Re: [Qucs-devel] QUCS help

[Raimund 'R. J. <ra...@lk...>]

Ced wrote:

evening, folks!

> implementation (bipolars mainly). To be honest, I'm
> not a programmer guru, but I may give you a hand in
> some not too in depth programmings areas.

go ahead and pick something:

http://qucs.sourceforge.net/road.html

i personally would add another item: qucs needs a regression test
collection that allows further improving the simulator without risking
the loss of existing functionality.

i have some very specific ideas on how such a test framework should work
(implemented one at work). i can write that down once someone volunteers
to implement something. but at the bottom there is a programm that has
to serve as the "test oracle": given two simulation results it has to
decide if they are "equal".

a regression test consists of a collection of schematics/netlists and
known results. the test suite runs the simulations and compares the
actual results against the expected ones, failing on mismatches.
however, the tricky part is that the test oracle cannot simply compare
the simulation results byte-by-byte since minor (acceptable) differences
may occur due to sunwind, algorithm improvements or hardware differences.

so, this part of the task consists of a qucs dataset parser and simple
"diff" ...

interested,
	Raimund

-- 
                  ___    ___   _____________
                 /  /|  /  /_ /  ____/  ___/\      Nothing useful for
                /  / / /   _ /  / __/  / __\/      more than a decade
               /  /_/_/    \/  /_/_/  /_/
{www.|raimi@} /______/__/\._\.____/\.____/\ .org
              \._____\._\/\._\.___\/\.___\/