UnsupportedType should not affect other variables

[rupertford]

In the current implementation, any variable that comes after an unsupported variable (in the list of a type declaration) will also have UnknownType. This should be fixed.

For example ...

real :: a=1.0, b will result in both a and b being of unknown type, however, real :: b, a=1.0 would only result in a being of unknown type.

[sergisiso]

This will be solve by:

• [ ] #1419
• [ ] #2137
• [ ] #2138

[arporter]

Hugo has also hit this when we have an array specified using unsupported dimension expressions, e.g.:

CODE_UNSUPPORTED_FTYPE_SCALAR="""subroutine foo()
     real*4 :: a(-1:10, 12), cff1
     a(1) = cff1
end"""

Issue #2138 will 'fix' this particular case by allowing the array declaration to be handled successfully.