Construct-name constraints

[OP1]

Does the fortran standard has rules regarding  construct-name conflicts with other entity names? I tried to figure this out, but given the size of the fortran standard it's quite likely that I overlooked the relevant part.

In other words, is a code like the one below compliant, or is ifx correct to issue the error message "error #6444: Not a valid construct name."?

INTEGER :: S1
S1 : BLOCK
END BLOCK S1
END

[Steve_Lionel · Accepted solution]

Yes, there are rules, in clause 19 (Scopes, association, and definition).  19.3.1 defines various "classes" of local identifiers.  Named variables and named constructs are both "Class 1" local identifiers, and paragraph 3 then goes on to say:

Within its scope, a local identifier of one class shall not be the same as another local identifier of the same class, except that a generic name may be the same as the name of a procedure as explained in 15.4.3.4 or the same as the name of a derived type (7.5.10). A local identifier of one class may be the same as a local identifier of another class.

Therefore, the compiler is correct in complaining, though I'd prefer that the message be clearer.

[Ron_Green]

Our Standards guru is out on vacation this week.  I can ask when he returns.  gfortran 13.2.0 accepts this code FWIW

As a data point, NAG does not like this syntax either.  I'm not saying it's the final answer, but NAG is pretty darn good at following the Standard:

nagfor block.f90
NAG Fortran Compiler Release 7.2(Shin-Urayasu) Build 7203
Error: block.f90, line 2: Multiply defined symbol S1
       detected at <end-of-statement>@S1
Error: block.f90, line 3: Inappropriate use of symbol S1
       detected at END BLOCK@S1
Warning: block.f90, line 4: Unused local variable S1
[NAG Fortran Compiler pass 1 error termination, 2 errors, 1 warning]