On 21 Jul, David Malcolm wrote:
> On Fri, 2017-07-21 at 19:58 +0200, Volker Reichelt wrote:
>> On 21 Jul, David Malcolm wrote:
>> > On Thu, 2017-07-20 at 18:35 +0200, Volker Reichelt wrote:
>> >> Hi,
>> >>
>> >> the following patch introduces a new C++ warning option
>> >> -Wduplicated-access-specifiers that warns about redundant
>> >> access-specifiers in classes, e.g.
>> >>
>> >> class B
>> >> {
>> >> public:
>> >> B();
>> >>
>> >> private:
>> >> void foo();
>> >> private:
>> >> int i;
>> >> };
>> >>
>> >> test.cc:8:5: warning: duplicate 'private' access-specifier [
>> >> -Wduplicated-access-specifiers]
>> >> private:
>> >> ^~~~~~~
>> >> -------
>> >> test.cc:6:5: note: access-specifier was previously given here
>> >> private:
>> >> ^~~~~~~
{...snip...]
>> > If you're going to implement a fix-it hint for this, there should
>> be a
>> > test case that tests it (probably as a separate file, e.g.
>> Wduplicated
>> > -access-specifiers-2.C, to allow for the extra option -
>> -fdiagnostics
>> > -show-caret, covering just one instance of the diagnostic firing,
>> for
>> > simplicity).
>>
>> I actually did try that, but dejagnu somehow gets confused.
>> To me it looks like the reason for this is that both multi-line
>> messages
>> are so similar. I'll give it a second try, though.
>
> I'm not sure what you mean by "both" multi-line messages; shouldn't
> there be just one multi-line message for the fix-it hint.
>
> Isn't this like e.g. g++.dg/cpp0x/auto1.C ? (an example of a testcase
> that verifies a deletion fix-it hint)
>
> Dave
There are two multi-line messages. One for the warning and one for the
note, see the example above the "[...snip...]". The message after the
note consists of the first two lines of the message after the warning.
This seems to confuse dejagnu. However, I managed to work around this
problem. I'll post an updated patch soon.
Regards,
Volker
