forums.silverfrost.com :: View topic

replica nfl jerseys replica nfl jerseys soccer jersey replica nfl jerseys forums.silverfrost.com :: View topic - Run-time Error 11

forums.silverfrost.com
Welcome to the Silverfrost forums

FAQ

Memberlist

Usergroups

Profile

Run-time Error 11
Goto page Previous 1, 2

forums.silverfrost.com Forum Index -> Support

View previous topic :: View next topic

Author

Message

star2066

Joined: 27 Jan 2015
Posts: 19

Posted: Wed Jan 28, 2015 5:13 am Post subject: Re:

davidb wrote:

mecej4 wrote:

Star2066: Old code such as the one that you posted, with dummy argument arrays with declared size of (1), as in your subroutine RANK, will not behave properly when you use a checking option such as /CHECK.

I would leave checking on. The compiler should be able to cope with the (1) sized arrays if the option \OLD_ARRAYS is enabled. In Plato this is on the language tab (Allow old arrays) in the Properties window. It effectively changes (1) to (*).

I didn't use your way, but thx for your reply.

mecej4

Joined: 31 Oct 2006
Posts: 1901

Posted: Wed Jan 28, 2015 1:21 pm Post subject: Re:

star2066 wrote:

Thank you for you answer, it works. But I have a new problem:

*** Error 112, Reference to undefined variable, array element or function result (/UNDEF)

TEST - in file freeformat2.f95 at line 39 [+0efb]

To compile with checks at the command line, use the /check or /undef options. In the IDE, select corresponding options.

In your main program, change

Code:

write(12,*)' x rank=', xr,' y rank=',yr

Code:

write(12,*)' x rank=', xr(:n),' y rank=',yr(:n)

The "reference to undefined" runtime error occurred because the subroutines assigned values only in the subscript range 1:n. Attempting to print the whole array causes an abort when xr(n+1) is referenced.

Last edited by mecej4 on Wed Jan 28, 2015 3:30 pm; edited 1 time in total

mecej4

Joined: 31 Oct 2006
Posts: 1901

Posted: Wed Jan 28, 2015 1:22 pm Post subject: Re:

You may not be aware that FTN95 comes with a subroutine, RSORT@, for creating a rank index array for a real array. Using that subroutine, your program can be abbreviated to

Code:

program test
implicit none
real :: r1,r2,c1,c2,e,x
integer, parameter :: NN = 50
real, dimension(NN) :: xp, yp
integer, dimension(NN) :: ir,jr
integer :: i,n1,n2,n
print *,'input numbers before change point to continue'
read *, n1
print *,'input numbers after change point to continue'
read *, n2
print *,'input coeff of rel before change point to continue'
read *, r1
print *,'input coef of rel after change point to continue'
read *, r2
c1 = sqrt(1-r1**2)
c2 = sqrt(1-r2**2)
write(*,*) 'c1=', c1,'c2=', c2
do i = 1,n1+n2
call random_number(e)
call random_number(x)
xp(i)=x
if (i<n1) then
yp(i) = x*r1+e*c1
else
yp(i) = x*r2+e*c2
endif
end do
n=n1+n2
call rsort@(ir,xp,n)
write(*,51)'x-price',(i,ir(i),xp(ir(i)),i=1,n)
call rsort@(jr,yp,n)
write(*,51)'y-price',(i,jr(i),yp(jr(i)),i=1,n)
51 FORMAT(/A,/,(i3,2x,i3,2x,F12.4))
end program test

davidb

Joined: 17 Jul 2009
Posts: 560
Location: UK

Posted: Wed Jan 28, 2015 9:09 pm Post subject: Re:

star2066 wrote:

I didn't use your way, but thx for your reply.

No worries star2066. I hope you made some good progress anyway.
_________________
Programmer in: Fortran 77/95/2003/2008, C, C++ (& OpenMP), java, Python, Perl

star2066

Joined: 27 Jan 2015
Posts: 19

Posted: Thu Jan 29, 2015 12:28 am Post subject: Re:

mecej4 wrote:

You may not be aware that FTN95 comes with a subroutine, RSORT@, for creating a rank index array for a real array. Using that subroutine, your program can be abbreviated to

Code:

I run your code, one thing, the xp and yp are same every time. should we use random seed(0) first?

mecej4

Joined: 31 Oct 2006
Posts: 1901

Posted: Thu Jan 29, 2015 12:38 am Post subject:

Quote:

I run your code, one thing, the xp and yp are same every time. should we use random seed(0) first?

Perhaps; does this remark not apply to your original program, as well? What is the objection to starting with the same seed every time?

I do not know the purpose of your program, and my comments and suggestions were weighted towards the programming aspects, and to indicate how you could concentrate on your problem proper by leveraging available library routines.

In a program that uses pseudo-random numbers, it is useful to start out with code that generates the same sequence every time, which makes debugging easier, after which you can generate a seed using the system clock reading or something similar.

Display posts from previous:

	forums.silverfrost.com Forum Index -> Support	All times are GMT + 1 Hour Goto page Previous 1, 2
Page 2 of 2

You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot vote in polls in this forum