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 - jandtr

Pages: [1]
1
Thanks Uffe, you made it all very clear.  :)
I followed your advice made the corrections and now all the tables are back in the Browser.
Too bad much of what you wrote is absent from the Sparx Documentation, especially the Profile 'influence'.

Thx

Jan

2
Hi Geert, Paolo,

@ Paolo yes the correction were executed the project reloaded same result some are invisible.

@Geert the Integrity checks dealt with connector (duplicate and wrong type) and one duplicate GUID on an ArchiMate_ApplicationComponent, I suspect an import or CSV upload. The correction was done, project reloaded, even a close and re-open, same result some are invisible.

What strikes me is that the visible tables all have t_object.Object_Type = 'Class' whereas the invisible ones have t_object.Object_Type = 'Table'.



GRB_ADP is invisible but GNBS_VlA2005 is visible.

Now, as I understand it the t_object.Object_Type is the same as EA.Element.Type, or am I missing something?

So when I do an csv export these two table both report CLass for Element.Type.?
What is happening here?

GNBS_VLA2005   Class{AD908E5F-3BB2-425d-A700-9BFD4BCB8C9F}   table   Implemented   EAUML::Table   EAUML::table   SQL Server 2012   
GRB_ADP   Class   {675834A8-2B9C-407f-ABD7-06F0BE70B2D2}   table   Implemented   EAUML::Table   EAUML::table   SQL Server 2012   

I can try now the obvious and with an SQL update set all the t_object.Object_Type to Class.
Especially since the t_object.Object_type on a different demo  DB I just uploaded all report Çlass for t_object.Object_Type.

Question then remains, how did itchange the t_object.Object_Type to Table?

Before I continue with the SQL, any thoughts?

Cheers,

Jan


3
Hi All,

This stumper of an issue is on v15.2 and v14.1.1428 ultimate versions

We're doing some development of micro services for encapsulating/obfuscating DBMS access. So we need the physical DBMS schema in the repository in order to use these in analysis and architecture diagrams describing the various micro services.

The DBMS schema evolves slightly and thus we do a 'show differences' in the DB Builder, this runs, show some minor differences, we import these and done. The generated diagram is somewhat clumsy, a half an hour later that is nicely formatted, the world looks great.

However when wanting to drag a table from the browser on a service architecture diagram I discover that the table is not present in the Project Browser, yet is clearly present in the diagram generated by the DB Builder. On closer inspection about 40% of the tables are not visible in the Project Browser.

When we search in the model these 'invisible' tables all show up in the result, we can ctrl+U and the diagrams are given correctly, the 'Find in Project Browser' however returns nothing.

We export CSV the Tables package and all the tables are present in the package csv export, even those 'invisible' ot visible in the Project Browser.

I write a quick script to dump the Package.Elements collection on the screen/file and again all the tables are present in the dump, also those not visible.

Last check via a SQL query on the repository (SQL Server) again all tables present, including the invisible ones.

Did an Integrity Check, some minor items, but nothing on the tables or the Database Package.
One peculiar thing was that when I did an XMI export of the Tables Package only the visible tables where exported.

Conclusion:
- all tables are correctly stored in the repository in the correct package;
- the generated diagram is complete and correct, the Project Browser is not having all tables.

A stumper if one wants to drag one of those ínvisible' tables onto a diagram, one cannot drag from diagram to diagram.

I've checked types, Stereotypes, FQStereotypes, Profile Types, PackageID, ...
Compared the diagram objects with the Package Elements and all that seems to be correct.
I've transferred repository to new server with new DBMS instance, went from DBMS based to eapx based, nothing helped.

Question is now: how to render those invisible tables again visible in the Project Browser.

Has anybody experienced this? And found a way out whitout reloading the DBMS again.
As a number of those tables are already in architecture diagrams an thus have a bunch of links and relations  to other elements in our repository, a clean reload anew from scratch would undo all that work, something I would like to avoid.

Has anyone an idea how Sparx populates the Project Browser? Can we script-activate this? And why some tables remain invisible, while apparently present in the Tables Package?

All help would be greatly appreciated.

Jan
 

Pages: [1]