Re: CVS commit: pkgsrc/devel/cmph

[jnemeth%victoria.tc.ca@localhost (John Nemeth)]

On Mar 4,  4:09pm, John Marino wrote:
} On 10/12/2012 21:10, Aleksej Saushev wrote:
} > John Marino<netbsd%marino.st@localhost>  writes:
} >
} > My point is straightforward, "-Werror" is there for reason.
} > Code is not something god blessed. If some compiler arrives that
} > provides more checks or performs more aggressive optimizations
} > and thus breaks traditional semantics code should be adapted.
} 
} Then the compiler is broken.  Although I will give you some leeway with 
} the C language itself being broken.
} With your logic, Werror should be on all packages, not ones that the 
} author decided it was necessary.

    I'm not totally sure what Aleksej meant here.  If he meant that GCC
deviates from the standard, then yes it is broken and should be fixed.
But if he meant that GCC liberally interperts things that are
implementation defined, then we are talking about code that is broken
and should be fixed.

} > GCC is notorious for becoming more and more aggressive over time in its
} > optimizations. If anything is allowed by language standard which has
} > drifted over time, GCC takes liberty of interpreting it in its own way.
} > Hopefully it emits a warning when it sees language usage "problematic"
} > in view of current GCC interpretation of language standard. The latter
} > may (and often does) differ from the standard applicable to the code.
} >
} > That's why I insist on either fixing the problem properly, or leaving
} > "-Werror" alone, or implementing your "fix" as a hack which it is in fact.
} 
} This is COMPLETELY unfair.  You've put me in a no-win situation.  My 
} original fix was to disable the warning that popped it.  Oh no, you 
} can't do that!  My compiler from 1995 wouldn't like it!  So then I 
} disable Werror which A) I was told to do and B) there is a ton of 
} precedent for it including on a similar package and C) people agree its 
} the best of the options.   Oh no!  You can't do that!  Some fantasy case 
} that I can't articulate might blow it up!
} 
} The only option you've left is to fix the author's crappy code.  That's 
} not what pkgsrc is supposed to do.  It's supposed to build packages.  If 

     Who says it isn't?  We fix/workaround bugs all the time.

} the package has bugs in it, kill the package.  I have no patience for a 

     Anything more complex then "hello world" is guaranteed to have
bugs.

} bad product, and I'm expecting the code to be released bug-free and to 
} be future-proof.  If that can't be delivered, it's subpar and worthy of 
} getting cut out.

     By this standard, we might as well just kill pkgsrc and be done
with it.

}-- End of excerpt from John Marino