Dynamic array names

[redrighthand]

Hi all

My question is that, is there a way to use a character as a name of a real array? In the following example "dir" is a character. According to the value of "idir", "dir" becomes, x, y or z. In the last statement, I am calculating "hgt" using the real arrays(x,y, or z arrays). x, y, and z real arrays are already defined in the code. Is there a way out for this problem.all I want to do is to use a dynamic name for the real arrays.

idir=2 ! 1=x, 2=y, 3=z
if (idir ==1) then
istart=istep
iend=istep2
write(dir,'(a)')'x'
elseif (idir == 2) then
istart=jstep
iend=jstep2
write(dir,'(a)')'y'
else
istart=kstep
iend=kstep2
write(dir,'(a)')'z'
endif

hgt = dir(iend)-dir(istart)

 

[xwb]

Are x, y and z the same size? If they are, you could use a 2D array. Say they have 20 items each

real, dimension(3,20):: dir
integer, parameter:: x = 1, y = 2, z = 3

... some code to set up istart and iend

hgt = dir(idir,iend) - dir(idir,istart)

 

[redrighthand]

xwb,

My arrays are not the same size, but i can't foresee a bad thing about setting them equal. . Your suggestion can really be helpful, thanks a lot. However, if somebody still has the solution for dynamic array names and share it, I would really appreciate it.

Thanks

 

[ArkM]

Try Fortran pointer/target attributes.
See. for example:

http://aeneas.ps.uci.edu/absoft/c12.htm

or google more links on "fortran pointer" topic.

 

[redrighthand]

Thanks ArkM
I followed your suggestion about pointer/target, and it worked so easy and great for me.
Thanks again