fails to compile the valid code

[Richard E Maine]

On Nov 22, 2005, at 4:52 PM, Uttam Pawar wrote:

> gfortran fails to compile the valid code (which has structure
> constructor and generic name)

As others have noted, this is not valid f95 code. I base that on the 
standard, rather than on how various compilers happen to behave. The 
xlf message cited from xlf in a later posting gives the relevant 
restriction of the standard. In general, you usually can't use the same 
name for two different things in the same scope. (The actual statement 
of the conditions in the standard goes on for most of a page, but it 
boils down to that as the simple version, with some exceptions).  There 
are a few exceptions, but not many. As the clf messages mentions, one 
of the exceptions is that you can use the same name for a generic and 
specific, but that's not what this code does.

F2003 adds another exception, which *IS* what this code does. In f2003, 
a generic can have the same name as a derived type, essentially 
providing a way to override or extend structure constructors (since 
function references look an awful lot like structure constructors). So 
perhaps you might want to consider this as one of your f2003 feature 
test cases, but it isn't valid f95 code.

-- 
Richard Maine                |  Good judgment comes from experience;
Richard.Maine@nasa.gov       |  experience comes from bad judgment.
                             |        -- Mark Twain