Re: [RFA] New testcase to evaluate Fortran substring expression

[Wu Zhou <woodzltc at cn dot ibm dot com>]

On Sun, 31 Jul 2005, Daniel Jacobowitz wrote:

> This is looks very good!  Two comments for you:
> 
>   - The magic (and undocumented) constants are not a good idea.  Rather
>     than being clever with abs(), how about using an enum saying what
>     sort of range it is?

What about adding the following enumeration definition to f-lang.h and 
keeping the use of abs to get the number of arguments following the F90 
subrange expression:

/* This enumeration type is to identify the sort of F90 subrange expression.
   If only the low bound is by default, set it to -1; if both bounds are by
   default, set it to 0; if only the high bound is by default, set it to 1;
   if no bound is by default, set it to 2.  The absolute value of the value
   is also the number of arguments following this expression */

enum f90_range_type
  {
    LOW_BOUND_DEFAULT=-1,
    BOTH_BOUND_DEFAULT,
    HIGH_BOUND_DEFAULT,
    NONE_BOUND_DEFAULT
  };

If use other values, it might looks ugly to me to set the nargs based on 
the type value respectively.  What is your point on this?  

> 
>   - You have a bunch of lines which are too long in eval.c; could you
>     fix that, please?

Sorry, I am not very sure what is the maximum limit of a line in the sources
and comments.  Maybe it is 72? or 80, or any other number?  Maybe I can 
choose to use the smaller one?  
 
Regards
- Wu Zhou