feature request: error-parser for external tools

Hello,

I use Mentor ModelSim to simulate VHDL code.
Therefore I configured sigasi to call ModelSim as an external compiler like descriped on help pages.

ModelSim output is displayed in eclipse console.
But of course without further efforts ModelSim output is not parsed for errors or warnings.

Is there a way to get output of external tools parsed and have errors and warnings added to sigasi error view ?

thx
Josef

coming up

Hi Josef,

We will make the ModelSim output "clickable" in the upcoming release. This means you can navigate to the correct file and line number by just clicking the error message. In the future, we will also mark the errors in the editor window, like the errors and warning our internal analyzer generates.

Philippe

great

Dear Philippe,

Thanks for this.
I'm really impressed of sigasi development team's work.
User suggestions are often considered instantly.

thx
Josef

modelsim libraries are not mapped

I tried the 20091127 release. Linking the compiler warnings to the source code seems to work. However, the directories are not linked to VHDL libraries (modelsim vmap command)

example from our compile scripts:
if (! (-d vhdl_build/tools_lib)) then
vlib vhdl_build/tools_lib;
endif
vmap tools_lib ${tests_dir}/vhdl_build/tools_lib

Work around: create a modelsim.ini file yourself with the correct mapping.

By the way: I had to do some research to find the enabling of the Makefile generation:
HDT / windows / preferences / VHDL / Makefile
And calling the compiler: define external tool, name = , location = , working directory = ${project_loc}, Arguments = -f ${project_loc}/Makefile.vsim all

Thanks, Jan-Willem for your

Thanks, Jan-Willem for your input. I'll check if we can integrate your suggestions in our next release.

We want to have a "run modelsim" button somewhere, but for now, the external builder you describe is probably the best way to do it.

kind regards

Philippe

Post new comment

The content of this field is kept private and will not be shown publicly.
By submitting this form, you accept the Mollom privacy policy.