Hello Oliver, I am using gcc version 6.2.0 20161005 (Ubuntu 6.2.0-5ubuntu12).
Most of the warnings do seem useful. Several return statements that are indented as if they belong to a python if/for loop even though they are outside the execution block is one of those things. Another is in m_abs.cc, where the indentation is such that it looks like an early if-statement goes on throughout the rest of the code even though it is just a check. I know we support several compiler. I hope they don't disagree in their warnings. This last iteration is just a bit nutty with all the text it generates for the warnings. I think some warnings are even repeated every time the code is included somewhere. //Richard 2016-11-07 22:55 GMT+09:00 Oliver Lemke <[email protected]>: > Hi Richard, > > In my opinion, the right solution would be a mixture of 1) and 2). If > possible, warnings should be fixed because usually there is a good reason > for them. However, if the warnings are completely uncritical (such as > indentation level), I think we should turn them off. The problem nowadays > is that every compiler in every version comes up with new things they warn > about (often useful, sometimes not). It is almost impossible to fix them > all. Even turning them off selectively is not easy. Because the compiler > options to turn them off are of course compiler-specific, one has to put > endless version checks in the cmake control file to turn them off > selectively. > > I'm happy (and usually try) to fix warnings if I see them. However, > currently I don't have access to a 16.10 installation. Which gcc version is > Ubuntu using now? Maybe the version is available in MacPorts, then I'm > happy to have a look at them. > > cheers, > /oliver > > > > On 7 Nov 2016, at 12:19, Richard Larsson <[email protected]> wrote: > > > > Hello all, > > > > The latest gcc iteration on Ubuntu is complaining on a level that it did > not do before when compiling ARTS. It lists indentation warnings, > conversion warnings, guarding warnings, and others. This floods the output > and hides real errors for me. > > > > I am somewhat confused by all these additional warnings and they hide > real problems. So I would like to ask if there is a way to: > > > > 1) Make the arts build scheme turn off the warnings by default? > > > > 2) Fix the warnings and enforce to follow the new coding standard > suggested by gcc? > > > > I assume this is just an option in the gcc compiler flag-list that they > considered "stable" this year but not last year. > > > > I am in favor of the latter but do not mind either way. I just want the > warnings gone since they are in the way of real errors. > > > > With hope for input, > > //Richard > > _______________________________________________ > > arts_dev.mi mailing list > > [email protected] > > https://mailman.rrz.uni-hamburg.de/mailman/listinfo/arts_dev.mi > >
_______________________________________________ arts_dev.mi mailing list [email protected] https://mailman.rrz.uni-hamburg.de/mailman/listinfo/arts_dev.mi
