Welcome to Telelogic Product Support
  Home Downloads Knowledgebase Case Tracking Licensing Help Telelogic Passport
Telelogic Rhapsody (steve huntington)
Decrease font size
Increase font size
Topic Title: Lattix and Rhapsody
Topic Summary: Request for info
Created On: 24-Jul-2008 10:37
Status: Read Only
Linear : Threading : Single : Branch
Search Topic Search Topic
Topic Tools Topic Tools
Subscribe to this topic Subscribe to this topic
E-mail this topic to someone. E-mail this topic
Bookmark this topic Bookmark this topic
View similar topics View similar topics
View topic in raw text format. Print this topic.
 24-Jul-2008 10:37
User is offline View Users Profile Print this message


Charlie Lane

Posts: 86
Joined: 16-Sep-2005

Does anyone have experience of using Lattix (Lattix web site) or Structure101 (Structure101 web site) with Rhapsody?

We're quite interested in getting a DSM view of a Rhapsody project and I've hacked a simple export utility to get dependency info from Rhapsody to Lattix for an evaluation trial.
Report this to a Moderator Report this to a Moderator
 24-Jul-2008 11:49
User is offline View Users Profile Print this message


Matthew Thomas

Posts: 15
Joined: 14-Feb-2008

Hi Charlie
Very interesting. Doesn't the Lattix solution work by analysing the code to produce the Dependency Structure Matrix (DSM).
Hence, one workflow would be: code generation -> Lattix analysis and re-architecting -> apply items in Lattix Workfile to model.
Or is there something more subtle amiss in my understanding.
Cheers
Matthew
Report this to a Moderator Report this to a Moderator
 28-Jul-2008 13:56
User is offline View Users Profile Print this message


Charlie Lane

Posts: 86
Joined: 16-Sep-2005

Hi Matthew, good to see your contributions in the forum again!

Yes, Lattix (and other similar tools) are based on the idea of looking at dependencies in code, so you can certainly use the workflow that involves code generation and then analysis of that code. Actually, you find that tools often rely on 3rd-party mechanisms to extract the dependencies: Doxygen, Microsoft BSC, etc.

However, there are a number of advantages of doing the analysis from the model rather than the generated code:
a) you can get a DSM even before you have something that compiles (the code analysis tools usually seem to require compilable code) -- this would typically be the situation at a preliminary design review,
b) you can get a DSM that covers all components/configurations in the model (e.g. our models often have different scopes for different builds, and it would be tedious to have to generate each one in order to get a DSM) -- this will be quite an advantage if overnight checking from a CM repository is done,
c) you can get a DSM from parts of the model that don't generate code (e.g. use cases, sequence diagrams, controlled files, etc) -- i.e. most of a Harmony/SysML model, and
d) you can get a DSM from a Rhapsody model that generates code that isn't currently supported by Lattix (e.g. your system to generate DOORS schema from Rhapsody -- I'll let you fill in the web address!).
e) It's quicker than generating code and then analysing that.

I'm also an advocate of doing as much as possible at model level as a general principle.

Regards, Charlie.
Report this to a Moderator Report this to a Moderator
 30-Jul-2008 11:25
User is offline View Users Profile Print this message


Matthew Thomas

Posts: 15
Joined: 14-Feb-2008

Hello Charlie
Thanks for you kind words and for making the advantages of deriving the DSM from the model clear.
Can you elaborate your requirements for the model-derived DSM? Would you want the same level of functionality you get with Lattix, i.e. matrix manipulation leading to a worklist? Did the hack you mentioned get you anywhere close to this?
Very best regards,
Matthew
Regarding "the other subjec"t I'll let you know when I can make an official announcement on that in cooperation with Telelogic (An IBM Company).
Report this to a Moderator Report this to a Moderator
 30-Jul-2008 21:03
User is offline View Users Profile Print this message


Charlie Lane

Posts: 86
Joined: 16-Sep-2005

Hi Matthew,

I didn't quite understand your question about "the same level of functionality you get with Lattix" because this was an experiment using Lattix, not an alternative to Lattix. To clarify, I wrote some VB that extracted dependency information from Rhapsody and dumped it into the file format that Lattix requires. We used Lattix to look at the DSM and manipulate the structure, leading to a worklist, which can then be used by the model designer to guide his changes to the model.
Actually, some of what we found didn't require manipulation of the structure in Lattix because it clearly showed where our model needed adjustment. For example we found that one unexpected (i.e. contrary to our intent) dependency from, say, class A to package B was because class A had a reception of event C which was defined in package B (wrongly, it should have been in the interfaces package). This is the sort of thing that would have resulted in a rule violation in Lattix if we had set up the rules for the intended project structure.
We also noted that the structure we set up in Lattix is approximately the inverse of what we set up in Rhapsody. What I mean is that in our Rhapsody models we like to have requirements at the top of the model, followed by architectural stuff, detailed design and then test harnesses. With Lattix the required organisation is to put the test harnesses and main builds at the top (nothing depends on them) and work down to the requirements (which are obviously what the whole model depends on).

So we've found some interesting things about our Rhapsody models from quite a short trial and I wondered whether other Rhapsody users had tried using Lattix or Structure101 with a model.

Regards, Charlie

Edited: 30-Jul-2008 at 21:05 by Charlie Lane
Report this to a Moderator Report this to a Moderator
 28-Aug-2008 09:01
User is offline View Users Profile Print this message


Charlie Lane

Posts: 86
Joined: 16-Sep-2005

Since the last posting I've also done a trial with Structure101, which also accepts a specialised XML format so could be driven with the same VB I did for Lattix, slightly modified.

Still hoping that someone else may have used Lattix and/or Structure101 with Rhapsody and may wish to comment.
Report this to a Moderator Report this to a Moderator
Statistics
20925 users are registered to the Telelogic Rhapsody forum.
There are currently 1 users logged in.
You have posted 0 messages to this forum. 0 overall.

FuseTalk Standard Edition v3.2 - © 1999-2009 FuseTalk Inc. All rights reserved.