Re: intel compiler problem

From: Matthew Nobes (manobes_at_sfu.ca)
Date: 10/16/03 

Date: Thu, 16 Oct 2003 21:11:41 +0000 (UTC)

Richard Maine <nospam@see.signature> wrote:
> Matthew Nobes <manobes@sfu.ca> writes:

>> master_vertex_list(count1)%entry_list(count2)%term_list(count3)%bac_label = bac_mom_label
>> ^
>> Error 24 at (320:gluon_ops/plaq_act_0b_2d.f90) : syntax error

> This isn't enough context to tell much. The error message seems
> pretty uninformative, but based solely on what it is pointing to, it
> sounds somewhat like it doesn't recognize that these are derived
> type objects (and thus that the % doesn't make sense). But I could
> be off base on that.

> Declarations are absolutely critical to diagnosing compilation
> problems like this. Just seeing the isolated code line doesn't help
> much. I'd need to see the declarations of all relevant things to be
> of much help here...and preferably as part of a compilable sample
> instead of as isolated lines.

Okay, I think the problem is that I'm using allocateable arrays of
defined types. I have managed to duplicate the first of this long
chain of syntax errors with the following test code

**BEGIN TEST CODE**
module test
  implicit none
  
  type, private :: exponant_entry
     integer :: index
  end type exponant_entry
  type, private :: sub_list_entry
     type(exponant_entry), dimension(:), allocatable :: term_list
  end type sub_list_entry
end module test
**END TEST CODE**

I am trying to compile this with just the -c flag and I get

**SCREEN OUTPUT**
matt@immoland:~/work/sdb$ if90 -c test.f90
   module TEST

     type(exponant_entry), dimension(:), allocatable :: term_list
                                         ^
Error 24 at (8:test.f90) : syntax error

1 Error
compilation aborted for test.f90 (code 1)
**SCREEN OUTPUT**

I'm at a loss. Like I mentioned in my last post, this compiles with the
portland group f90 compiler without error. As far as I can tell it's
proper f90 style. 

-- 
Matthew Nobes
c/o Physics Dept. Simon Fraser University, 8888 University
Drive Burnaby, B.C., Canada
http://burbland.phys.sfu.ca/~matt

Relevant Pages

  • Re: Assignments in Modules
    ... implicit none ... end module test ... end program tester ... But.loading the data run time gives a fast compile: ...
    (comp.lang.fortran)
  • Re: Odd problem with Visual Studio
    ... I rebooted (so nothing of my test code is running). ... compile the code from the command line repeatedly without a problem. ... target directory, nor can the target file be deleted until I exist from the ... I am posting more detail in another posting as I have made some progress. ...
    (microsoft.public.dotnet.languages.csharp)
  • ld terminated with signal 11
    ... arm-gcc 2.95.2 and glibc 2.1.3. ... When I try to compile ... the simple test code shown below, the linker terminates with a signal ...
    (comp.arch.embedded)
  • Re: Assembler Editor vs. Editor Assembler
    ... ASM/ED is dog slow! ... I didn't do much assembly programming but if I wrote anything larger ... I used to test code after just about every line entered. ... After working with the ASM/ED cart I started writing routines that actually did stuff before I would test them for a clean compile. ...
    (comp.sys.atari.8bit)
  • Re: Testing classes individually
    ... >I remember that in Java you can compile and run each class individually ... I then instantiated my class and added some test code in the Mainmethod. ... Prev by Date: ...
    (microsoft.public.dotnet.languages.csharp)