Show Posts

This section allows you to view all posts made by this member. Note that you can only see posts made in areas you currently have access to.


Messages - Sunshine

Pages: 1 ... 27 28 [29] 30 31 ... 46
421
General Board / Re: how to grant user in a sqlserver datbase
« on: September 10, 2014, 07:16:07 am »
If you are having trouble setting up a SQL database then suggest seeking help from a SQL DBA or go to the MS website as there is a lot of help there on the subject. Sorry but its too technical to explain to a novice on this forum. You need to understand about  the workings of the DBMS and security models such as impersonation, trusted subsystem, single sign on etc. Most places I've worked seem to set up the database with trusted subsystem. i.e. use a username and password to access the database and that is used in the connection string. Once that's done you enable security in Sparx EA Model and create user names and passwords for each user giving them the various access rights to model.
However the you can use single sign on which passes the security token from the windows logon. Again you still need to enable security in Sparx EA model and add the users using the active directory (assuming thats being used) user names and access rights to the model.
Hope that helps  :)

422
General Board / Re: Shared Project: Diagrams looks different by Us
« on: September 10, 2014, 07:08:10 am »
Another thing as well as those mentioned above is that user settings for MDG selections can also effect the presentation. i.e. if someone has BPMN enabled and another does not then they won't see the same thing.

So just to recap check ;
  • you have the same Graphics settings on the PC
  • Have the same Sparx EA settings
  • [-] Namespace settings etc in tools|options
    [-] MDG settings

423
General Board / Model Views Font Size?
« on: August 29, 2014, 01:21:39 pm »
Just started using the new model view artifacts in V11 but I find I can't seem to fix the font size and it varies depending on the zoom. I've also found that what I see on the screen isn't the same as what gets printed. i.e. If I have a list of elements say 1-20 showing when I print out it shows 1-18 or something less.
Has anyone else experienced the same sort of issues?

424
General Board / Re: Data modelling
« on: August 01, 2014, 07:04:05 am »
IMHO you shouldn't be putting primary keys or foreign keys in to the logical model. That's what the physical model is for isn't it? The simple reason is that a logical model can be transformed to many types of physical models such as database or EJB or C#, XML etc. But I guess this is where the Object Oriented World vs the Data World differ.
If you need constraints on attributes or classes then one possible way forward is to abstract it out to be a stereo type and apply the constraint when the logical model is transformed to the physical based on the stereotype. That way you can apply the constraint in a platform specific way. It also means if you need to change it you can change it in the transform script and apply it globally to the physical models.
Hope that makes sense but I know from past experience a lot of people just simply don't get the concept.

425
General Board / Re: Data modelling
« on: July 30, 2014, 07:41:57 pm »
Here is one way
Conceptual - use packages for subject areas
Logical - use classes
Physical - use tables with primary and foreign keys

Can use DDL transformations to go from logical to physical.

Simple aye? :)

426
General Board / Re: EA as an ALM tool
« on: August 13, 2014, 06:52:47 am »
Just a further thought. If you are approving requirements in bulk in a package then you can use the update package status feature. Right click on package>advanced>update package status. You get a dialog where you can update all the elements status and version at the same time.
Not sure if that helps, hope it does. :)

427
General Board / Re: EA as an ALM tool
« on: August 08, 2014, 08:29:24 am »
Quote
Create workflows (e.g. requirement shall not be added as approved without review and approval)
Haven't used it myself but I understand you might be able to do this via workflow scripts. Have a look in user manual and search for workflow scripts for more details.
Quote
Create user groups per project
Yes kind of. You can create groups to restrict who can do what and you can create a group to lock a project. So horizontal and vertical restrictions in one model repository (i.e. central model paradigm). You could create a repository for each project and create a user group for each repository. Not such a good approach if you want to share things across projects.
Quote
Customize who can do what (for example developer shall not be allowed to change requirement status to approved)
No but you could put the requirements in a package and apply a lock that just lets a group say "Business Analysts" edit elements in that package.  
Quote
Automatic versioning of approved requirements
Not sure but it might be possible via workflow scripts. Hopefully someone else can answer this one.
Hope that helps  :)

428
General Board / Re: Gap Analysis - Baseline and Target Packages
« on: August 13, 2014, 06:43:24 am »
Automatic gap analysis mmh? Sounds like good idea I think there is potential for a new feature request in that thought. Maybe combine the compare utility with the gap matrix?
In the meantime you may want try looking at the compare utility. From the online help it says you can compare the following;
Quote
  • Compare a model branch in Enterprise Architect with a Baseline created using the Baseline functionality (Corporate, Business and Software Engineering, System Engineering and Ultimate editions)
  • Compare a model branch in Enterprise Architect with a  Baseline stored in a different model
  • Compare a model branch in Enterprise Architect with an XML 1.1 file on disk created previously using the Enterprise Architect XML export facility (user selects file)
  • Compare a model branch in Enterprise Architect with the current version-controlled XMI 1.1 file on disk as created when using Version Control in Enterprise Architect (file automatically selected)
  • Compare a .EAP file to another .EAP file
  • Compare a .EAP file to a DBMS-based repository
  • Compare two DBMS repositories
So you could base line the package then updated it to the future state and compare. The problem is if you are still developing the current state in parallel with the future.
Or you could put you current state in one model and future state in another.

429
General Board / Re: Gap Analysis - Baseline and Target Packages
« on: August 08, 2014, 09:09:15 am »
Not used the gap analysis matrix in anger but how I interpret how to use it is that it cross references elements in baseline to target. So in essence you are duplicating elements if they exist in both states. Where elements exist only in the baseline but not the target you have a column where you can create a gap element to say it is missing or was eliminated. Where you have an element only in the target state you have a row where you can add a gap element to say something about the gap. Finally where an element exists in both you can either say nothing about it or add a note to say something about it like it was upgraded in someway or no change.

430
General Board / Re: Limiting TOC Heading Level Output
« on: July 28, 2014, 06:55:23 am »
The only way I've managed to get this to work is to export the TOC template and edit in MS Word. Toggle field codes on the TOC and change {TOC \o "1-9"} to {TOC \o "1-3"} and save. Then import the updated rtf file back into Sparx.
Hope that helps :)

431
General Board / Re: Assigning images to custom stereotypes
« on: July 17, 2014, 02:24:06 pm »
Quote
That worked perfectly, thanks for that.

Is there any way to print this on the outside of the image box so that it doesn't sit on top of my nice pretty image?

In addition to the previous suggestion you could try putting the print statement in a shape label to get floating text label.
Code: [Select]
...
shape label
{
    print("#name#")// draw a floating text label
}

RTM: http://www.sparxsystems.com/enterprise_architect_user_guide/10/extending_uml_models/writing_scripts.html

432
General Board / Re: Assigning images to custom stereotypes
« on: July 14, 2014, 07:06:40 am »
Sounds like your going though the pain barrier. Keep up the good work you'll get through it Jayson.

To get the image which you should have included in the MDG generation process you'll need to add the MDG name space of the image. This is the name you call the MDG. i.e. what you fill in on the 'Technology' field of the MDG Technology Creation Wizard Dialog.
So if you have a MDG Technology called 'MyMDG' you need to modify your code to something like this

Code: [Select]
shape main
{
   image("MyMDG::imagename",0,0,100,100);
}

Good luck Mr Speer  ;)

433
General Board / Re: Different Levels of Detail (Aggregation)
« on: July 10, 2014, 07:50:04 am »
Quote
Quote
Dear Sunshine,

I followed you advice and found the picture in http://www.sparxsystems.com/resources/mda/. This is a good starting point for what we are looking for. Unfortunately I we do not have 1:1 mappings between models. We need it more detailed and more flexibel.

V
Been busy working for a living so haven't responded earlier.
I guess what you are asking is how do you map the PIM elements to those transformed in the PSM?
You simply put the elements on a diagram draw the relationship. If you need it down to attribute level right click on the link and use the link element to feature to choose the attribute.
If you used the built in transforms then somewhere in the database the GUIDs are stored to keep the relationship between the source and target. If you have some programming experience you could probably parse those GUIDs with javascript and create the relationships between the elements from the PIM to the PSM automagically.
Hope that helps in some way  :)


434
General Board / Re: Different Levels of Detail (Aggregation)
« on: July 10, 2014, 07:47:41 am »
Quote
Dear Sunshine,

I followed you advice and found the picture in http://www.sparxsystems.com/resources/mda/. This is a good starting point for what we are looking for. Unfortunately I we do not have 1:1 mappings between models. We need it more detailed and more flexibel.

V
Been busy working for a living so haven't responded earlier.
I guess what you are asking is how do you map the PIM elements to those transformed in the PSM?
You simply put the elements on a diagram draw the relationship. If you need it down to attribute level right click on the link and use the link element to feature to choose the attribute.
If you used the built in transforms then somewhere in the database the GUIDs are stored to keep the relationship between the source and target. If you have some programming experience you could probably parse those GUIDs with javascript and create the relationships between the elements from the PIM to the PSM.
Hope that helps in some way  :)


435
General Board / Re: Different Levels of Detail (Aggregation)
« on: July 08, 2014, 07:08:05 am »
Think I understand you have a model with N:M cardinality and you want to transform to an ERM and insert a join table.
i.e.

Partner(n)->(m)Contract where n & m are cardinality

Transform to ERM inserting helper entity

Partner->(n)HelperEntity(m)->Contract

Been a while since I looked at it but I think DDL transform built into Sparx does something similar when transforming PIM to PSM (DDL) so have a look at that and perhaps modify it to suit your needs.

Pages: 1 ... 27 28 [29] 30 31 ... 46