Code review comment for lp:~maddevelopers/mg5amcnlo/3.0.1

Hi all,
I have just noticed this (fks_singular, line 3818)
      call Qterms_reduced_timelike(j_type,i_type,ch_i,ch_j,t,z,Q)
it should be instead
      call Qterms_reduced_timelike(j_type,i_type,ch_j,ch_i,t,z,Q)
(switch ch_i <--> ch_j), shouldn't it?

It may be that in the end there is no effect, since, for the relevant cases when either ch_i or ch_j ==0 and the other is not one has always Q=0

c g/a ->qqbar splitting
         Qterms(1) = 4d0 * TR * z*(1d0-z)**2
         Qterms(2) = 4d0 * dble(abs(col1)) * ch1**2 * z*(1d0-z)**2

      elseif ((abs(col1).eq.3 .and. col2.eq.8) .or.
     & (dabs(ch1).gt.0d0 .and. dabs(ch2).eq.0d0)) then
c q->q g/a splitting
         Qterms(1) = 0d0
         Qterms(2) = 0d0

Let me know

Cheers,

Marco