residu() 1st example fails on h2norm() since Scilab 5.3.x (REGRESSION)
Reported by Samuel GOUGEON (@sgougeon)
BUG DESCRIPTION:
----------------
s = poly(0,'s');
H = [s/(s+1)^2,1/(s+2)];N=numer(H);D=denom(H);
w = residu(N.*horner(N,-s),D,horner(D,-s)); //N(s) N(-s) / D(s) D(-s)
sqrt(sum(w)) //This is H2 norm
h2norm(tf2ss(H))
ERROR LOG:
----------
With 5.3.3:
----------
-->h2norm(tf2ss(H))
Warning !!!
Scilab has found a critical error (EXCEPTION_ACCESS_VIOLATION)
with "linmeq" function.
Save your data and restart Scilab.
!--error 10
Multiplication incohérente.
at line 23 of function h2norm called by :
h2norm(tf2ss(H))
With 5.5.2:
----------
-->h2norm(tf2ss(H))
!--error 10000
h2norm: Wrong type for argument #1: In continuous time expected.
at line 18 of function h2norm called by :
h2norm(tf2ss(H))
HOW TO REPRODUCE THE BUG:
-------------------------
See above
OTHER INFORMATION:
------------------
Was still OK with 5.3.0.
Bugs with 5.3.3
Still bugs with 6.0.1