears when the enumeration literal compartment of an enumeration is displayed. Hides the compartment.

  • Show Enumeration Literal Compartment. Only appears when the enumeration literal compartment of an enumeration is hidden. Displays the compartment.

  • Show All Edges. Only appears on a class. Displays all associations (to shown model elements) that are not shown yet. This is the same function as the "add to Diagram" on the asociation in the explorer context menu. currently.

  • Hide All Edges. Only appears on a class. Hides all associations. This is the same function as “Remove from Diagram” on all the associations of this class.

  • Hide Stereotype. Only appears when the Stereotype of a package is displayed. Hides the stereotype.

  • Show Stereotype. Only appears when the Stereotype of a package is hidden. Displays the stereotype.

  • Hide Visibility. Only appears when the visibility of a package is displayed. Hides the visibility.

  • Show Visibility. Only appears when the visibility of a package is hidden. Displays the visibility.

12.10.5.  Modifiers

This sub-menu only appears with class, interface, package and use case model elements. It is used to set or clear the values of the various modifiers available.

  • Abstract. Set for an abstract model element.

  • Leaf. Set for a final model element, i.e. one with no sub-model elements.

  • Root. Set for a root model element, i.e. one with no super-model elements.

  • Active. Set for a model element with dynamic behavior.

    [Note]Note

    This really ought to be set automatically for model elements with state machines or activity diagrams.

12.10.6.  Multiplicity

This sub-menu only appears with association model elements, when clicking at one end of the association. It is used to control the multiplicity at the end of the association nearest the mouse click point. There are only four entries, a sub-set of the range of multiplicities that are available through the property sheet of a association end (see Section 17.6, “ Association End).

  • 1

  • 0..1

  • 1..*

  • 0..*

12.10.7.  Aggregation

This sub-menu only appears with association model elements, when clicking at one end of the association. It is used to control the aggregation at the end of the association nearest the mouse click point. There are three entries.

  • none. Remove any aggregation.

  • aggregate. Make this end a shared aggregation (loosely known as an “aggregation”).

  • composite. Make this end a composite aggregation (loosely known as a “composition”).

[Caution]Caution

UML requires that an end with a composition relationship must have a multiplicity of 1 (the default).

12.10.8.  Navigability

This sub-menu only appears with association model elements, when clicking at one end of the association. It is used to control the navigability of the association. There are three entries.

  • bidirectional. Make the association navigable in both directions.

  • <class1> to <class2>. Make the association navigable only from <class1> to <class2>. In other words <class1> can reference <class2> but not the other way round.

  • <class2> to <class1>. Make the association navigable only from <class2> to <class1>. In other words <class2> can reference <class1> but not the other way round.

[Note]Note

UML does permit an association to be non-navigable in both directions. ArgoUML will allow this, but you will have to set each of the association ends navigation property, reached from the property tab of the association - and the diagram does not show any arrows in this case.

This is considered bad design practice (it will trigger a critic in ArgoUML), so is only of theoretical interest.

[Note]Note

UML does not permit navigability from an interface to a class. ArgoUML does not prevent this.

12.11.  Notation

Notation is the textual representation on the diagram of a modelelement or its properties.

12.11.1.  Notation Languages

ArgoUML supports showing notation in different languages. By default, all text is shown in UML notation, but the menus contain an item to select between Java and UML. With plugin modules, it is even possible to select other languages, such as C++.

Figure 12.7, “ A class in UML notation shows a class in UML notation, while Figure 12.8, “ A class in Java notation shows the same class in Java notation.

Figure 12.7.  A class in UML notation

A class in UML notation

Figure 12.8.  A class in Java notation

A class in Java notation

12.11.2.  Notation Editing on the diagram

Most text shown on a diagram may be edited by double-clicking button 1 on the text. This causes a edit box to be shown, with the previous text selected, ready for amending.

Also, the status bar of ArgoUML (i.e. the small area at the bottom of the ArgoUML window), shows an help text that indicates the syntax of the text to be entered. Text entry can be concluded by pressing F2, or for single-line fields, by pressing the enter key. Additionally, editing can be concluded by clicking somewhere in the diagram outside the edit area.

Editing notation on the diagram is a very powerful way to enter a lot of model-information in a very compact way. It is e.g. possible to create an operation, its stereotype, all parameters and their types, and operation properties (visibility, concurrency), all at once by typing:

        	+Order(customerID : int,items : List) : void {sequential}
        

An association (e.g. between two classes) is showing many texts close to its middle and ends, so it deserves some extra explanation. Figure 12.9, “ A couple of associations with adornments shows two associations to clarify the following:

Figure 12.9.  A couple of associations with adornments

A couple of associations with adornments

The association on the right shows that invisible fields where text can be entered become visible once the modelelement is selected. The fields are indicated by blue rectangles - double-click on them with mouse button 1 to start editing.

The visibility (the +, -, # or ~) is shown together with the association-end name, but it is not shown for an unnamed association end.

Likewise, the multiplicity is not shown if it is 1, unless the setting Show "1" multiplicities in the menu File=>Properties is checked.

The example figure does not demonstrate this, but stereotypes of an association are shown on the diagram, but are not editable. And stereotypes of association-ends are shown together with the association-end name.

12.11.3.  Notation Parsing

(to be written)

Chapter 13.  The Details Pane

13.1.  Introduction

Figure 13.1, “ Overview of the details pane shows the ArgoUML window, with the details pane highlighted.

Figure 13.1.  Overview of the details pane

Overview of the details pane

For any model element within the system, this pane is where all its associated data is viewed and entered.

The Pane has a series of tabs at the top, which are selected by button 1 click The body of a tab is a menu of items to be checked, selected or entered specific to the particular tab selected.

Of these, the Properties Tab is by far the most complex, with a different presentation for each model element within the system. The detailed descriptions of the properties tab for each model element are the subject of separate chapters covering the model elements that may appear on the various diagrams (see Chapter 16, Top Level Model Element Reference through Chapter 23, Deployment Diagram Model Element Reference ).

13.2.  To Do Item Tab

This tab provides control over the various to-do items created by the user, or raised automatically by the ArgoUML critics (discussed in more detail in the section on the Critique menu—see Section 10.9, “ The Critique Menu). Figure 13.2, “ Example of the To Do Item tab on the properties pane shows a typical pane. The to-do item is selected with button 1 in the to-do pane (see Chapter 14, The To-Do Pane ) or by using the Critiques context sensitive pop-up menu on the editing pane.

Figure 13.2.  Example of the To Do Item tab on the properties pane

Example of the To Do Item tab on the properties pane

Customization of the critics behaviour is possible through the Browse critics... menu (see Section 10.9.4, “ Browse Critics...).

The body of the tab describes the problem found by the critic and outlines how it can be fixed. To the left are four buttons.

  • New To Do Item... This launches a dialog box (see Figure 13.3, “ Dialog box for New To Do Item ), which allows you to create your own to-do item, with its own headline (which appears in the to-do pane), priority for the to-do pane, reference URL and detailed description for further information.

    Figure 13.3.  Dialog box for New To Do Item

    Dialog box for New To Do Item

  • Resolve Item... This pops up a dialog allowing the user to resolve the selected to-do item (see Figure 13.4, “ Dialog box for Resolve Item ). This is an important dialog, because it allows you to deal with to-do items in ways other than the recommendation of the to-do item (which is the whole point of their being advisory).

    This dialog box is intended to be used for the following reasons: deleting todo items that were manually created, preventing a single critic to trigger on a single object, and dismissing categories of todo items by lowering design concerns or design goals.

    Figure 13.4.  Dialog box for Resolve Item

    Dialog box for Resolve Item

    At the top are three radio-buttons, of which by default the last is selected, labeled 1) It is not relevant to my goals, 2) It is not of concern at the moment, and 3) Reason given below. If you choose the third of these you should enter a reason in the main text box.

    [Tip]Tip

    If you wish to resolve a to-do item (that is generated by a critic) by following its recommendations, just make the recommended changes and the to-do item will disappear of its own accord. There is no need to use this dialog.

    [Warning]Warning

    The V0.20 version of ArgoUML implementation is incomplete: The reason given is not stored when the project is saved. And there is no way to retrieve todo items that were resolved. So, it is not usefull to give a reason at all.

    When a todo item generated by a critic is resolved, then there is no way to undo this (unless by re-creating the object that triggered the critic).

  • Snooze Critic This suspends the activity of the critic that generated the current to-do item. The to-do item (and all others generated by the critic) will disappear from the to-do pane.

    The critic will wake up after a period of time. Initially this period is 10 minutes, but it doubles on each successive application of the Snooze button. The critic can be awakened explicitly through the Critique > Browse Critics... menu (see Section 10.9.4, “ Browse Critics...).

    [Tip]Tip

    Some common critics can fire the whole time as you build a big diagram. Some users find it useful to snooze these critics until the diagram has been completed.

13.2.1.  Wizards

Some of the more common critics have a “wizard” available to help in fixing the problem. The wizard comprises a series of pages (one or more) in the ToDo Item tab that step you through the changes. Start the wizard by clicking the Next> button.

Figure 13.5.  Example of a Wizard

Example of a Wizard

The wizard is driven through the first three buttons at the bottom of the ToDo Item tab.

  • <Back. This will take you back to the previous step in the wizard. Grayed out if this is the first step.

  • Next>. This will take you back to the next step in the wizard. Grayed out if this is the last step.

  • Finish. This will commit the changes you have made through the wizard in previous steps, and/or use the defaults for all next steps.

[Note]Note

Not all to-do items have wizards. If there is no wizard all three buttons will remain grayed out.

The ArgoUML wizards are non-modal, i.e. once started, you may select other todo items, or do some other actions, and all the while the wizard will remeber where it was, so if you return to the todo item, the wizard will indicate the same step it was on when you left it.

13.2.2.  The Help Button

There is one remaining button at the bottom of the To Do Item tab, labeled Help. This will fire up a browser to a URL with further help.

13.3.  Properties Tab

Through this tab, the properties of model elements selected in the explorer or editing pane may be set. The properties of an model element may be displayed in one of the following ways:

  1. Selection of the model element in the explorer or editing panes, followed by selection of the properties tab in the details pane; or

  2. Navigation buttons cause different model elements to be selected. I.e. the Go Up button on the properties tab, the Navigate Back and Navigate Forward buttons in the main tool bar, and the various menu-items under Edit - Select.

Figure 13.6, “ A typical Properties tab on the details pane shows a typical properties tab for a model element in ArgoUML (in this case a class).

Figure 13.6.  A typical Properties tab on the details pane

A typical Properties tab on the details pane

At the top left is the icon and name of the type of model element (i.e. the UML metaclass, not the actual name of this particular model element). In this example the property tab is for a class.

To the right of this is a toolbar of icons relevant to this property tab. The first one is always navigation Go up. The last is always Delete to delete the selected model element from the model. The ones in between depend on the model element.

The remainder of the tab comprises fields, laid out in two or three columns. Each field has a label to its left. The fields may be text boxes, text areas, drop down selectors, radio boxes and check boxes. In most (but not all cases) the values can be changed. In the case of text boxes this is sometimes by just typing the required value.

However for many text boxes and text areas, data entry is via a context sensitive pop-up menu (using button 2 click), which offers options to add a new entry, delete an entry or move entries up and down (in text areas with multiple entries).

The first field is almost always a text field Name, where the name of the specific model element can be entered. The remaining fields vary depending on the model element selected.

The detailed property sheets for all ArgoUML model elements are discussed in separate chapters for each of the diagram types (use case diagram (Chapter 17, Use Case Diagram Model Element Reference , class diagram (Chapter 18, Class Diagram Model Element Reference , sequence diagram (Chapter 19, Sequence Diagram Model Element Reference , statechart diagram ( Chapter 20, Statechart Diagram Model Element Reference , collaboration diagram ( Chapter 21, Collaboration Diagram Model Element Reference , activity diagram ( Chapter 22, Activity Diagram Model Element Reference , deployment diagram ( Chapter 23, Deployment Diagram Model Element Reference ). Property sheets for model elements that are common to all diagram types have their own chapter (Chapter 16, Top Level Model Element Reference ).

[Caution]Caution

ArgoUML will always try to squeeze all fields on to the property sheet. If the size of the property tab is too small, it may become unusable. The solution is to either enlarge the property tab by enlarging the main window, or by moving the dividers to left and top.

13.4.  Documentation Tab

Within the UML 1.4 standard, all model elements are children of the Element metaclass. The Element metaclass defines a tagged value documentation for comment, description or explanation of the element to which it is attached. Since this tagged value applies to every model element, it is given its own tab in the details pane, rather than being part of the Tagged Values tab.

Figure 13.7, “ A typical Documentation tab on the details pane shows a typical documentation tab for a model element in ArgoUML.

Figure 13.7.  A typical Documentation tab on the details pane

A typical Documentation tab on the details pane

As you can see, many more fields have been added to the Documentation field alone. The other fields similarly store their information under tagged values: author, version, since, deprecated, see.

The fields on this tab are the same for all model elements.

Since UML comments are a kind of documentation, they are also shown on this tab, with name and body.

  • Author: A text box for the author of the documentation.

  • Version: A text box for the version of the documentation.

  • Since: A text box to show how long the documentation has been valid.

  • Deprecated: A check box to indicate whether this model element is deprecated (i.e. planned for removal in future versions of the design model).

  • See: Pointers to documentation outside the system.

  • Documentation: Literal text of any documentation.

  • Comment Name: The names of all comments attached to the modelelement.

  • Body: The bodies of all comments attached to this modelelement.

[Tip]Tip

ArgoUML is not primarily a documentation system. For model elements that require heavy documentation, notably use cases, the use of the See: field to point to external documents is more practical.

13.5.  Presentation Tab

This tab provides some limited control over the graphical representation of model elements in the diagram in the editing pane.

Model elements that do not have any specific direct graphical representation on the screen (beyond their textual description) do not have style tabs of their own. For example the style sheet of an operation on a class will be downlighted.

Style sheets vary a little from model element to model element, but Figure 13.8, “ A typical Presentation tab on the details pane shows a typical style tab for a model element in ArgoUML (in this case a class).

Figure 13.8.  A typical Presentation tab on the details pane

A typical Presentation tab on the details pane

There may be further fields in some cases, e.g. for a package, but most fields are common to many model elements.

  • Path This checkbox allow to display or hide the path in front of the name of the modelelement. It is shown in UML notation with :: seperators. E.g. the ArgoUML Main class would be shown as: org::argouml::application::Main.

  • Attributes This checkbox allows to hide or show the attributes compartment of a class.

  • Operation This checkbox allows to hide or show the operations compartment of a class or interface.

  • Stereotype This checkbox allows to reveal or hide the stereotypes of a package, shown above the name.

  • Visibility This checkbox allows to hide the visibility of a package. The visibility is shown in UML notation as +, -, # or ~.

  • Extension Points This checkbox allows to reveal or hide the extensions points compartment of a usecase.

  • Bounds: This defines the corners of the bounding box for a 2D model element. It comprises four numbers separated by commas. These four numbers are respectively: i) the X coordinate of the upper left corner of the box; ii) the Y coordinate of the upper left corner of the box; iii) the width of the box; and iv) the height of the box. All units are pixels on the editing pane.

    This field has no effect on 1D model elements that link other model elements (associations, generalizations etc), since their position is constrained by their connectedness. In this case the field is downlighted.

  • Fill: This drop-down selector specifies the fill color for 2D model elements. It is not present for line model elements. Selecting No Fill makes the model element transparant. Selecting Custom allows to create other colors then the ones listed. It causes the color selector dialog box to appear, see Figure 13.9, “ The Custom Fill/Line Color dialog box .

  • Line: This drop-down selector specifies the line color for model elements. Selecting No Fill makes the model element transparant. Selecting Custom allows to create other colors then the ones listed. It causes the color selector dialog box to appear, see Figure 13.9, “ The Custom Fill/Line Color dialog box .

Figure 13.9.  The Custom Fill/Line Color dialog box

The Custom Fill/Line Color dialog box

Figure 13.10.  The Custom Fill/Line Color dialog box

The Custom Fill/Line Color dialog box

Figure 13.11.  The Custom Fill/Line Color dialog box

The Custom Fill/Line Color dialog box

13.6.  Source Tab

This tab shows the source code that will be generated for this model element, in the selected language. ArgoUML generates the code e.g. for classes and interfaces. The code shown here, may be saved in the indicated files with the aid of the functions in the Generation menu.

Figure 13.12.  The Source Tab of a class.

The Source Tab of a class.

Any code you add will be lost - that is not the intention of ArgoUML - use an IDE instead.

The dropdown at the right allows selection of the output file. This function is not very useful for languages that generate all code for a class within one file, but serves its purpose for e.g. C++, where a .h and .cpp file are generated. See the figure below.

Figure 13.13.  A C++ example.

A C++ example.

13.7.  Constraints Tab

Constraints are one of the extension mechanisms provided for UML. ArgoUML is equipped with a powerful constraint editor based on the Object Constraint Language (OCL) defined in the UML 1.4 standard.

[Caution]Caution

The OCL editor implementation for ArgoUML V0.26 doesn't support OCL constraints for elements other than Classes and Features.

This is something of a general restriction of OCL. Although the UML specification claims that there may be a constraint for every model element, the OCL specification only defines classes/interfaces and operations as allowable contexts.

It is not before OCL 2.0 that a more general definition of allowable contexts is introduced. The key issue is that for each context definition you need to define what is the contextualClassifier, i.e., the classifier that will be associated with the self keyword. The creators of the OCL specification claim that this is not an issue for the OCL specification, but rather for UML or some integration task force. Conversely, it seems that the UML specification people seem to expect this to be defined in the OCL specification (which is why we did a first step in that direction in OCL 2.0).

So, to cut a long story short, it appeared that the simplest solution for ArgoUML at the moment would be to enable the OCL property panel only for those model elements for which there actually exists a definition of the contextualClassifier in OCL 1.4. These are (s. above) Class/Interface and Feature.

The standard pre-defines a small number of constraints (for example the xor constraint over a set of associations indicating that only one may be manifest for any particular instance).

The standard also envisages a number of circumstances where general purpose constraints may be useful:

  • To specify invariants on classes and types in the class model;

  • To specify type invariants for stereotypes;

  • To describe pre-conditions and post-conditions on operations and methods;

  • To describe guards;

  • As a navigation language; and

  • To specify constraints on operations.

Figure 13.14, “ A typical Constraints tab on the details pane shows a typical constraint tab for a model element in ArgoUML (in this case a class).

Figure 13.14.  A typical Constraints tab on the details pane

A typical Constraints tab on the details pane

Along the top of the tab are a series of icons.

  • New Constraint. This creates a new constraint and launches the constraint editor in the Constraints tab for that new constraint (see Section 13.7.1, “ The Constraint Editor). The new constraint is created with a context declaration for the currently selected model element.

    [Warning]Warning

    It seems logical, that when a new constraint is created, it needs to be edited. But ArgoUML V0.26 fails to start the OCL editor upon creation; you have to do this by primo selecting the new constraint first, secundo rename it, and tertio press the Edit Constraint button. It is essental for successfully creating a constraint to follow these 4 steps accurately: create, select, rename, edit. The step to rename is necessary, because the validity check will refuse the constraint if its name differs from the name mentioned in the constraint text. For the same reason, renaming a constraint afterwards is impossible.

  • Delete Constraint. The constraint currently selected in the Constraint Name box (see below) is deleted.

    [Caution]Caution

    In V0.26 of ArgoUML this button is not downlighted when it is not functional, i.e. when no constraint is selected.

  • Edit Constraint. This launches the constraint editor in the Constraints tab (see Section 13.7.1, “ The Constraint Editor). The editor is invoked on the constraint currently selected in the Constraint Name box.

    [Caution]Caution

    In V0.26 of ArgoUML this button is not downlighted when it is not functional, i.e. when no constraint is selected.

  • Configure Constraint Editor. This a dialog to configure options in the constraint editor (see Figure 13.15, “ Dialog box for configuring constraints ).

    Figure 13.15.  Dialog box for configuring constraints

    Dialog box for configuring constraints

    The dialog box has a check box for the following option.

    • Check type conformance of OCL constraints. OCL is strictly typed. At the early stages of design it may be helpful to disable type checking, rather than follow through all the detailed specification needed to get type consistency.

    At the bottom are two buttons, labeled OK (to accept the option changes) and Cancel (to discard the changes).

The main body of the constraints tab comprises two boxes, a smaller to the left and a larger one to the right. The two are separated by two small arrow buttons which control the size of the boxes.

  • Shrink Left. Button 1 click on this icon shrinks the box on the left. Its effect may be reversed by use of the Shrink Right button (see below).

  • Shrink Right. Button 1 click on this icon shrinks the box on the right. Its effect may be reversed by use of the Shrink Left button (see above).

Finer control can be achieved by using button 1 motion to drag the dividing bar to left and right.

The box on the left is titled Constraint Name and lists all the constraints (if any) so far defined for the selected model element. A constraint may be selected by button 1 click.

The box on the right is labeled Preview and contains the text of the constraint. This box only shows some contents if a constraint is selected. Where a constraint is too large for the box, a scroll bar is provided to the right.

13.7.1.  The Constraint Editor

This is invoked through the use of the Edit Constraint button on the main Constraints tab. The constraint editor takes up the whole tab (see Figure 13.16, “ Dialog box for configuring constraints).

Figure 13.16.  Dialog box for configuring constraints

Dialog box for configuring constraints

Along the top of the tab are a series of icons.

  • Cancel Edit Constraint. This exits the constraint editor without saving any changes and returns to the main Constraints tab.

  • Check OCL Syntax. This button invokes a full syntax check of the OCL written in the editor. If the syntax is valid, the constraint is saved, and control returns to the main Constraints tab. If the syntax is not valid, a dialog box explains the problem.

    [Warning]Warning

    Whether type checking is included should be configurable with the Configure Constraint Editor button (see below). But ArgoUML V0.20 does always check, and refuses to accept any constraint with the slightest error.

  • Configure Constraint Editor. This a dialog to configure options in the constraint editor. It is also available in the main Constraints tab and is discussed in detail there (see Section 13.7, “ Constraints Tab ).

To the right of the toolbar is a check box labeled Syntax Assistant (unchecked by default), which will enable the syntax assistant in the constraint editor.

If the syntax assistant is enabled, six drop down menus are provided in a row immediately below the toolbar. These provide standard templates for OCL that, when selected, will be inserted into the constraint being edited.

The syntax assistant can be made floating in a separate window by button 1 motion on the small divider area to the left of the row of drop-down menus.

  • General. General OCL constructors. Entries: inv (inserts an invariant); pre (inserts a pre-condition); post (inserts a post-condition); self (inserts a self-reference); @pre (inserts a reference to a value at the start of an operation); and result (inserts a reference to a previous result).

  • Basic Operators. Relational operators and parentheses. Entries: =; <>; <; >; <=; >=; and ().

  • Numbers. Arithmetic operators and functions. Entries: +; -; *; /; mod; div; abs; max; min; round; and floor.

  • Strings. String functions. Entries: concat; size; toLower; toUpper; and substring.

  • Booleans. Logical functions. Entries: or; and; xor; not; implies; and if then else.

  • Collections. Operators and functions on collections—bags, sets and sequences. The large number of functions are organized into sub-groups.

    • General. Functions that apply to all types of collection. Entries: Collection {} (insert a new collection); Set {} (insert a a new set); Bag {} (insert a new bag); Sequence {} (insert a new sequence); size; count; isEmpty; notEmpty; includes; includesAll; iterate; exists; forAll; collect; select; reject; union; intersection; including; excluding; and sum.

    • Sets. Operators and functions that apply only to sets. Entries: - (set difference); and symmetricDifference.

    • Sequences. Functions that apply to sequences. Entries: first; last; at; append; prepend; and subSequence.

The remainder of the tab comprises a writable text area containing the text to be edited. The mouse buttons have their standard behavior within an editable text area (see Section 8.2, “ General Mouse Behavior in ArgoUML).

In addition, cut, copy and paste operations may be invoked through the keyboard shortcuts Ctrl-X, Ctrl-C and Ctrl-V respectively.

13.8.  Stereotype Tab

This tab shows the available and applied stereotypes for the currently selected modelelement. It consists of 2 panels and 2 buttons. The buttons allow to move the stereotypes from one list to the other.

Figure 13.17.  An example of a stereotype tab for a class.

An example of a stereotype tab for a class.

In the lists, between [] the baseclass of the stereotypes is shown. E.g. in the figure above, the thread [Classifier] stereotype may be applied to all types of classifiers, such as Class, UseCase,...

13.9.  Tagged Values Tab

Tagged values are another extension mechanism provided by UML. The user can define name-value pairs to be associated with model elements which define properties of that model element. The names are known as tags. UML pre-defines a number of tags that are useful for many of its model elements.

[Note]Note

The tag documentation is defined for the top UML metaclass, Element and is so available to all model elements. In ArgoUML documentation values are provided through the Documentation tab, rather than by using the Tagged Values tab.

The Tagged Values tab in ArgoUML comprises a two column table, with a combo-box on the left to select the tagdefinition and an editable box on the right for the associated value. There is always at least one empty row available for any new tag.

The button at the top of this tab allows creation of a new tagdefinition. After clicking this button, go to the properties tab first to set the name of the new tagdefinition.

The mouse buttons have their standard behavior within the editable value area (see Section 8.2, “ General Mouse Behavior in ArgoUML). In addition, when in the value field, cut, copy and paste operations may be invoked through the keyboard shortcuts Ctrl-X, Ctrl-C and Ctrl-V respectively.

13.10.  Checklist Tab

Conducting design reviews and inspections is one of the most effective ways of detecting errors during software development. A design review typically consists of a small number of designers, implementers, or other project stakeholders holding a meeting to review a software development artifact. Many development organizations have developed checklists of common design problems for use in design review meetings. Recent research indicated that reviewers inspecting code without meeting, makeing use of these checklists, are just as effective as design review meetings.

Hence, a checklist feature has been added to ArgoUML, that is much in the spirit of design review checklists. However, ArgoUML's checklists are integrated into the design tool user interface and the design task.

A software designer using ArgoUML can see a review checklist for any design element. The “Checklist” tab presents a list of check-off items that is appropriate to the currently selected design element. For example, when a class is selected in a design diagram, the checklist tab shows items that prompt critical thinking about classes. See the figure below. Designers may check off items as they consider them. Checked items are kept in the list to show what has already been considered, while unchecked items prompt the designer to consider new design issues. ArgoUML supplies many different checklists with many possible items.

Figure 13.18.  An example of a checklist for a class.

An example of a checklist for a class.

[Caution]Caution

In the V0.22 release of ArgoUML, this tab is not completely implemented. E.g. the checks are not saved.

Chapter 14.  The To-Do Pane

14.1.  Introduction

Figure 14.1, “ Overview of the to-do pane shows the ArgoUML window with the to-do pane highlighted.

Figure 14.1.  Overview of the to-do pane

Overview of the to-do pane

This pane provides access to the advice that comes from the critics processes running within ArgoUML.

A selector box at the top allows a choice of how the data is presented, a button allows the display of the hierarchy to be changed, and there is an indicator of the number of to-do items identified.

More information on critics can be found in the discussion of the Critique menu (see Section 10.9, “ The Critique Menu).

14.2.  Mouse Behavior in the To-Do Pane

Behavior of the mouse in general, and the naming of the buttons is covered in the chapter on the overall user interface (see Chapter 8, Introduction ).

14.2.1.  Button 1 Click

This action is generally used to select an item for subsequent operations.

Within the hierarchical display, elements which have sub-hierarchies may be indicated by when the hierarchy is hidden and when the hierarchy is open.

When these icons are displayed, the display of the hierarchy is toggled by button 1 click on these icons.

Button 1 click over the headline of any to-do item will cause its details to be shown in the To Do Item tab of the details pane. That tab is automatically selected if it is not currently visible.

14.2.2.  Button 1 Double Click

When applied to the folder icon alongside a hierarchy category, this will cause the display of that hierarchy to be toggled.

When applied to a headline, button 1 double click will show the diagram for the model element to which the to-do item applies in the editing pane and select the model element on the diagram using an appropriate clarifier (the model element may be highlighted, underlined with a wavy line or surrounded by a colored box as appropriate).

14.2.3.  Button 2 Actions

There are no button 2 functions in the to-do pane.

14.2.4.  Button 2 Double Click

There are no button 2 functions in the to-do pane.

14.3.  Presentation Selection

At the top of the pane is a drop-down selector controlling how the to-do items are presented. The to-do items may be presented in six different ways. This setting is not stored persistently, i.e. it is on its default vallue when ArgoUML is started.

  • By Priority. This is the default setting. The to-do items are organized into three hierarchies by priority: High, Medium and Low. The priority associated with the to-do items generated by a particular critic may be altered through the Critique > Browse Critics... menu (see Section 10.9.4, “ Browse Critics...).

  • By Decision. The to-do items are organized into 17 hierarchies by design issue: Uncategorized, Class Selection, Behavior, Naming, Storage, Inheritance, Containment, Planned Extensions, State Machines, Design Patterns, Relationships, Instantiation, Modularity, Expected Usage, Methods, Code Generation and Stereotypes. The details of the critics in each category are discussed in Section 10.9.2, “ Design Issues....

  • By Goal. ArgoUML has a concept that critics may be grouped according to the user goals they affect. This presentation groups the to-do items into hierarchies by goal.

    [Caution]Caution

    In the current release of ArgoUML there is only one goal, Unspecified and all to-do items will appear under this heading.

  • By Offender. The to-do items are organized into a hierarchy according to the model element that caused the problem. Todo items that were manually created with the "New ToDo item" button (i.e. not by a critic), are not listed here.

  • By Poster. The to-do items are organized into a hierarchy according to which critic generated the to-do item. The class name of the critic is listed instead of just its headline name since the former is guaranteed to be a unique name.

  • By Knowledge Type. ArgoUML has the concept that a critic reflects a deficiency in a category of knowledge. This presentation option groups the critics according to their knowledge category: Designer's, Correctness, Completeness, Consistency, Syntax, Semantics, Optimization, Presentational, Organizational, Experiencial and Tool. The former category (Designer's) contains the manually entered todo items.

14.4.  Item Count

To the right of the flat/hierarchical button is a count of the number of to-do items currently found. It will be highlighted in yellow when the number of to-do items grows above 50 todo items, and red when above 100.

Chapter 15.  The Critics

15.1.  Introduction

The key feature that distinguishes ArgoUML from other UML CASE tools is its use of concepts from cognitive psychology. The theory behind this is well described in Jason Robbins' PhD dissertation http://argouml.tigris.org/docs/robbins_dissertation/.

Critics are one of the main ways in which these ideas are implemented. Running in the background they offer advice to the designer which may be accepted or ignored. A key point is that they do not impose a decision on the designer.

[Note]Note

The critics are asynchronous processes that run in parallel with the main ArgoUML tool. Changes typically take a second or two to propagate as the critics wake up.

15.1.1.  Terminology

The critics are background processes, which evaluate the current model according to various “good” design criteria. There is one critic for every design criterion.

The output of a critic is a critique—a statement about some aspect of the model that does not appear to follow good design practice.

Finally a critique will generally suggest how the bad design issue it has identified can be rectified, by raising a to-do item.

15.1.2.  Design Issues

ArgoUML categorizes critics according the the design issue they address (some critics may be in more than one category). At present there are 16 such categories.

Within this manual the descriptions of critics are grouped in sections by design issue.

15.2.  Uncategorized

These are critics that do not fit into any other category.

ArgoUML has no critics in this category. Maybe some will be added in later versions.

15.3.  Class Selection

These are critics concerning how classes are chosen and used.

ArgoUML has the following critics in this category.

15.3.1.  Wrap DataType

DataTypes are not full classes within UML 1.4. They can only have enumeration literals as values, and only support query operations (that is operations that do not change the DataType's state).

DataTypes cannot be associated with classes, unless the DataType is part of a composite (black diamond) aggregation. Such an association relects the tight binding of a collection of DataType instances to a class instance. In effect such a DataType is an attribute of the class with multiplicity.

Good OOA&D depends on careful choices about which entities to represent as full objects and which to represent as attributes of objects.

There are two options to fix this problem.

  • Replace the DataType with a full class.

  • or change the association aggregation to composite relationship at the DataType end.

15.3.2.  Reduce Classes in namespace <namespace>

Suggestion to improve understandability by having fewer classes in one namespace. If one namespace (such as the model, a package, or a class) has too many classes it may become very difficult for humans to understand. Defining an understandable set of namespaces is an important part of your design.

The Wizard of this critic allows setting of the treshold, i.e. the maximum number of classes allowed before this critic fires.

[Caution]Caution

This number is not stored persistently, and there is no way to reduce it after it has been set higher, except by creating more classes until the critic fires again. Restarting ArgoUML resets this number to its default: 20.

15.3.3.  Clean Up Diagram

Suggestion that the diagram could be improved by moving model elements that are overlapping.

15.4.  Naming

These are critics concerning the naming of model elements. The current version of ArgoUML has 18 critics in this category.

15.4.1.  Resolve Association Name Conflict

Suggestion that two association names in the same namespace have the same name. This is not permitted in UML.

15.4.2.  Revise Attribute Names to Avoid Conflict

Suggestion that two attribute names of a class have the same name. This is not permitted in UML.

[Note]Note

The problem may be caused by inheritance of an attribute through a generalization relationship.

15.4.3.  Change Names or Signatures in a model element

Two operations in <model element> have the same signature. This means their name is the same, and the list of parameters has the same type.

Where there are conflicting signatures, correct code cannot be generated for mainstream OO languages. It also leads to very unclear semantics of the design.

In comparing signatures, this critic considers:

  1. the name;

  2. the list of in, out and in-out parameter types in order; and

Only if these all match in both type and order, will the signatures be considered as the same.

This follows the line of Java/C++ in ignoring the return parameters for the signature. This may be unsatisfactory for some functional OO languages.

[Note]Note

Some purists would argue that the comparison should really differentiate between in, out and in-out parameters. However no practical programming language can do this when resolving an overloaded method invocation, so this critic lumps them all together.

15.4.4.  Duplicate End (Role) Names for an Association

The specified association has two (or more) ends (roles) with the same name. One of the well-formedness rules in UML 1.4 for associations, is that all end (role) names must be unique.

This ensures that there can be unambiguous reference to the ends of the association.

To fix this, manually select the association and change the names of one or more of the offending ends (roles) using the button 2 pop-up menu or the property sheet.

15.4.5.  Role name conflicts with member

A suggestions that good design avoids role names for associations that clash with attributes or operations of the source class. Roles may be realized in the code as attributes or operations, causing code generation problems.

15.4.6.  Choose a Name (Classes and Interfaces)

The class or interface concerned has been given no name (it will appear in the model as Unnamed). Suggestion that good design requires that all interfaces and classes are named.

15.4.7.  Name conflict in a namespace

The namespace (e.g. package) concerned contains a class or interface specified that has the same name as another (in the same namespace), which is bad design and will prevent valid code generation.

15.4.8.  Choose a Unique Name for a model element (Classes and Interfaces)

Suggestion that the class or interface specified has the same name as another (in the namespace), which is bad design and will prevent valid code generation.

15.4.9.  Choose a Name (Attributes)

The attribute concerned has been given no name (it will appear in the model as (Unnamed Attribute)). Suggestion that good design requires that all attributes are named.

15.4.10.  Choose a Name (Operations)

The operation concerned has been given no name (it will appear in the model as (Unnamed Operation)). Suggestion that good design requires that all operations are named.

15.4.11.  Choose a Name (States)

The state concerned has been given no name (it will appear in the model as (Unnamed State)). Suggestion that good design requires that all states are named.

15.4.12.  Choose a Unique Name for a (State related) model element

Suggestion that the state specified has the same name as another (in the current statechart diagram), which is bad design and will prevent valid code generation.

15.4.13.  Revise Name to Avoid Confusion

Two names in the same namespace have very similar names (differing only by one character). Suggestion this could potentially lead to confusion.

[Caution]Caution

This critic can be particularly annoying, since at times it is useful and good design to have a series of model elements var1, var2 etc.

It is important to remember that critics offer guidance, and are not always correct. ArgoUML lets you dismiss the resulting to-do items through the to-do pane (see Chapter 14, The To-Do Pane ).

15.4.14.  Choose a Legal Name

All model element names in ArgoUML must use only letters, digits and underscore characters. This critic suggests an entity has not met this requirement.

15.4.15.  Change a model element to a Non-Reserved Word

Suggestion that this model element's name is the same as a reserved word in UML (or within one character of one), which is not permitted.

15.4.16.  Choose a Better Operation Name

Suggestion that an operation has not followed the naming convention that operation names begin with lower case letters.

[Caution]Caution

Following the Java and C++ convention most designers give their constructors the same name as the class, which begins with an upper case character. In ArgoUML, this will trigger this critic, unless the constructor is stereotyped «create».

It is important to remember that critics offer guidance, and are not always correct. ArgoUML lets you dismiss the resulting to-do items through the to-do pane (see Chapter 14, The To-Do Pane ).

15.4.17.  Choose a Better Attribute Name

Suggestion that an attribute has not followed the naming convention that attribute names begin with lower case letters.

15.4.18.  Capitalize Class Name

Suggestion that a class has not followed the naming convention that classes begin with upper case letters.

[Note]Note

Although not triggering this critic, the same convention should apply to interfaces.

15.4.19.  Revise Package Name

Suggestion that a package has not followed the naming convention of using lower case letters with periods used to indicated sub-packages.

15.5.  Storage

Critics concerning attributes of classes.

The current version of ArgoUML has the following critics in this category.

15.5.1.  Revise Attribute Names to Avoid Conflict

This critic is discussed under an earlier design issues category (see Section 15.4.2, “ Revise Attribute Names to Avoid Conflict ).

15.5.2.  Add Instance Variables to a Class

Suggestion that no instance variables have been specified for the given class. Such classes may be created to specify static attributes and methods, but by convention should then be given the stereotype «utility».

15.5.3.  Add a Constructor to a Class

You have not yet defined a constructor for class class. Constructors initialize new instances such that their attributes have valid values. This class probably needs a constructor because not all of its attributes have initial values.

Defining good constructors is key to establishing class invariants, and class invariants are a powerful aid in writing solid code.

To fix this, add a constructor manually by clicking on class in the explorer and adding an operation using the context sensitive pop-up menu in the property tab, or select class where it appears on a class diagram and use the Add Operation tool.

In the UML 1.4 standard, a constructor is an operation with the stereotype «create». Although not strictly standard, ArgoUML will also accept «Create» as a stereotype for constructors.

By convention in Java and C++ a constructor has the same name as the class, is not static, and returns no value. ArgoUML will also accept any operation that follows these conventions as a constructor even if it is not stereotyped «create».

[Caution]Caution

Operators are created in ArgoUML with a default return parameter (named return). You will need to remove this parameter to meet the Java/C++ convention.

15.5.4.  Reduce Attributes on a Class

Suggestion that the class has too many attributes for a good design, and is at risk of becoming a design bottleneck.

The Wizard of this critic allows setting of the treshold, i.e. the maximum number of attributes allowed before this critic fires.

[Caution]Caution

This number is not stored persistently, and there is no way to reduce it after it has been set higher, except by creating more attributes until the critic fires again. Restarting ArgoUML resets this number to its default: 7.

15.6.  Planned Extensions

Critics concerning interfaces and subclasses.

[Note]Note

It is not clear why this category has the name “Planned Extensions”.

The current version of ArgoUML has three critics in this category.

15.6.1.  Operations in Interfaces must be public

Suggestion that there is no point in having non-public operations in Interfaces, since they must be visible to be realized by a class.

15.6.2.  Interfaces may only have operations

Suggestion that an interfaces has attributes defined. The UML standard defines interfaces to have operations.

[Caution]Caution

ArgoUML does not allow you to add attributes to interfaces, so this should never occur in the ArgoUML model. It might trigger if a project has been loaded with XMI created by another tool.

15.6.3.  Remove Reference to Specific Subclass

Suggestion that in a good design, a class should not reference its subclasses directly through attributes, operations or associations.

15.7.  State Machines

Critics concerning state machines.

ArgoUML has the following critics in this category.

15.7.1.  Reduce Transitions on <state>

Suggestion given state is involved in so many transitions it may be a maintenance bottleneck.

The Wizard of this critic allows setting of the treshold, i.e. the maximum number of transitions allowed before this critic fires.

[Caution]Caution

This number is not stored persistently, and there is no way to reduce it after it has been set higher, except by creating more transition until the critic fires again. Restarting ArgoUML resets this number to its default: 10.

15.7.2.  Reduce States in machine <machine>

Suggestion that the given state machine has so many states as to be confusing and should be simplified (perhaps by breaking into several machines, or using a hierarchy).

The Wizard of this critic allows setting of the treshold, i.e. the maximum number of states allowed before this critic fires.

[Caution]Caution

This number is not stored persistently, and there is no way to reduce it after it has been set higher, except by creating more states until the critic fires again. Restarting ArgoUML resets this number to its default: 20.

15.7.3.  Add Transitions to <state>

Suggestion that the given state requires both incoming and outgoing transitions.

15.7.4.  Add Incoming Transitions to <model element>

Suggestion that the given state requires incoming transitions.

15.7.5.  Add Outgoing Transitions from <model element>

Suggestion that the given state requires outgoing transitions.

15.7.6.  Remove Extra Initial States

Suggestion that there is more than one initial state in the state machine or composite state, which is not permitted in UML.

15.7.7.  Place an Initial State

Suggestion that there is no initial state in the state machine or composite state.

15.7.8.  Add Trigger or Guard to Transition

Suggestion that a transition is missing either a trigger or guard, one at least of which is required for it to be taken.

15.7.9.  Change Join Transitions

Suggestion that the join pseudostate has an invalid number of transitions. Normally there should be one outgoing and two or more incoming.

15.7.10.  Change Fork Transitions

Suggestion that the fork pseudostate has an invalid number of transitions. Normally there should be one incoming and two or more outgoing.

15.7.11.  Add Choice/Junction Transitions

Suggestion that the branch (choice or junction) pseudostate has an invalid number of transitions. Normally there should be at least one incoming transition and at least one outgoing transition.

15.7.12.  Add Guard to Transition

Suggestion that the transition requires a guard.

[Caution]Caution

It is not clear that this is a valid critic. It is perfectly acceptable to have a transition without a guard—the transition is always taken when the trigger is invoked.

15.7.13.  Clean Up Diagram

This critic is discussed under an earlier design issues category (see Section 15.3.3, “ Clean Up Diagram ).

15.7.14.  Make Edge More Visible

Suggestion that an edge model element such as an association or abstraction is so short it may be missed. Move the connected model elements apart to make the edge more visible.

15.7.15.  Composite Association End with Multiplicity >1 Zusammengesetztes Assoziationsende mit der Kardinalität >1

An instance may not belong by composition to more than one composite instance. You must change the multiplicity at the composite end of the association to either 0..1 or 1..1 (1) for your model to make sense.

Remember that composition is the stronger aggregation kind and aggregation is the weaker. The problem can be compared to a model where a finger can be an integral part of several hands at the same time.

This is the second well-formedness rule on AssociationEnd in UML 1.4.

15.8.  Design Patterns

Critics concerning design pattern usage in ArgoUML.

These relate to the use of patterns as described by the so called “Gang of Four”. ArgoUML also uses this category for critics associated with deployment and sequence diagrams. The current version of ArgoUML has the following critics in this category.

15.8.1.  Consider using Singleton Pattern for <class>

The class has no non-static attributes nor any associations that are navigable away from instances of this class. This means that every instance of this class will be identical to every other instance, since there will be nothing about the instances that can differentiate them.

Under these circumstances you should consider making explicit that you have exactly one instance of this class, by using the singleton Pattern. Using the singleton pattern can save time and memory space. Within ArgoUML this can be done by using the «singleton» stereotype on this class.

If it is not your intent to have a single instance, you should define instance variables (i.e. non-static attributes) and/or outgoing associations that will represent differences bewteen instances.

Having specified class as a singleton, you need to define the class so there can only be a single instance. This will complete the information representation part of your design. To achieve this you need to do the following.

  1. You must define a static attribute (a class variable) holding the instance. This must therefore have class as its type.

  2. You must have only private constructors so that new instances cannot be made by other code. The creation of the single instance could be through a suitable helper operation, which invokes this private constructor just once.

  3. You must have at least one constructor to override the default constructor, so that the default constructor is not used to create multiple instances.

For the definition of a constructor under the UML 1.4 standard, and extensions to that definition accepted by ArgoUML see Section 15.5.3, “ Add a Constructor to a Class .

15.8.2.  Singleton Stereotype Violated in <class>

This class is marked with the «singleton» stereotype, but it does not satisfy the constraints imposed on singletons (ArgoUML will also accept «Singleton» stereotype as defining a singleton). A singleton class can have at most one instance. This means that the class must meet the design criteria for a singleton (see Section 15.8.1, “ Consider using Singleton Pattern for <class> ).

Whenever you mark a class with a stereotype, the class should satisfy all constraints of the stereotype. This is an important part of making a self-consistent and understangle design. Using the singleton pattern can save time and memory space.

If you no longer want this class to be a singleton, remove the «singleton» stereotype by clicking on the class and selecting the blank selection on the stereotype drop-down within the properties tab.

To apply the singleton pattern you should follow the directions in Section 15.8.1, “ Consider using Singleton Pattern for <class> .

15.8.3.  Nodes normally have no enclosers

A suggestion that nodes should not be drawn inside other model elements on the deployment diagram, since they represent an autonomous physical object.

15.8.4.  NodeInstances normally have no enclosers

A suggestion that node instances should not be drawn inside other model elements on the deployment diagram, since they represent an autonomous physical object.

15.8.5.  Components normally are inside nodes

A suggestion that components represent the logical entities within physical nodes, and so should be drawn within a node, where nodes are shown on the deployment diagram.

15.8.6.  ComponentInstances normally are inside nodes

A suggestion that component instances represent the logical entities within physical nodes, and so should be drawn within a node instance, where node instances are shown on the deployment diagram.

15.8.7.  Classes normally are inside components

A suggestion that classes, as model elements making up components, should be drawn within components on the deployment diagram.

15.8.8.  Interfaces normally are inside components

A suggestion that interfaces, as model elements making up components, should be drawn within components on the deployment diagram.

15.8.9.  Objects normally are inside components

A suggestion that objects, as instances of model elements making up components, should be drawn within components or component instances on the deployment diagram.

15.8.10.  LinkEnds have not the same locations

A suggestion that a link (e.g. association) connecting objects on a deployment diagram has one end in a component and the other in a component instance (since objects can be in either). This makes no sense.

15.8.11.  Set classifier (Deployment Diagram)

Suggestion that there is an instance (object) without an associated classifier (class, datatype) on a deployment diagram.

15.8.12.  Missing return-actions

Suggestion that a sequence diagram has a send or call action without a corresponding return action.

15.8.13.  Missing call(send)-action

Suggestion that a sequence diagram has a return action, but no preceding call or send action.

15.8.14.  No Stimuli on these links

Suggestion that a sequence diagram has a link connecting objects without an associated stimulus (without which the link is meaningless).

[Warning]Warning

Triggering this critic indicates a serious problem, since ArgoUML provides no mechanism for creating a link without a stimulus. It probably indicates that the diagram was created by loading a corrupt project, with an XMI file describing a link without a stimulus, possibly created by a tool other than ArgoUML.

15.8.15.  Set Classifier (Sequence Diagram)

Suggestion that there is an object without an associated classifier (class, datatype) on a sequence diagram.

15.8.16.  Wrong position of these stimuli

Suggestion that the initiation of send/call-return message exchanges in a sequence diagram does not properly initiate from left to right.

15.9.  Relationships

Critics concerning associations in ArgoUML.

The current version of ArgoUML has the following critics in this category.

15.9.1.  Circular Association

Suggestion that an association class has a role that refers back directly to itself, which is not permitted.

[Warning]Warning

This critic is meaningless in the V0.14 version of ArgoUML which does not support association classes.

15.9.2.  Make <association> Navigable

Suggestion that the association referred to is not navigable in either direction. This is permitted in the UML standard, but has no obvious meaning in any practical design.

15.9.3.  Remove Navigation from Interface via <association>

Associations involving an interface can be not be navigable in the direction from the interface. This is because interfaces contain only operation declarations and cannot hold pointers to other objects.

This part of the design should be changed before you can generate code from this design. If you do generate code before fixing this problem, the code will not match the design.

To fix this, select the association and use the Properties tab to select in turn each association end that is not connected to the interface. Uncheck Navigable for each of these ends.

The association should then appear with a stick arrowhead pointed towards the interface

When an association between a class and interface is created in ArgoUML, it is by default navigable only from the class to the interface. However, ArgoUML does not prevent to change the navigability afterwards into a wrong situation. Which will cause this critic to be triggered.

15.9.4.  Add Associations to <model element>

Suggestion that the specified model element (actor, use case or class) has no associations connecting it to other model elements. This is required for the model element to be useful in a design.

15.9.5.  Remove Reference to Specific Subclass

This critic is discussed under an earlier design issues category (see Section 15.6.3, “ Remove Reference to Specific Subclass ).

15.9.6.  Reduce Associations on <model element>

Suggestion that the given model element (actor, use case, class or interface) has so many associations it may be a maintenance bottleneck.

The Wizard of this critic allows setting of the treshold, i.e. the maximum number of associations allowed before this critic fires.

[Caution]Caution

This number is not stored persistently, and there is no way to reduce it after it has been set higher, except by creating more associations until the critic fires again. Restarting ArgoUML resets this number to its default: 7.

15.9.7.  Make Edge More Visible

This critic is discussed under an earlier design issues category (see Section 15.7.14, “ Make Edge More Visible ).

15.10.  Instantiation

Critics concerning instantiation of classifiers in ArgoUML.

The current version of ArgoUML has no critics in this category.

15.11.  Modularity

Critics concerning modular development in ArgoUML.

The current version of ArgoUML has the following critics in this category.

15.11.1.  Classifier not in Namespace of its Association

One of the well-formedness rules in UML 1.4 for associations, is that all the classifiers attached to the ends of the association should belong to the same namespace as the association.

If this were not the case, there would be no naming, by which each end could refer to all the others.

This critic is triggered when an association does not meet this criterion. The solution is to delete the association, and recreate it on a diagram, whose namespace includes those of all the attached classifiers.

[Caution]Caution

In the current implementation of ArgoUML this critic does not handle hierarchical namespaces. As a consequence it will trigger for associations where the immediate namespaces of the attached classifiers is different, even though they are part of the same namespace hierarchy.

15.11.2.  Add Elements to Package <package>

Suggestion that the specified package has no content. Good design suggests packages are created to put things in.

[Note]Note

This will always trigger when you first create a package, since you cannot create one that is not empty!

15.12.  Expected Usage

Critics concerning generally accepted good practice in ArgoUML.

The current version of ArgoUML has one critic in this category.

15.12.1.  Clean Up Diagram

This critic is discussed under an earlier design issues category (see Section 15.3.3, “ Clean Up Diagram ).

15.13.  Methods

Critics concerning operations in ArgoUML.

The current version of ArgoUML has the following critics in this category.

15.13.1.  Change Names or Signatures in <model element>

This critic is discussed under an earlier design issues category (see Section 15.4.3, “ Change Names or Signatures in a model element ).

15.13.2.  Class Must be Abstract

Suggestion that a class that inherits or defines abstract operations must be marked abstract.

15.13.3.  Add Operations to <class>

Suggestion that the specified class has no operations defined. This is required for the class to be useful in a design.

15.13.4.  Reduce Operations on <model element> Reduzieren Sie die Anzahl der Operationen im <Modellelement>

Suggestion that the model element (class or interface) has too many operations for a good design, and is at risk of becoming a design bottleneck.

The Wizard of this critic allows setting of the treshold, i.e. the maximum number of operations allowed before this critic fires.

[Caution]Caution

This number is not stored persistently, and there is no way to reduce it after it has been set higher, except by creating more operations until the critic fires again. Restarting ArgoUML resets this number to its default: 20.

15.14.  Code Generation

Critics concerning code generation in ArgoUML.

The current version of ArgoUML has one critic in this category.

15.14.1.  Change Multiple Inheritance to interfaces

Suggestion that a class has multiple generalizations, which is permitted by UML, but cannot be generated into Java code, because Java does not support multiple inheritance.

15.15.  Stereotypes

Critics concerning stereotypes in ArgoUML.

The current version of ArgoUML has no critics in this category.

15.16.  Inheritance

Critics concerning generalization and specialization in ArgoUML.

The current version of ArgoUML has the following critics in this category.

15.16.1.  Revise Attribute Names to Avoid Conflict

This critic is discussed under an earlier design issues category (see Section 15.4.2, “ Revise Attribute Names to Avoid Conflict ).

15.16.2.  Remove <class>'s Circular Inheritance

Suggestion that a class inherits from itself, through a chain of generalizations, which is not permitted.

[Caution]Caution

This critic is marked inactive by default in the current release of ArgoUML (the only one so marked). It will not trigger unless made active.

15.16.3.  Class Must be Abstract

This critic is discussed under an earlier design issues category (see Section 15.13.2, “ Class Must be Abstract ).

15.16.4.  Remove final keyword or remove subclasses

Suggestion that a class that is final has specializations, which is not permitted in UML.

15.16.5.  Illegal Generalization

Suggestion that there is a generalization between model elements of different UML metaclasses, which is not permitted.

[Caution]Caution

It is not clear that such a generalization can be created within ArgoUML. It probably indicates that the diagram was created by loading a corrupt project, with an XMI file describing such a generalization, possibly created by a tool other than ArgoUML.

15.16.6.  Remove Unneeded Realizes from <class>

Suggestion that the specified class has a realization relationship both directly and indirectly to the same interface (by realization from two interfaces, one of which is a generalization of the other for example). Good design deprecates such duplication.

15.16.7.  Define Concrete (Sub)Class

Suggestion that a class is abstract with no concrete subclasses, and so can never be realized.

15.16.8.  Define Class to Implement <interface>

Suggestion that the interface referred to has no influence on the running system, since it is never implemented by a class.

15.16.9.  Change Multiple Inheritance to interfaces

This critic is discussed under an earlier design issues category (see Section 15.14.1, “ Change Multiple Inheritance to interfaces ).

15.16.10.  Make Edge More Visible

This critic is discussed under an earlier design issues category (see Section 15.7.14, “ Make Edge More Visible ).

15.17.  Containment

Critics concerning containment in ArgoUML, that is where one model element forms a component part of another.

The current version of ArgoUML has the following critics in this category.

15.17.1.  Remove Circular Composition

Suggestion that there is a series of composition relationships (associations with black diamonds) that form a cycle, which is not permitted.

15.17.2.  Duplicate Parameter Name

Suggestion that a parameter list to an operation or event has two or more parameters with the same name, which is not permitted.

15.17.3.  Two Aggregate Ends (Roles) in Binary Association

Only one end (role) of a binary association can be aggregate or composite. This a well-formedness rule of the UML 1.4 standard.

Aggregation and composition are used to indicate whole-part relationships, and by definition, the “part” end cannot be aggregate.

To fix this, identify the “part” end of the association, and use the critic wizard (the Next> button, or manually set its aggregation to none using the button 2 pop-up menu or the property sheet.

Composition (more correctly called composite aggregation) is used where there is a whole-part relationship that is one-to-one or one-to-many, and the lifetime of the part is inextricably tied to the lifetime of the whole. Instances of the whole will have responsibility for creating and destroying instances of the associated part. This also means that a class can only be a part in one composite aggregation.

An example of a composite aggregation might be a database of cars and their wheels. This is a one-to-four relationship, and the database entry for a wheel is associated with its car. When the car ceases to exist in the database, so do its wheels.

Aggregation (more correctly called shared aggregation) is used where there is a whole-part relationship, that does not meet the criteria for a composite aggregation. An example might be a database of university courses and the students that attend them. There is a whole-part relationship between courses and students. However there is no lifetime relationship between students and course (a student continues to exist even after a course is finished) and the relationship is many-to-many.

15.17.4.  Aggregate End (Role) in 3-way (or More) Association

Three-way (or more) associations can not have aggregate ends (roles). This a well-formedness rule of the UML 1.4 standard.

Aggregation and composition are used to indicate whole-part relationships, and by definition can only apply to binary associations between model elements.

To fix this, manually select the association, and set the aggregation of each of its ends (roles) to none using the button 2 pop-up menu or the property sheet.

15.17.5.  Wrap DataType

This critic is discussed under an earlier design issues category (see Section 15.3.1, “ Wrap DataType ).

Part 3.  Model Reference

Chapter 16.  Top Level Model Element Reference

16.1.  Introduction

This chapter describes each model element that can be created within ArgoUML. The chapter covers top-level “general” model elements. The following chapters (see Chapter 17, Use Case Diagram Model Element Reference through Chapter 23, Deployment Diagram Model Element Reference ) cover each of the ArgoUML diagrams.

There is a close relationship between this material and the properties tab of the details pane (see Section 13.3, “ Properties Tab). That section covers properties in general, in this chapter they are linked to specific model elements.

16.2.  The Model

The model is the top level model element within ArgoUML. In the UML meta-model it is a sub-class of package. In many respects within ArgoUML it behaves similarly to a package (see Section 18.2, “ Package).

[Note]Note

ArgoUML is restricted to one model within the tool.

Standard data types, classes and packages are loaded (the default, see Chapter 24, Profiles ) as sub-packages of the model. These sub-packages are not initially present in the model but are added to the model when used.

16.2.1.  Model Details Tabs

The details tabs that are active for the model are as follows.

ToDoItem

Standard tab.

Properties

See Section 16.2.2, “ Model Property Toolbar and Section 16.2.3, “ Property Fields For The Model below.

Documentation

Standard tab. See Section 13.4, “ Documentation Tab.

Stereotype

Standard tab. This contains a a list of the stereotypes applied to this model, and a list of available stereotypes that may be applied to the model.

Tagged Values

Standard tab. In the UML meta-model, Model has the following standard tagged values defined.

  • derived (from the superclass, ModelElement).

    Values true, meaning the class is redundant — it can be formally derived from other elements, or false meaning it cannot.

    Derived models have their value in analysis to introduce useful names or concepts, and in design to avoid re-computation.

16.2.2.  Model Property Toolbar

Go up

Navigate up through the composition structure of the model.

Since the model is the top package nothing can happen, and this button is allways downlighted.

New Package

This creates a new Package (see Section 18.2, “ Package) within the model (which appears on no diagram), navigating immediately to the properties tab for that package.

[Tip]Tip

While it can make sense to create Packages of the model this way, it is usually a lot clearer to create them within diagrams where you want them.

New DataType

This creates a new DataType (see Section 16.3, “ Datatype) within the model (which appears on no diagram), navigating immediately to the properties tab for that DataType.

New Enumeration

This creates a new Enumeration (see Section 16.4, “ Enumeration) within the model (which appears on no diagram), navigating immediately to the properties tab for that Enumeration.

New Stereotype

This creates a new Stereotype (see Section 16.6, “ Stereotype) within the model, navigating immediately to the properties tab for that stereotype.

Delete

This tool is always downlighted, since it is meaningless to delete the model!

16.2.3.  Property Fields For The Model

Name

Text box. The name of the model. The name of a model, like all packages, is by convention all lower case.

[Note]Note

The default name supplied to a new model by ArgoUML, untitledModel, is thus erroneous and guarantees that ArgoUML always starts up with at least one problem being reported by the design critics.

Stereotype

Drop down selector. Model is provided by default with the UML standard stereotypes for model ( systemModel and metamodel) and package ( facade, framework, stub).

Stereotyping models is a useful thing, although it is of limited value in ArgoUML where you have only a single model.

Navigate Stereotype

icon. If a stereotype has been selected, this will navigate to the stereotype property panel (see Section 16.6, “ Stereotype).

Namespace

Text box. Records the namespace for the model. This is the package hierarchy. However since the model is at the top of the hierarchy in ArgoUML, this box is always empty.

Visibility

Radio box, with entries public, private, protected, and package.

Records the visibility for the model. Since ArgoUML only permits one model, this has no meaningful use.

Modifiers

Check box, with entries Abstract, Leaf and Root.

  • abstract is used to declare that this model cannot be instantiated, but must always be specialized.

    The meaning of abstract applied to a model is not that clear. It might mean that the model contains interfaces or abstract classes without realizations. Since ArgoUML only permits one model, this is not a meaningful box to check.

  • Leaf indicates that this model can have no further subpackages, while root indicates it is the top level model.

    Within ArgoUML root only meaningfully applies to the Model, since all packages sit within the model. In the absence of the topLevel stereotype, this could be used to emphasize that the Model is at the top level.

Generalizations

Text area. Lists any model that generalizes this model.

[Note]Note

Since there is only one model in ArgoUML there is no sensible specialization or generalization that could be created.

Specializations

Text box. Lists any specialized model (i.e. for which this model is a generalization.

[Note]Note

Since there is only one model in ArgoUML there is no sensible specialization or generalization that could be created.

Owned Elements

Text area. A listing of the top level packages, classes, interfaces, datatypes, actors, use cases, associations, generalizations, and stereotypes within the model.

Button 1 double click on any of the model elements yields navigating to that model element.

16.3.  Datatype

Datatypes can be thought of as simple classes. They have no attributes, and any operations on them must have no side-effects. A useful analogy is primitive datatypes in a language like Java. The integer “3” stands on its own—it has no inner structure. There are operations (for example addition) on the integers, but when I perform 3 + 4 the result is a new number, “3” and “4” are unchanged by the exercise.

Within UML 1.4, DataType is a sub-class of the Classifier metaclass. It embraces the predefined primitive types ( byte, char, double, float, int, long and short), the predefined enumeration, boolean and user defined enumeration types.

[Note]Note

Also void is implemented as a datatype within ArgoUML

Within ArgoUML new datatypes may be created using the New datatype button on the property tabs of the model and packages (in which case the new datatype is restricted in scope to the package), as well as the properties tab for datatype. Datatypes can also be created with the tool in the diagram toolbar of a class diagram.

The UML 1.4 standard allows user defined datatypes to be placed on class diagrams to define their inheritence structure. This is also possible in ArgoUML. It is represented on the diagram by a box with two compartments, of which the top one is marked with «datatype», and contains the name. The lower one contains operations.

16.3.1.  Datatype Details Tabs

The details tabs that are active for datatypes are as follows.

ToDoItem

Standard tab.

Properties

See Section 16.3.2, “ Datatype Property Toolbar and Section 16.3.3, “ Property Fields For Datatype below.

Documentation

Standard tab. See Section 13.4, “ Documentation Tab.

Source

Standard tab. Unused. One would expect a class declaration for the new datatype to support code generation.

Tagged Values

Standard tab. In the UML metamodel, Datatype has the following standard tagged values defined.

  • persistence (from the superclass, Classifier). Values transitory, indicating state is destroyed when an instance is destroyed or persistent, marking state is preserved when an instance is destroyed.

    [Tip]Tip

    Since user defined datatypes are enumerations, they have no state to preserve, and the value of this tagged value is irrelevant.

  • semantics (from the superclass, Classifier). The value is a specification of the semantics of the datatype.

  • derived (from the superclass, ModelElement). Values true, meaning the class is redundant—it can be formally derived from other elements, or false meaning it cannot.

    [Tip]Tip

    While formally available, a derived datatype does not have an obvious value, and so datatypes should always be marked with derived=false.

16.3.2.  Datatype Property Toolbar

Go up

Navigate up through the package structure.

New datatype

This creates a new datatype (see Section 18.6, “ Class) within the same package as the current datatype.

[Tip]Tip

While it can make sense to create datatypes this way, it can be clearer to create them within the package or model where you want them.

New Enumeration

This creates a new Enumeration (see Section 16.4, “ Enumeration) in the same package as the datatype, navigating immediately to the properties tab for that Enumeration.

New Operation

This creates a new operation within the datatype, navigating immediately to the properties tab for that operation.

New Stereotype

This creates a new Stereotype (see Section 16.6, “ Stereotype) within the same package as the datatype, navigating immediately to the properties tab for that stereotype.

Delete

This deletes the datatype from the model.

16.3.3.  Property Fields For Datatype

Name

Text box. The name of the datatype. The primitive datatypes all have lower case names, but there is no formal convention.

[Note]Note

The default name supplied for a newly created datatype is the empty string “”. Datatypes with empty string names will appear with the name (Unnamed Datatype) in the explorer.

Namespace

Drop down selector with navigate button. Allows changing the namespace for the datatype. This is the package hierarchy.

Modifiers

Check box, with entries Abstract, Leaf and Root.

  • Abstract is used to declare that this datatype cannot be instantiated, but must always be specialized.

    [Note]Note

    ArgoUML provides no mechanism for specializing datatypes, so this check box is of little use.

  • Leaf indicates that this datatype can have no further sub-types, while Root indicates it is a top level datatype.

    [Tip]Tip

    You can define the specialization of datatypes in a class diagram by drawing generalizations between them.

Visibility

Radio box, with entries public, private, protected, and package.

Records the visibility for the Datatype.

Client Dependencies

Text area. Lists any elements that depend on this datatype.

[Caution]Caution

It is not clear that dependencies between datatypes makes much sense.

Supplier Dependencies

Text area. Lists any elements that this datatype depends on.

[Caution]Caution

It is not clear that dependencies between datatypes makes much sense.

Generalizations

Text area. Lists any datatype that generalizes this datatype.

Specializations

Text box. Lists any specialized datatype (i.e. for which this datatype is a generalization.

Operations

Text area. Lists all the operations defined on this datatype. Button 1 double click navigates to the selected operation. button 2 click brings up a pop up menu with two entries.

  • Move Up. Only available where there are two or more operations, and the operation selected is not at the top. It is moved up one.

  • Move Down. Only available where there are two or more operations listed, and the operation selected is not at the bottom. It is moved down one.

See Section 18.8, “ Operation for details of operations.

[Caution]Caution

ArgoUML treats all operations as equivalent. Any operations created here will use the same mechanism as operations for classes. Remember that operations on datatypes must have no side effects (they are read-only). This means the query modifier must be checked for all operations.

16.4.  Enumeration

An enumeration is a primitive datatype that can have a fixed short list of values. It has no attributes, and any operations on them must have no side-effects. A useful analogy is the primitive datatype boolean in a language like Java. The boolean stands on its own—it has no inner structure. There are operations (for example logical xor) on the booleans, but when I perform true xor true the result is a new boolean, and the original 2 booleans “true” are unchanged by the exercise.

Within UML 1.4, Enumeration is a sub-class of the DataType metaclass.

The big difference with other DataTypes, is that an Enumeration has EnumerationLiterals. E.g. the Enumeration “boolean” is defined as having 2 EnumerationLiterals, “true” and “false”.

Within ArgoUML new enumerations may be created using the New Enumeration button on the property tabs of the model and packages (in which case the new enumeration is restricted in scope to the package), as well as the properties tab for datatype and enumeration. Enumerations can also be created with the tool in the diagram toolbar of a class diagram.

The UML 1.4 standard allows user defined enumerations to be placed on class diagrams to define their inheritence structure. This is also possible in ArgoUML. It is represented on the diagram by a box with three compartments, of which the top one is marked with «enumeration», and contains the name. The middle compartment shows the enumeration literals. The lower one contains operations.

16.4.1.  Enumeration Details Tabs

The details tabs that are active for enumerations are as follows.

ToDoItem

Standard tab.

Properties

See Section 16.4.2, “ Enumeration Property Toolbar and Section 16.4.3, “ Property Fields For Enumeration below.

Documentation

Standard tab. See Section 13.4, “ Documentation Tab.

Presentation

Standard tab.

Source

Standard tab.

Stereotype

Standard tab. The UML metamodel has the following stereotypes defined by default for a Classifier, which also apply to an Enumeration:

  • metaclass (from the superclass, Classifier).

  • powertype (from the superclass, Classifier).

  • process (from the superclass, Classifier).

  • thread (from the superclass, Classifier).

  • utility (from the superclass, Classifier).

Tagged Values

Standard tab. In the UML metamodel, Enumeration has no standard tagged values defined.

16.4.2.  Enumeration Property Toolbar

Go up

Navigate up through the composition structure.

New datatype

This creates a new datatype (see Section 18.6, “ Class) within the same package as the current enumeration.

New enumeration

This creates a new enumeration within the same namespace as the current enumeration, navigating immediately to the properties tab for new enumeration.

New enumeration literal

This creates a new enumeration literal within the enumeration, navigating immediately to the properties tab for that literal.

New Operation

This creates a new operation within the enumeration, navigating immediately to the properties tab for that operation.

New Stereotype

This creates a new Stereotype (see Section 16.6, “ Stereotype) within the same package as the enumeration, navigating immediately to the properties tab for that stereotype.

Delete from Model

This deletes the datatype from the model.

16.4.3.  Property Fields For Enumeration

Name

Text box. The name of the enumeration. The primitive enumerations all have lower case names, but there is no formal convention.

[Note]Note

The default name supplied for a newly created datatype is the empty string “”. Enumerations with empty string names will appear with the name (Unnamed Enumeration) in the explorer.

Namespace

Drop down selector with navigation button. Allows changing the namespace for the enumeration. This is the composition hierarchy.

Modifiers

Check box, with entries Abstract, Leaf and Root.

  • Abstract is used to declare that this enumeration cannot be instantiated, but must always be specialized.

  • Leaf indicates that this enumeration can have no further sub-types, while Root indicates it is a top level enumeration.

Visibility

Radio box, with entries public, private, protected, and package.

Records the visibility for the Enumeration.

Client Dependencies

Text area. Lists any elements that depend on this enumeration. Button 1 double click navigates to the selected modelelement. Button 2 click brings up a pop up menu with following entry.

  • Add.... This brings up a dialog box that allows to create dependencies from other modelelements.

Supplier Dependencies

Text area. Lists any elements that this enumeration depends on. Button 1 double click navigates to the selected modelelement. Button 2 click brings up a pop up menu with the following entry.

  • Add.... This brings up a dialog box that allows to create dependencies to other modelelements.

Generalizations

Text area. Lists any enumeration that generalizes this enumeration.

Specializations

Text box. Lists any specialized enumerations (i.e. for which this enumeration is a generalization.

Operations

Text area. Lists all the operations defined on this enumeration. Button 1 double click navigates to the selected operation. Button 2 click brings up a pop up menu with two entries.

  • Move Up. Only available where there are two or more operations, and the operation selected is not at the top. It is moved up one.

  • Move Down. Only available where there are two or more operations listed, and the operation selected is not at the bottom. It is moved down one.

See Section 18.8, “ Operation for details of operations.

[Caution]Caution

ArgoUML treats all operations as equivalent. Any operations created here will use the same mechanism as operations for classes. Remember that operations on enumerations must have no side effects (they are read-only). This means the query modifier must be checked for all operations.

Literals

Text area. Lists all the enumeration literals defined for this enumeration. Button 1 double click navigates to the selected literal, button 2 click brings up a pop up menu with two entries.

  • Move Up. Only available where there are two or more literals, and the literal selected is not at the top. It is moved up one.

  • Move Down. Only available where there are two or more literals listed, and the literal selected is not at the bottom. It is moved down one.

16.5.  Enumeration Literal

An enumeration literal is one of the predefined values of an Enumeration.

16.6.  Stereotype

Stereotypes are the main extension mechanism of UML, providing a way to derive specializations of the standard metaclasses. Stereotype is a sub-class of GeneralizableElement in the UML metamodel. Stereotypes are supplemented by constraints and tagged values.

New stereotypes are added from the property tab of almost any model element. Properties of existing stereotypes can be reached by selecting the property tab for any model element with that stereotype and using the navigate button ( ) within the property tab.

16.6.1.  Stereotype Details Tabs

The details tabs that are active for stereotypes are as follows.

ToDoItem

Standard tab.

Properties

See Section 16.6.2, “ Stereotype Property Toolbar and Section 16.6.3, “ Property Fields For Stereotype below.

Documentation

Standard tab. See Section 13.4, “ Documentation Tab.

Stereotype

Standard tab.

[Warning]Warning

Here you can set stereotypes of stereotypes, not a very usefull thing to do.

Tagged Values

Standard tab. In the UML metamodel, Stereotype has the following standard tagged values defined.

  • derived (from the superclass, ModelElement). Values true, meaning the class is redundant—it can be formally derived from other elements, or false meaning it cannot.

    [Note]Note

    This indicates any element with this stereotype has the derived tag set accordingly.

[Caution]Caution

Tagged values for a stereotype are rather different to those for elements in the UML core architecture, in that they apply to all model elements to which the stereotype is applied, not just the stereotype itself.

16.6.2.  Stereotype Property Toolbar

Go up

Navigate up through the package structure of the model.

Add stereotype

This creates a new stereotype (see Section 16.6, “ Stereotype) within the model (which appears on no diagram), navigating immediately to the properties tab for that stereotype.

New Tag Definition

This creates a new tag definition (see Section 16.7, “ Tag Definition) within the model (which appears on no diagram), navigating immediately to the properties tab for that tagdefinition.

Delete

This deletes the stereotype from the model.

16.6.3.  Property Fields For Stereotype

Name

Text box. The name of the stereotype. There is no convention for naming stereotypes, beyond starting them with a lower case letter. Even the standard UML stereotypes vary between all lower case (e.g. metamodel), bumpy caps (e.g. systemModel) and space separated (e.g. object model).

[Note]Note

ArgoUML does not enforce any naming convention for stereotypes

Base Class

Drop down selector. Any stereotype must be derived from one of the metaclasses in the UML metamodel or the model element classes that derive from them. The stereotype will then be available to model elements that derive from that same metaclass or that model element.

Namespace

Drop down selector with navigation button. Records the namespace for the stereotype. This is the package hierarchy.

Modifiers

Check box, with entries Abstract, Leaf and Root.

  • Abstract is used to declare that model elements that use this stereotype cannot be instantiated, but must always be specialized.

  • Leaf indicates that model elements that use this stereotype can have no further sub-types, while Root indicates it is a top level model element.

[Caution]Caution

Remember that these modifiers apply to the model elements using the stereotype, not just the stereotype.

[Warning]Warning

ArgoUML neither imposes, nor checks that model elements using a stereotype adopt the stereotype's modifiers.

Visibility

Radio box, with entries public, private, protected, and package.

Records the visibility for the stereotype.

Generalizations

Text area. Lists any stereotype that generalizes this stereotype.

Specializations

Text area. Lists any specialized stereotype (i.e. for which this stereotype is a generalization.

Tag Definitions

Text area. Lists any tag definitions that are defined for this stereotype.

Extended Elements

Text area. Lists all modelelements that are stereotyped by this stereotype.

16.7.  Tag Definition

(To Be Written)

16.8.  Diagram

The UML standard specifies eight principal diagrams, all of which are supported by ArgoUML.

  • Use case diagram. Used to capture and analyse the requirements for any OOA&D project. See Chapter 17, Use Case Diagram Model Element Reference for details of the ArgoUML use case diagram and the model elements it supports.

  • Class diagram. This diagram captures the static structure of the system being designed, showing the classes, interfaces and datatypes and how they are related. Variants of this diagram are used to show package structures within a system (the package diagram) and the relationships between particular instances (the object diagram).

    The ArgoUML class diagram provides support for class and package diagrams. See Chapter 18, Class Diagram Model Element Reference for details of the model elements it supports. The object diagram is suported on the Deployment diagram.

  • Behavior diagrams. There are four such diagrams (or strictly speaking, five, since the use case diagram is a type of behavior diagram), which show the dynamic behavior of the system at all levels.

    • Statechart diagram. Used to show the dynamic behavior of a single object (class instance). This diagram is of particular use in systems using complex communication protocols, such as in telecommunications. See Chapter 20, Statechart Diagram Model Element Reference for details of the ArgoUML statechart diagram and the model elements it supports.

    • Activity diagram. Used to show the dynamic behavior of groups of objects (class instance). This diagram is an alternative to the statechart diagram, and is better suited to systems with a great deal of user interaction. See Chapter 22, Activity Diagram Model Element Reference for details of the ArgoUML activity diagram and the model elements it supports.

    • Interaction diagrams. There are two diagrams in this category, used to show the dynamic interaction between objects (class instances) in the system.

      • Sequence diagram. Shows the interactions (typically messages or procedure calls) between instances of classes (objects) and actors against a timeline. Particularly useful where the timing relationships between interactions are important. See Chapter 19, Sequence Diagram Model Element Reference for details of the ArgoUML sequence diagram and the model elements it supports.

      • Collaboration diagram. Shows the interactions (typically messages or procedure calls) between instances of classes (objects) and actors against the structural relationships between those instances. Particularly suitable where it is useful to relate interactions to the static structure of the system. See Chapter 21, Collaboration Diagram Model Element Reference for details of the ArgoUML collaboration diagram and the model elements it supports.

  • Implementation diagrams. UML defines two implementation diagrams to show the relationship between the software components that make up a system (the component diagram) and the relationship between the software and the hardware on which it is deployed at run-time (the deployment diagram.

    The ArgoUML deployment diagram provides support for both component and deployment diagrams, and additionally for object diagrams. See Chapter 23, Deployment Diagram Model Element Reference for details of the diagram and the model elements it supports.

Diagrams are created using the Create drop down menu (see Section 10.6, “ The Create Menu ), or with the tools on the toolbar (see Section 9.4, “ Create operations), or with the pop-up menus in the explorer.

[Note]Note

ArgoUML uses its deployment diagram to create the UML 1.4 component, deployment and object diagrams.

[Caution]Caution

Statechart and activity diagrams are associated with a particular class or operation (or the latter also with a package), and can only be created when this modelelement has been selected.

[Warning]Warning

In ArgoUML version 0.26, the UML 1.4 object diagram as a variant of the class diagram is not directly supported. However, it is possible to create object diagrams within the ArgoUML deployment diagram.

16.8.1.  Diagram Details Tabs

The details tabs that are active for diagrams are as follows.

ToDoItem

Standard tab.

Properties

See Section 16.8.3, “ Property Fields For Diagram below.

16.8.2.  Diagram Property Toolbar

Go up

Navigate up through the package structure of the model.

Delete

This deletes the diagram from the model. As a consequence, in case of a statechart diagram or an activity diagram, all contained elements are deleted, too.

16.8.3.  Property Fields For Diagram

Name

The name of the diagram. There are no conventions for naming diagrams. By default, ArgoUML uses the (space separated) diagram name and a sequence number, thus Use Case Diagram 1.

[Tip]Tip

This name is used to generate a filename when activating the “Save Graphics...” menu-item.

Home Model

The Home Model of the diagram is not something defined in the UML specification. The Home Model is the modelelement represented by the diagram. Hence its type depends on the type of diagram: e.g. it is the namespace represented by a class diagram, or the statemachine in case of a Statechart diagram.

Chapter 17.  Use Case Diagram Model Element Reference

17.1.  Introduction

This chapter describes each model element that can be created within a use case diagram. Note that some sub-model elements of model elements on the diagram may not actually themselves appear on the diagram.

There is a close relationship between this material and the properties tab of the details pane (see Section 13.3, “ Properties Tab). That section covers properties in general, in this chapter they are linked to specific model elements.

Figure 17.1, “ Typical model elements on a use case diagram. shows a use case diagram with all typical model elements displayed.

Figure 17.1.  Typical model elements on a use case diagram.

Typical model elements on a use case diagram.

17.1.1.  ArgoUML Limitations Concerning Use Case Diagrams

Use case diagrams are now well supported within ArgoUML. There still are some minor limitations though, especially with extension points.

[Note]Note

Earlier versions of ArgoUML (0.9 and earlier) implemented extend and include relationships by using a stereotyped dependency relationship. Although such diagrams will show correctly on the diagram, they will not link correctly to the use cases, and should be replaced by proper extend and include relationships using the current system.

17.2.  Actor

An actor represents any external entity (human or machine) that interacts with the system, providing input, receiving output, or both.

Within the UML metamodel, actor is a sub-class of classifier.

The actor is represented by a “stick man” figure on the diagram (see Figure 17.1, “ Typical model elements on a use case diagram.).

17.2.1.  Actor Details Tabs Detail-Register Akteur

The details tabs that are active for actors are as follows.

ToDoItem

Standard tab.

Properties

See Section 17.2.2, “ Actor Property Toolbar and Section 17.2.3, “ Property Fields For Actor below.

Documentation

Standard tab. See Section 13.4, “ Documentation Tab.

Presentation

Standard tab. The fill color is used for the stick man's head.

Source

Standard tab. Usually, no code is provided for an actor, since it is external to the system.

Stereotype

Standard tab.

Tagged Values

Standard tab. In the UML metamodel, Actor has the following standard tagged values defined.

  • persistence (from the superclass, Classifier). Values transitory, indicating state is destroyed when an instance is destroyed or persistent, marking state is preserved when an instance is destroyed.

    [Tip]Tip

    Actors sit outside the system, and so their internal behavior is of little concern, and this tagged value is best ignored.

  • semantics (from the superclass, Classifier). The value is a specification of the semantics of the actor.

  • derived (from the superclass, ModelElement). Values true, meaning the actor is redundant—it can be formally derived from other elements, or false meaning it cannot.

    [Note]Note

    Derived actors have limited value, since they sit outside the system being designed. They may have their value in analysis to introduce useful names or concepts.

Checklist

Standard tab for a Classifier.

17.2.2.  Actor Property Toolbar

Go up

Navigate up through the package structure of the model.

Add Actor

This creates a new actor within the model, (but not within the diagram), navigating immediately to the properties tab for that actor.

[Tip]Tip

This method of creating a new actor may be confusing. It is much better to create an actor on the diagram.

New Reception

This creates a new reception within the model,(but not within the diagram), navigating immediately to the properties tab for that rception.

[Tip]Tip

A reception is a declaration that the actor handles a signal, but the actual handling is specified by a state machine.

Delete

This deletes the selected actor from the model.

[Warning]Warning

This is a deletion from the model not just the diagram. To delete an actor from the diagram, but keep it within the model, use the main menu Remove From Diagram (or press the Delete key).

17.2.3.  Property Fields For Actor

Name

Text box. The name of the actor. The diagram shows this name below the stick man figure. Since an actor is a classifier, it would be conventional to Capitalize the first letter (and initial letters of any component words), e.g. RemoteSensor.

[Note]Note

ArgoUML does not enforce any naming convention for actors

Namespace

Text box with navigation button. Records the namespace for the actor. This is the package hierarchy.

Modifiers

Check box, with entries Abstract, Leaf and Root.

  • Abstract is used to declare that this actor cannot be instantiated, but must always be specialized.

    [Caution]Caution

    While actors can be specialized and generalized, it is not clear that an abstract actor has any meaning. Perhaps it might be used to indicate an actor that does not itself interact with a use case, but whose children do.

  • leaf indicates that this actor can have no further children, while Root indicates it is a top level actor with no parent.

Generalizations

Text area. Lists any actor that generalizes this actor.

Button 1 double click navigates to the generalization and opens its property tab.

Specializations

Text box. Lists any specialized actor (i.e. for which this actor is a generalization. The specialized actors can communicate with the same use case instances as this actor.

Button 1 double click navigates to the generalization and opens its property tab.

Association Ends

Text area. Lists any association ends of associations connected to this actor.

Button 1 double click navigates to the selected entry.

17.3.  Use Case

A use case represents a complete meaningful “chunk” of activity by the system in relation to its external users (actors), human or machine. It represents the primary route through which requirements are captured for the system under construction

Within the UML metamodel, use case is a sub-class of classifier.

The use case icon is an oval (see Figure 17.1, “ Typical model elements on a use case diagram.). It may be split in two, with the lower compartment showing extension points

[Caution]Caution

By default ArgoUML does not show the extension point compartment. It may be revealed by the context sensitive Show menu (using button 2 click), or from the Presentation tab.

17.3.1.  Use Case Details Tabs

The details tabs that are active for use cases are as follows.

ToDoItem

Standard tab.

Properties

See Section 17.3.2, “ Use Case Property Toolbar and Section 17.3.3, “ Property Fields For Use Case below.

Documentation

Standard tab. See Section 13.4, “ Documentation Tab.

Presentation

Standard tab. The Fill color is used for the use case oval.

The Display: Extension Points check box is used to control whether an extension point compartment is displayed.

Source

Standard tab. It would not be usual to provide any code for a use case, since it is primarily a vehicle for capturing requirements about the system under construction, not creating the solution.

Stereotype

Standard tab.

Tagged Values

Standard tab. In the UML metamodel, UseCase has the following standard tagged values defined.

  • persistence (from the superclass, Classifier). Values transitory, indicating state is destroyed when an instance is destroyed or persistent, marking state is preserved when an instance is destroyed.

    [Tip]Tip

    In general the instantiation of use cases is not a major aspect of any design method (they are mostly concerned with requirements capture. For most OOA&D methodologies, this tag can safely be ignored.

  • semantics (from the superclass, Classifier). The value is a specification of the semantics of the use case.

  • derived (from the superclass, ModelElement). Values true, meaning the use case is redundant—it can be formally derived from other elements, or false meaning it cannot.

    [Note]Note

    Derived use cases still have their value in analysis to introduce useful names or concepts.

Checklist

Standard tab for a Classifier.

17.3.2.  Use Case Property Toolbar

Go up

Navigate up through the package structure of the model.

New use case

This creates a new use case within the model, (but not within the diagram), and shows immediately the properties tab for that use case.

[Tip]Tip

This method of creating a new use case can be confusing. It is much better to create a new use case on the diagram of your choice.

New extension point

This creates a new use extension point within the namespace of the current use case, with the current use case as its associated use case, navigating immediately to the properties tab for that extension point.

New Attribute

This creates a new attribute within the current use case, navigating immediately to the properties tab for that attribute.

New Operation

This creates a new operation within the current use case, navigating immediately to the properties tab for that operation.

New Reception

This creates a new reception within the current use case, navigating immediately to the properties tab for that reception.

New Stereotype

This creates a new stereotype within the current use case, navigating immediately to the properties tab for that stereotype.

Delete

This deletes the selected use case from the model.

[Warning]Warning

This is a deletion from the model not just the diagram. To delete a use case from the diagram, but keep it within the model, use the main menu Remove From Diagram (or press the Delete key).

17.3.3.  Property Fields For Use Case

Name

Text box. The name of the use case. Since a use case is a classifier, it would be conventional to Capitalize the first letter (and initial letters of any component words), e.g. RemoteSensor. The name is shown inside the oval representation of the use case on the diagram.

[Note]Note

ArgoUML does not enforce any naming convention for use cases

Namespace

Text box with navigation button. Records the namespace for the use case. This is the package hierarchy.

Modifiers

Check box, with entries Abstract Leaf and Root.

  • Abstract is used to declare that this actor cannot be instantiated, but must always be specialized. .

  • Leaf indicates that this use case can have no further children, while Root indicates it is a top level use case with no parent.

Client Dependencies

Text area. Lists the “depending” ends of the relationship, i.e. the end that makes use of the other end.

Button 1 double click navigates to the dependency and opens its property tab.

Button 2 click shows a pop-up menu with one entry Add... that opens a dialog box where you can add and remove depending modelelements.

Supplier Dependencies

Text area. Lists the “supplying” ends of the relationship, i.e. the end supplying what is needed by the other end.

Button 1 double click navigates to the dependency and opens its property tab.

Button 2 click shows a pop-up menu with one entry Add... that opens a dialog box where you can add and remove dependent modelelements.

Generalizations

Text area. Lists use cases which are generalizations of this one. Will be set whenever a generalization is created on the from this Use Case. Button 1 Double Click on a generalization will navigate to that generalization.

Specializations

Text box. Lists any specialized use case (i.e. for which this use case is a generalization.

Button 1 double click navigates to the generalization and opens its property tab.

Extends

Text box. Lists any class that is extended by this use case.

Where an extends relationship has been created, button 1 double click will navigate to that relationship.

Includes

Text box. Lists any use case that this use case includes.

Where an include relationship has been created, button 1 Double Click will navigate to that relationship.

Attributes

Text area. Lists all the attributes (see Section 18.7, “ Attribute) defined for this use case. Button 1 double click navigates to the selected attribute. Button 2 gives a pop up menu with two entries, which allow reordering the attributes.

  • Move Up. Only available where there are two or more attributes listed, and the attribute selected is not at the top. It moves the attribute up one position.

  • Move Down. Only available where there are two or more attributes listed, and the attribute selected is not at the bottom. It moves the attribute down one position.

Association Ends

Text box. Lists any association ends (see Section 18.12, “ Association) of associations connected to this use case.

Button 1 double click navigates to the selected entry.

Operations

Text area. Lists all the operations (see Section 18.8, “ Operation) defined on this use case. Button 1 click navigates to the selected operation. Button 2 gives a pop up menu with two entries, which allow reordering the operations.

  • Move Up. Only available where there are two or more operations listed, and the operation selected is not at the top. It moves the operation up one position.

  • Move Down. Only available where there are two or more operations listed, and the operation selected is not at the bottom. It moves the operation down one position.

Extension Points

Text box. If this use case is, or can be extended, this field lists the extension points for the use case.

[Note]Note

Extension points are listed by their location point rather than their name.

Where an extension point has been created (see below), button 1 Double Click will navigate to that relationship. Button 2 gives a pop up menu with the following entries.

  • New. Add a new extension point and navigate to it, making this use case the owning use case of the extension point.

  • Move Up. Only available where there are two or more extension points listed, and the extension point selected is not at the top. It moves the extension point up one position.

  • Move Down. Only available where there are two or more extension points listed, and the extension point selected is not at the bottom. It moves the extension point down one position.

17.4.  Extension Point

An extension point describes a point in a use case where an extending use case may provide additional behavior.

Examples for a travel agent sales system might be the use case for paying for a ticket, which has an extension point in the specification of the payment. Extending use cases may then extend at this point to pay by cash, credit card etc.

Within the UML metamodel, Extension Point is a sub-class of ModelElement. A use case may display an extension point compartment (see Section 17.3, “ Use Case for details), in which extension points are shown with the following syntax.

name :location.

17.4.1.  Extension Point Details Tabs

The details tabs that are active for extension points are as follows.

ToDoItem

Standard tab.

Properties

See Section 17.4.2, “ Extension Point Property Toolbar and Section 17.4.3, “ Property Fields For Extension Point below.

Documentation

Standard tab. See Section 13.4, “ Documentation Tab.

Stereotype

Standard tab.

Extensionpoints do not have any stereotypes defined by default.

Tagged Values

Standard tab. In the UML metamodel, ExtensionPoint has the following standard tagged values defined.

  • derived (from the superclass, ModelElement). Values true, meaning the extension point is redundant—it can be formally derived from other elements, or false meaning it cannot.

    [Note]Note

    It is not clear how derived extension points could have any value in analysis.

17.4.2.  Extension Point Property Toolbar

Go up

Navigate up to the use case which owns this extension point.

New Extension Point

This creates a new Extension Point below the selected extension point, navigating immediately to the properties tab of the newly created extension point.

New Stereotype

This creates a new Stereotype (see Section 16.6, “ Stereotype) for the selected extension point, navigating immediately to the properties tab for that stereotype.

Delete

This deletes the selected extension point from the model.

17.4.3.  Property Fields For Extension Point

Name

Text box. The name of the extension point.

[Tip]Tip

It is quite common to leave extension points unnamed in use case analysis, since they are always listed (within use cases and extend relationships) by their location.

[Note]Note

ArgoUML does not enforce any naming convention for extension points.

Location