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.

Topics - Mr. Sanders

Pages: [1] 2 3 4
General Board / floating license expiration
« on: February 01, 2016, 11:52:38 pm »

we set the Floating license expiration to 5 days.
When we look at the checked out licenses they are mentioned to expire in 1 day. ???

Is the second issue intended?
How can we know, when the license checkout expires than?


Suggestions and Requests / segregate/decompose large sequence diagrams
« on: September 17, 2009, 09:01:09 pm »

we would be very pleased about that feature:

When sequence diagrams grow through their live cycle it is often later recognised, which part of it can be a sequence diagram of its own to be referenced in other diagrams (like loops, optional parts, etc.)

Currently one has to draw this  parts  as a separate diagram and then can add it as a reference to the original diagram and has to remove the messages out of the original diagram.

This is very cumbersome.

It would be much more easy to add a fragment element to the original large sequence diagram and than being able to create a new diagram from all objects and messages in this fragment by converting it to a referenced diagram.

This is the straigth forward way to cut large diagrams into smaller pieces.


Suggestions and Requests / partial C# classes
« on: March 19, 2008, 10:11:43 pm »

there is an ugly behavior when one has to handle partial classes.
Especially some UI wizards generate partial classes to separate the
user modifyable code and the always regenerated code from the wizard.

The problem occurs when these partial classes are e.g. reverse engineered.

Doing it this way, one will get n class elements in its package for the n parts existing in the n implementation files.

The ugly behavior is, when one has to do some sequence or class diagrams.

You never know in which class element the method, association, property etc. is, you currently need.

That means that you select in 50% (when having 2 parts) the wrong element. A worse case is, that you have n elements with the same name on your diagrams to be able showing the information you want to elaborate.

It would be better if EA wrappes these parts into a single element, which offers all properties, methods, attributes of all partial implementations.

That would be a great help.

What do you recon?


Suggestions and Requests / reverse engineering
« on: January 19, 2008, 08:08:21 am »

last week I had some discussions with sparx.
I was able to convince them, that the reverse engineering has some problematic features.

The said, they will change it in the following way:

In future it wönt be necessary to checkout every pacakge from namespace root down to the package that shall be reverse engineered.

In a first step only the namespace root and the direct package must be checked out.

In a second step, it is even not necessary to checkout the namespace root.

In a third step, EA will take care about the existing namespace package and when you reverse engineer A.B.C.MyClass  from package C it will be engineered into NSR.A.B.C and not in NSR.A.B.C.A.B.C as in the current implementation.

I hope the will fix this soon.


Suggestions and Requests / reverse engineering and renamed classes in code
« on: November 08, 2007, 04:02:37 am »

We have the following scenario:

We use kind of 3rd party classes, which we reverse engineered to be able using these classes in our model.

Now our model is infused with hundreds of sequence diagrams using these classes.

Now we got a new release of the 3rd party classes, where some very core classes were renamed.

What happens at reverse engineering?
EA find the new renamed classes and provides you a dialog where you can delete the old class from the model, because EA thinks the old classes are gone from code. (because of the missing ids in the source code, rose!)

Deleting the old class corupts all diagrams in our model and this is a very rude mechanism.

Not deleting it results in two classes.
The new renamed one (not in the diagrams) and the still in the model existing old class (still in the diagrams).

What we need are additional buttons/dialogs/mechanisms where we are able to say
e.g. replace old class with new class (so internal uids stay the same), so the diagrams are not corupted.

Of course there are still other problems to solve, e.g.
methods/attibutes/properties are renamed, ...
In an additional release we would expect to have the same kind of matching of old to new methods/attributes/.. too.

But for a first shot, it would be very helpful that hundreds of diagrams are not corupted.

It's much easier to redraw the new messages than to redraw the whole diagram or diagram part concerning the deleted class.

Please give us a solution for that.


Suggestions and Requests / Database icon
« on: November 08, 2007, 07:13:17 am »

can you please provide a database symbol/icon?

I am missing it very often and it's such a basic thing.


Suggestions and Requests / Parts and Ports
« on: November 06, 2007, 07:34:33 am »

when drag'n'drop classes to components you can define that the class is part, port or neither of them.

When you select Port or Part you have to give them a name
and the default proposal that EA gives is Port  or Part.

It would be very helpful if the name would be the same as the class I drop.

In most cases this is good enough.

In the way EA implements it, you always have to change that default, because Port and Part explains nothing in the diagram.

If I don't want to have it as a port or part,
I would expect that the dropped class is shown in the project explorer as a sub element of the component, but not as a copy, rather than as a reference to the original class.

Thus I would circumvent the port/part level, which we don't find necessary. It would be very nice, if the class references are directly shown below the component node in the project browser.


Suggestions and Requests / dependencies on packages
« on: November 05, 2007, 08:46:56 am »

I wonder if there is a possibility to show the dependencies between packages.

Asume you have


so ClassA is in namespace/package A
and ClassB is in namespace/package B.
Now draw an aggregation in a way that
ClassA aggregates ClassB.

If you have modelled it this way you have an
implicit dependency from package A to package B.

If you now create a package diagram and drag and drop
package A and B on it, NO dependency is shown.

This is wrong.

And reverse engineering doesn't elaborate the depencies on this level.

This is bad.

Because that's what I want to see for refactoring and it would be an ease for such a tool to draw this dependencies.

So please create this dependencies at reverse engineering and modeling automatically.

This information is very worthy.


Suggestions and Requests / deleting namespaces
« on: October 17, 2007, 11:33:35 pm »

the behaviour of deleting namespaces changed from EA 6.5 to EA 7.0.

Settings -> Namespaces

formerly you had the possibility to delete namespaces.
Now this possibility seems to be gone.

Is it unfeatured or bugged or am I too simple minded to find the old feature?


Suggestions and Requests / feature request: favourites
« on: October 17, 2007, 11:17:56 pm »

I would like to have something like "favourites".

Either as an own pinnable tab, or as a subpackage in the project explorer as a root node.

Inside this favourites I would place all my stuff
I am currently working on, e.g. diagrams, classes, docs, ...

This stuff in that folder are references to the original stuff and NOT a copy. Thus there wouldn't be a need to
drill down the model structure to find my current work every day.

Just work on the favourites folder and nobody would care about its real location in the model.

Of course there will be so problems, like refacturing the model, e.g. move one class from one package to another.
There must be an automatic update of the favourites internal structure, so the favourites folder is not broken.

But these are the sparxians problems to solve not mine  ;)


Suggestions and Requests / additional overlay for resource selection
« on: October 29, 2007, 02:08:14 am »

you can specify three types of people:

Authors, Resources and Clients.

For this you have three dialogs for defining them.

When assigning people to elements in the project management dialog, all members of the three groups are pooled(merged)into a comobobox labeled as "Resource".

1. Please rename the label "Resoure" into "People", because
that's what it currently is. It is not only a resource, but an author or a client too.

2. The information what it is (Author, Resource, Client) is lost.
So please provide an overlay icon or an additional letter in each entry in the comobobox (A, R, C) or ...
to be able distinguishing them in the combobox.



when you open the project management window, there are two bugs in my point of view.

1. When assiging an author to an element in the resoures allocation, the default role the author belongs to is NOT selected.

Instead nothing is selected.

This is wrong.
Why can I assign a role to an author in the settings | people| authors dialog, when it is not selected here as default?

Of course it must be possible to assign the author a different role in the project management dialog.

But the default must be selected!

2. When I assign a resource to an element in the resource allocation, I can assign a role to the resource.

But this roles are NOT available for the resources when the resource is defined in the settings | people | resource dialog.

And the role I assigned to the resource is not available here, so this functionality is totally confusing.

Try the following:

Create a resource called Mr. X with the role Scrum Master.

Goto the project management dialog and assign this Mr. X to an element.

-> The data role scrum master is NOT available.



when you add Authors, Resources and Clients in
Settings | People

the dialogs for Authors, Resources and Clients are different and must be homogenised.

I can't see, why a project author has no contact information like a resource or customer (phone, mobile, fax, email) .

I cant't see, why you can select authors from the windows user managment, but resources and clients not.

I can't see why I can defines roles in the project roles dialog, but I can only reuse them in the Authors dialog and not in resources and clients. This is cumbersome.

This is not usable and must be adapted, that each dialog offers the other possibilities too.


Suggestions and Requests / overlay icon of namespace root  not cleared
« on: October 17, 2007, 11:08:36 pm »

when clearing the namespace root of a package via context menu of the package, the overlay icon stays on package.

You have to select another package, than it is cleared.

Do you have the same behaviour?


Suggestions and Requests / notes and sequence diagrams
« on: October 24, 2007, 02:14:42 am »

I very often miss being able to attach a note with a notelink to a message in a sequence diagram.

That would be very helpful to us.


Pages: [1] 2 3 4