Hi Art,

So why JS said it is turned "on" by default? I'm in favor of leaving it "off".

Sukender
PVLE - Lightweight cross-platform game engine - http://pvle.sourceforge.net/


Le Tue, 30 Dec 2008 10:45:28 +0100, Art Tevs <stud_in...@yahoo.de> a écrit:

> Hi all,
>
> some longer time ago there was a discussion about the warning levels. Robert 
> wasn't happy about using very pendantic warning levels, since it could 
> introduce more errors than we currently have. Maybe take a look into list 
> archives.
> Hence I am not really sure that OSG_USE_AGGRESSIVE_WARNINGS should be turned 
> on per default, because this is an opposite of what Robert want to have...
>
> Cheers
> art
>
>
> --- Sukender <suky0...@free.fr> schrieb am Di, 30.12.2008:
>
>> Von: Sukender <suky0...@free.fr>
>> Betreff: Re: [osg-users] Warning level
>> An: "OpenSceneGraph Users" <osg-users@lists.openscenegraph.org>
>> Datum: Dienstag, 30. Dezember 2008, 9:36
>> Hi JS,
>>
>> IMHO, MSVC's /W3 is already quite
>> "aggressive", but I already had to go to /W4 to
>> find a few bugs. However, I agree that /W4 is "very
>> very very aggressive". I guess the "-Wall" on
>> gcc is roughly equivalent to "/W3", and
>> "-Wall -pedantic -all other that are not in Wall"
>> is roughly equivalent to "/W4"; Correct me if
>> I'm wrong. Does OSG_USE_AGGRESSIVE_WARNINGS generates
>> all gcc warnings (I mean -Wall plus all others)?
>>
>> So what? Well, as I don't know why
>> OSG_USE_AGGRESSIVE_WARNINGS has been turned on by default, I
>> can't be sure. But I suggest MSVC's
>> OSG_USE_AGGRESSIVE_WARNINGS to be turned off by default,
>> since it's only used for in-depth analysis and generates
>> lots of false positives.
>>
>> Sukender
>> PVLE - Lightweight cross-platform game engine -
>> http://pvle.sourceforge.net/
>>
>>
>> Le Mon, 29 Dec 2008 20:52:39 +0100, Jean-Sebastien Guay
>> <jean-sebastien.g...@cm-labs.com> a écrit:
>>
>> > Hi all, hi Robert, hope you are all having a good
>> holiday. :-)
>> >
>> > I was just building after an svn update today and
>> happened to delete
>> > my CMakeCache.txt, and the new default setting for
>> > OSG_USE_AGGRESSIVE_WARNINGS started taking effect. (it
>> seems that my
>> > old cache was overriding the value before)
>> >
>> > On Visual C++ 2005, with the
>> "non-aggressive" warning level (which
>> > gives /W3, which is a good warning level) all the libs
>> and plugins
>> > compile without warnings since about a week or two
>> ago, when Robert
>> > did a big warning blitz. Using the new default setting
>> (which turns on
>> > OSG_USE_AGGRESSIVE_WARNINGS and seems to equate to /W4
>> and perhaps
>> > some other settings which I didn't check), Visual
>> Studio now spits out
>> > a huge number of warnings in the default header files
>> like vector,
>> > xlocale, etc.
>> >
>> > Now I'm sure the first reaction will be to say
>> that MS should make
>> > their code warning-clean under even the highest level
>> of warnings. I'd
>> > tend to agree. But nevertheless, this warning level is
>> just not useful
>> > on Windows.
>> >
>> > I'd like to suggest that unless someone else has a
>> better idea,
>> > OSG_USE_AGGRESSIVE_WARNINGS do nothing on Windows /
>> Visual C++. /W3 is
>> > high enough to find most bad uses (unused /
>> uninitialized variables,
>> > bad type conversions, etc) and /W4 doesn't seem
>> useful at all since it
>> > spits out so much noise that we wouldn't be able
>> to find useful
>> > warnings in the lot even if we tried.
>> >
>> > Another possibility would be to make
>> OSG_USE_AGGRESSIVE_WARNINGS
>> > default to OFF on Windows / Visual C++, so that the
>> setting is still
>> > there if someone wants to use it but it doesn't
>> start spewing out all
>> > those warnings by default without conscious action
>> from the person
>> > running CMake.
>> >
>> > Comments?
>> >
>> > J-S
>> > --
>> > ______________________________________________________
>> > Jean-Sebastien Guay    jean-sebastien.g...@cm-labs.com
>> >                        http://whitestar02.webhop.org/
>> > _______________________________________________
>> > osg-users mailing list
>> > osg-users@lists.openscenegraph.org
>> >
>> http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org
>>
>> _______________________________________________
>> osg-users mailing list
>> osg-users@lists.openscenegraph.org
>> http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org
>
>
> 
> _______________________________________________
> osg-users mailing list
> osg-users@lists.openscenegraph.org
> http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org

_______________________________________________
osg-users mailing list
osg-users@lists.openscenegraph.org
http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org

Reply via email to