In article <754165592snz@ncc1701.demon.co.uk> STeve@ncc1701.demon.co.uk writes:
>In article <2cu1uk$bm7@avdms8.msfc.nasa.gov>
>           brandon@gaia.msfc.nasa.gov "Brandon S. Dewberry" writes:
>
>[stuff deleted]
>
>>
>>  CVS_SYSTEM cvs();
>>
>                 ^^
>
>The parentheses are the problem. Stroustrup's reference manual
>(r.8.4 Initializers) states that the above "is not a declaration of an object
>of type [whatever], but a function taking no argument and returning
>[whatever]".

I wonder if it had ever occured to anyone on the C++ standardization
committee that this particular (unfortunate) syntactic ambiguity could
have been eliminated from the language simply be requiring that functions
accepting no arguments be declared with an argument list of `(void)'.

Anyone know if this point was ever discussed?

(Follow-ups to comp.std.c++ please.)

--

-- Ronald F. Guilmette, Sunnyvale, California -------------------------------
------ domain address: rfg@netcom.com ---------------------------------------
------ uucp address: ...!uunet!netcom.com!rfg -------------------------------