Subject: Re: [boost] Bug report rejected as conformant
From: Edward Diener (eldiener_at_[hidden])
Date: 2018-01-13 21:13:27
On 1/13/2018 3:40 PM, Olaf van der Spek via Boost wrote:
> On Sat, Jan 13, 2018 at 7:45 PM, Edward Diener via Boost
> <boost_at_[hidden]> wrote:
>> I recently reported a preprocessor bug in Oracle C++ 12.6 on their online
>> forum when compiling a C program example. I even cited the C11 standard in
>
> What bug is it about?
Oracle C++ fails two preprocessor tests. I boiled it down to:
#define SOME_MACRO(x,y) some_result
int main() { SOME_MACRO(); return 0; }
Of course calling SOME_MACRO without two arguments is a preprocessor
error. Oracle C++ gives only a warning.
My point is not about an actual bug but that a compiler decides a bug is
not a bug because it gives a warning rather than an error. I find this
ludicrous. I know that the VC++ preprocessor is much worse and is not
even remotely C or C++ standard conformant. But no one from Microsoft
would seriously argue that their preprocessor conformed to the C++
standard. The fact that the Oracle developer actual argues this case is
what bothered me and caused my to give up trying to test that compiler
or report anything to them.