I have a part of my program which can be called by various events. Each event however does something different before making use of this part. How can I represent these using a diagram? I was thinking of a flowchart but as far as I know a flow chart can have one start terminal, right?
Thanks a lot for the help,
Regards,
Krt_Malta
UML Activity Diagram.
http://www.ibm.com/developerworks/rational/library/2802.html
You have a "wait for event" node that has multiple incoming events. Each event has a different state transition.
Related
I am trying to create an activity diagram and i need to represent an update activity. So is it possible to do this in activity diagram or there is another alternatives to do so ???
thanks for your help.
You can use a receive event like this:
Probably you shouldn't bother with interruptible regions here.
I was wondering, how do you implement a different process in an activity diagram? For example, I have following diagram:
Now, if I want to show in the diagram that the system should backup all files on my computer every 10 pm, how would I do this?
EDIT:
Take a look at the Accept Time Event Action as depicted below.
Use a decision symbol to state two separate works, and the hour glass for waiting.
i`m wondering what is the way to display an event which appears after some period of time on a use-case realization diagram. For example lets say an actor taps a display of his smart phone, the display turns on and in 5 minutes it turns off. Is it allowed in UML to write a message on event like "DisplayOFF(5min)",if not, what is most proper way?
What do you mean by "use-case realization diagram"? UML use case diagrams are unaware of sequence flows. It is also quite solution-independent, because it is rather meant to show objectives of a system. Have a look at UML sequence diagrams and check whether or not they satisfy your prescribed need.
I would like to model "open" processes in BPMN, that is, processes designed to receive messages at points in their execution, from an arbitrary entity that will be specified at a second moment. This would be best written in my opinion as a collaboration diagram with hanging message flows that have empty source or target. Is this standard BPMN? Does it make sense? Are there other ways to solve the problem?
This seems to be best done in BPMN using "black box" pools that contain unknown processes, as explained for example here:
http://www.visual-paradigm.com/product/bpva/tutorials/bpmn2.jsp
I have a question about UML diagrams.
Can be displayed relations between loose coupled components on sequence diagrams?
It is required in loose coupled environment (MQ) or optional?
For example, component "A" put some message into queue "X", then component "B" takes this message from queue "X". Should be this relation displayed on sequence diagram?
Thanks for help.
The one sequence diagram is not good place for it because according your question there are TWO asynchronous interactions.
The best way is to use TWO sequence diagrams or state chart diagram (you can use it to describe how queue works)
Yes, you can do that. A sequence diagram demonstrates how components/classes interact in a use case. As long as the resulting diagram clearly demonstrated the interaction you are modelling then you should show the interactions. If the diagram looks very cluttered and difficult to read then you could split in to 2 seperate diagrams; first shows 'A' put the message on the queue and the second diagram shows 'B' pull the message off the queue. Ultimately, you need to make the concepts in the diagram easy for somebody to read, so less clutter the better.