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 - heba

Pages: [1] 2
Bugs and Issues / "Prefer property dialogs" not working in sequence charts?
« on: November 16, 2018, 10:29:06 pm »
Hi all,

there is this feature "Prefer property dialogs" in Visual Styles which helps me to get back to the good old way to e.g. edit operations and attributes in classes:  it pops up the dialog window.

Now, when I have a sequence chart and double click a message I still get that new style docked property window. This is rather useless since it is too small in my setup and I have to scroll horizontally to get to the "Operations" button...

So, is there another to switch back to a dialog or should it work with the setting above and it is just a bug?


You can. In v14.1 there is an option to bring back the old features dialog.

Visual Style | Prefer property dialogs

Thank you!  That saved my day, I guess ;-)

Yes, this is the solution!  Thank you!

I never used this feature and I wonder how he switched it on.  Anyway, it works and now he can start working  ;)


Hello Geert,

The option does not show up in his EA.

My menu looks quite complete with some options greyed out:
 - Apply/release user lock...
 - Configure
 - Package Baselines
 - (greyed) Save package ...
 - (greyed) Load package ...
 - (greyed) View package ...
 - Compare package with XMI
 - Merge package with XMI
 - (greyed) Import a model branch
 - Version control settings

His menu:
 - Merge package with XMI
Nothing else.

Kind regards

Bugs and Issues / One user in the team not able to lock packages / elements
« on: December 14, 2017, 07:50:57 pm »
Hi all,

We work in a team with EA on a SQL server-based model.  Security is enabled, we use windows authentication and have a EA group with common permissions.  These permissions include the "Lock Elements" permission.

Now one user in the team cannot lock packages or elements while all other can. I checked the permissions, the group rights and did not find a difference to others.  I also created a local user for him but it still did not work.

Does anyone have ideas where I can look further into the problem?  I don't have any idea how to debug this .)  Maybe there are local settings which I'm not aware of?

Thanks in advance!

Hi all,

I have some diagrams with notes linked to elements (classes) and their content linked to the element feature "Element note". 
The class was created using reverse engineering from source.

After running reverse engineering again the class gets updated. Its element note remains as it is but text from the note disappears. :(

I can "repair" this by opening the dialog "Link this Note to an element feature..." and just pressing OK. So the text is not lost and there must be something wrong with the link.

Did anyone else run into this issue or has some more information?

Thanks in advance!

Automation Interface, Add-Ins and Tools / Create formatted element 'notes'
« on: February 08, 2017, 10:38:10 pm »
Hi all,

When I use the rich text features of the EA notes editor I get some HTML-like formatting in the element description, e.g. <ul> for the beginning of an unordered list or 
 for line breaks.

I want to generate element notes via automation which contain some formatting like this but I have not found a way to prevent EA from quoting the "&".  It seems that assigning  Element.Notes = "
"  always converts it to "
" which basically converts the formatting tags into plain text.

Does someone know how to mark the '&' as special char not to be converted?

Thank you!

Okay, thanks a lot for pointing this out.

If I remove the name of the CallBehaviour action the project browser also shows the name of the called behaviour so this correction does not look too bad.   I think I can live with that and will remove the names of all these elements.



in my activity charts I sometimes use "Call behaviour" action elements.

Before EA 13, they had the name resp. caption "<element name>: <activity name>".  For example, when I name the activity "GetLock" and the element "Call" it read "Call: GetLock".

With EA 13, the elements only have their names as captions but do not show the name of the activity where they are an instance of.  In the above example, it just reads "Call" now.

Is there a new option to show or hide the name of the base element?   When I open the model in an older EA version it appears with this name.

Thank you!

They are stored in the file %appdata%\Sparx Systems\EA\paths.txt

Yes, but should we really use system file access to read these variables? I was hoping to find something more internal...

Passing the variable to the "dir" parameter in the "ExtraOptions", e.g. "Dir=%TRUNK%\\generated\\" does not work. It generates but the files do not seem to appear anywhere :o   Although this is not documented as supported, could it be that it should work but is a bug?  :)


Ok, thank you!  I managed to get this to work.

Now I'm left with one final question for which my searches did not return any useful result:

To use relative paths our classes normally have a local variable like %TRUNK% in their Filename.  Is it possible to query this variable's value using the API?

Btw, I wonder why GenerateClass() does not use the "ExtraOptions". Some parameters like overwrite, dir or even lang would be useful - at least to me ;-)


Thank you!

Am I right that I cannot pass my desired language even for the class?

  GenerateClass (string ElementGUID, string ExtraOptions) --> ExtraOptions: [..] currently unused

and GeneratePackage () does not provide such a feature in the ExtraOptions.

I suppose the language is taken from the GenType property and the only solution is to change that before generating?  Which again means I have to lock the package if security is enabled, right?  That could make things more complicated  ???


Hi all,

I want to generate code for classes using an automation script and I'm searching for the respective method.  There should be something like Element.Generate(File, Language) but I cannot find it. And I'm quite sure that I have seen something like this  :-\

Use case is to generate a package resp. some elements of it two times with a different language in each run.


General Board / Re: What is the base template for "typedef"?
« on: May 02, 2016, 06:18:32 pm »
Thank you, Simon!  For some reason I never saw these overrides. They look very promising to what I want to do :)


General Board / What is the base template for "typedef"?
« on: April 23, 2016, 12:53:03 am »
Hi all,

I have a question regarding code generation.

In my model I have a UML class "MyNiceClass", stereotype = "struct" with some attributes.  I set the tagged values "typeDef" = "true" and "typeSynonyms" = "MyNiceClassType".  Now when I generate C code I mainly get what I expect:

typedef struct MyNiceClass {
  /* some attributes with doxygen documentation */
} MyNiceClassType;

So attributes and the attribute notes have been modified by me and this works well.  What I now want to do is also modify the "typedef struct" (the "frame" of the attributes) but I cannot find any template to modify.  Nothing in the "class" area influences the result here. 

Does anyone know what the correct base template is or if EA has some internal magic processing these "typedef = true" classes?

Many thanks!

Pages: [1] 2