Re: XL Fortran on PowerPC- or x86-arch

Thomas C. Lang <tcl@xxxxxx> wrote:

format descriptors without field-length.

If you find *ANY* f90 compiler that does not at least have an option to
complain about a missing field length in an edit descritor in a format
statement, submit a bug report to the vendor. The standard requires the
capability to diagnose that kind of nonconformance in a format
statement. I just now tried

program me
write (*,100) 42
100 format(i)
end

with all 3 of the f90 compilers that I have on this system right now
(nag, g95, gfortran). All 3 bitched with default compiler options. No
bug reports needed there.

If the format is a character expression instead of a format statement,
the standard doesn't require diagnosis, though som ecompilers do it in
the simple and common cases (like literal constants). With the same 3
compilers, I tried

program me
write (*,'(i)') 42
end

and they also all bitched, though with different severity levels.

Or in case one uses a present()
statement and checking for the value of an optional argument in the same
if-statement.

If I have correctly parsed what you are saying there, along with your
comment about sometimes getting informed via program crashes, this error
wasn't "detected" at all. You might have run into a distinction between
whether or not the compiler short-circuits expression evaluation. That's
not a matter of being "strict" or not. That's just a happenstance of
particular implementation choices, all of which are valid. But maybe I
misunderstood, because I did have a bit of trouble interpreting that
statement.

Yes, xlf does have a reputation for being fairly strict (or "picky",
depending which slant one wants to put on it). To constructively answer
your question, the Nag compiler is also on the strict/picky side.

--
Richard Maine | Good judgment comes from experience;
email: my first.last at org.domain| experience comes from bad judgment.
org: nasa, domain: gov | -- Mark Twain
.

Relevant Pages

  • Re: vm02 update
    ... Instant Pascal was a misguided attempt at creating an IDE with automatic formatting and almost instant feedback. ... Editing the source in a tokenized format would keep the memory constraints lower and ease the first pass of the compiler. ... compiled to machine code at all. ...
    (comp.sys.apple2.programmer)
  • Re: OT: ASCII files and Intel P4 MMX instructions
    ... ASCII ... The file format is clean, one number per line, no fuzz. ... While I have played a bit with the settings in the compiler, ... are often dominated by the conversion from and to the external character ...
    (comp.dsp)
  • Re: replace statement and free format
    ... doesn't) that a vendor PRODUCE a compile-time listing of the source code. ... Free form reference format allows for each source line to have 255 characters ... - Depending on how many "passes" your compiler does of source code, ...
    (comp.lang.cobol)
  • Re: Deleting un(?)reachable code
    ... CDR form. ... Usually when I get these "deleting unreachable code" errors its because I did something stupid, generally rather far from the code which is flagged. ... Why does the compiler single out the format strings as ...
    (comp.lang.lisp)
  • Re: Next project, C programming problem.
    ... > Bill Gunshannon wrote: ... > condemn VMS for a non-existent problem is stupid and pejorative. ... > COBOL source files onto your VAX in a format that the VAX COBOL ... > file that does the necessary stuff before invoking the compiler. ...
    (comp.os.vms)