Re: Trailing semicolon at the end of namespaces

Hi Petr,

Yes you are right, I didn’t know that! We’d better be standard conformant and remove the semicolons!

I actually used to think that namespaces need a semicolon just like classes. Recently I accidentally figured out that was not true when I forgot to write the semicolon in the packet-gen templates.

It would actually be great to always add -std=c++98 -pedantic to the compiler flags when using gcc under linux. Actually putting the following in tester.pri, for example, will do the job:

linux-g++{
QMAKE_CXXFLAGS += -std=c++98 -pedantic
}

Aside from complaining about the semicolons, the compiler complained a lot about long long which seems to be non-standard. I don’t know how we could work around the long long… it would have been nice to be able to compile with full standard conformance.

I digged an old post from Atul talking about some other compiler switches (see http://www.calitko.org/source-talk/583) which might be interesting to check.

Regards,

Peter

Would you like to post a relpy?


This post is a reply to:
Trailing semicolon at the end of namespaces
Hi Peter, I've just noticed (haven't paid attention to it before) that we have an extra semicolon at the end of all namespaces but there should be no semicolon (according to (more...)

Follow-ups:
Re: Trailing semicolon at the end of namespaces
Hi Peter, It would actually be great to always add -std=c++98 -pedantic to the compiler flags when using gcc under linux. Actually putting the following in tester.pri, for example, will do (more...)