How can I get the error message from sldvrun when programmatically using Simulink Design Verifier?

I am using sldvrun in a script to run design error checking analyses on several models. When models encounter issues during the analysis, I want to record these errors so I can go back later after all the models are processed, and see if they can be addressed.

The sldvrun function will output the result of the analysis as 0 if there is an error, however, as far as I can see, the actual error is shown in the UI only. The information I am looking for is in green in the image below. Is there a way to get this information programmatically? A try/catch with sldvrun did not work.

Answer 1

[Pat Canny]

Hi Monika,

Apologies for the delay.

You can use the ERRMSGS return field with sldvcompat:

[STATUS, ERRMSGS] = sldvcompat(MODEL)

returns 1 (true) if the Simulink model MODEL is compatible with Simulink Design Verifier and 0 (false) otherwise, and also returns a list of incompatibilities.

For example, if your model uses 'variable step solver' then the returned error msg has following format.

struct with fields:

source: 'solver_sel'

sourceFullName: 'solver_sel'

objH: 0.0033

reportedBy: 'simulink'

msg: 'Simulink Design Verifier cannot be used with a variable-step solver. You must configure the solver options for a fixed-step solver. ?See documentation.'

msgid: 'Sldv:Compatibility:UnsupSolver'

Note: this is available in the "help sldvcompat" documentation, but not in the Documentation page for sldvcompat. I have passed this discrepancy along to our Documentation team.

Another option is to use the Simulink Diagnostic Viewer API's diary method.

An example:

model = 'sldvdemo_flipflop' 
open_system(model)
set_param(model,'SolverType','Variable-step')
sldiagviewer.diary('errorlog.txt')
sldvrun(model,[],true)
sldiagviewer.diary('off')
edit errorlog.txt

This will capture every message (as plain text) sent to the Diagnostic Viewer while logging is on, whether or not it is produced by Simulink Design Verifier.

[Monika Jaskolka]

Hi Pat,

The first option isn't ideal for me because I don't want to run sldvcompat. The reason being that subsequently running sldvrun performs the compatibility check a second time, which isn't time efficient. I have larger models where the compatibility check takes quite some time.

The second option worked, but I had to add in

sldiagviewer.diary('on');