[Developers] Re: [CactusMaint] IDAxiBrillBH no longer compiles on Absoft F90

[Jonathan Thornburg]

Hi,

>> This is on the Mac 'norah.aei.mpg.de', running Darwin 7.9.0,
>> Absoft Pro FORTRAN 7.0
>> ---------------------------------------------------------------------
>> -- f90 -N11 -s -O -c -o $current_wd/IDAxiBrillBH.F.o
>> $current_wd/IDAxiBrillBH.f
>>
>>          write(message_buffer, '(A,I)')
>>                                     ^
>> cf90-306 f90fe: ERROR IDAXIBRILLBH, File =
>> /Users/swhite/Cactus/configs/new/build/IDAxiBrillBH/IDAxiBrillBH.f,
>> Line = 277, Column = 36
>>    Integer constant expected where ")" is present.
>
> Saying "I" to output an integer variable is a Fortran 90 feature.  From
> its file extention, IDAxiBrissBH.F is a Fortran 90 source code, hence
> it should work.  I call the compiler broken, but there is at least one
> other such compiler (PGI, if I remember).
>
> You can say instead "I6" or something, where the number is the field
> width, i.e., the number of digits that should be output.  You could
> also ask Jonathan for help, I think he may have introduced this.

I was indeed the one who introduced this.  I had _thought_ that
it was ok to omit the field width, and the Intel fortran compiler
seemed to grok that ok.  But a quick check in Metcalf & Reid's Fortran 90
book doesn't mention that feature, so I just committed a fix to say
"I8" in all such places.  Hopefully that should make the code
strictly-legal and ok on all compilers...

ciao,

-- 
-- Jonathan Thornburg <jthorn at aei.mpg.de>
    Max-Planck-Institut fuer Gravitationsphysik (Albert-Einstein-Institut),
    Golm, Germany, "Old Europe"     http://www.aei.mpg.de/~jthorn/home.html
    "Washing one's hands of the conflict between the powerful and the
     powerless means to side with the powerful, not to be neutral."
                                       -- quote by Freire / poster by Oxfam