I had been using my code(fortran 90) on linux openSuse, and it was working perfectly. Yet, now that I have switched to ubuntu and I have installed gfortran, it gives me error as sollows:

make

--- Compiling Fortran-90 source file [analysis.f90]
f95 -O3 -c analysis.f90
analysis.f90:18:4:

data vk /1,1,1, -1,1,1, 1,-1,1, 1,1,-1, -1,-1,-1, 1,-1,-1, -1,1,-1, -1,-1,1/
1
Error: DATA statement at (1) has more variables than values
makefile:21: recipe for target 'analysis.o' failed
make: *** [analysis.o] Error 1


I really appreciate if someone could help me with this problem. Thanks a lot.

Hi. For future reference it is best to include enough code in support of your question. And then also place code as well as things like command line outputs in within [code][/code] tags to maintain the spacing and formatting and also make it easier for readers of your questions.

The error report seems to echo the line in your source it has a problem with.

As with all syntax, you should look at that line, see if it matches correct FORTRAN syntax.

I feel you should check the space between "data" and "vk" and determine what it is and what it should be, whether it is an actual SPACE versus TAB.

This may be a newly edited line where the editor on your new system acts differently when you do certain keystrokes.

I'll move this to the Programming forum where I feel you'll get better assistance from readers. (Also renamed the question since "I" is not really helpful for people to see what the nature of your problem is.)

1 members found this post helpful.

It depends on the definition of variable 'vk'

My problem got resolved, thank you so much.

Hi,

Glad you did resolve it.

Two things which will help future persons who experience similar issues are:

1. If you can mark the thread as solved, which is in the Tread Tools menu which is part of the form
2. If you can also provide some statements as to what solved the problem that will be helpful for other future seekers of solutions:
   1. It could have been the definition of the variable
   2. It could have been the interpretation of white spaces
   3. It could have been an entirely different thing which resolved this, and this would be helpful for people to know

Best Regards!

3 members found this post helpful.

Yes: please take the time to "write for the future." Your post will be seen for years to come. Flag it as [SOLVED] and write a short "closing comment" that is intended to be beneficial to the next by giving him or her the that solved the same problem for you.

In particular, you imply that the same Fortran-90 source code was acceptable to Intel's compiler but not to Gnu's. That's very interesting. And, that's what we'd like to know more about from you in your closing-comments to this thread. (Anyone could be "stuck, and stuck for a very long time" by a subtle difference between two compilers.)

2 members found this post helpful.

My data had only 8 entries, whereas, I had double precision:: vk(3,10),vk1(3,10),vkr,ar, ai,theta, when I changed it to
double precision:: vk(3,8),vk1(3,8),vkr,ar, ai,theta, it worked perfectly.