Re: [PUSHED 4/4] Re: [PATCH 0/4] Remove some unused methods

On Tue, Feb 07, 2012 at 01:26:18AM +0100, Michael Stahl wrote:
> On 06/02/12 23:05, Greg Kroah-Hartman wrote:
> > Here are 4 patches removing some unused methods in the tree.
> >
> > All of these patches are contributed under the LGPLv3+ / MPL.
>
> great!
>
> > Note, I also removed the line in the unusedcode.easy file at the same
> > time, if I shouldn't have done that, let me know and I'll redo these
> > patches.
>
> doesn't matter too much as that will get updated anyway next time
> callcatcher tool is run.
> (also, unusedcode.easy only shows the situation on Linux, see below...)

Ah, didn't realize that.

> > If I messed anything up, or the patches need more information within the
> > body of the changelog, please let me know, and I'll be glad to respin
> > them.
> >
> > thanks,
> >
> > greg k-h
>
> these "tools" things are wicked, and have traps prepared for the unwary;
> it turns out that both DirEntry::GetDevice() and FileStat::FileStat(void
> const*) have 2 implementations, and the WNT one is actually used in
> WNT-only code:
>
> > tools/source/fsys/dirent.cxx:        aFromDevice=aFromDevice.GetDevice();
> > tools/source/fsys/dirent.cxx:        aToDevice=aToDevice.GetDevice();
>
> > tools/source/fsys/wntmsc.cxx:            pTemp->ImpSetStat( new FileStat( (void*) pDosDir ) );
> > tools/source/fsys/wntmsc.cxx:                FileStat *pNewStat = new FileStat( (void*) pDosDir 
> > );
>
> (and one of these is inside an "#ifdef FEAT_FSYS_DOUBLESPEED" ... you
> can't make these things up i guess...)

Ugh, sorry about that, I should have caught that in a grep, my fault.

> so i've only applied and pushed the part of these patches that removes
> the Unix implementation (that should hopefully deter people from adding
> new callers...); please try to be more careful next time (though
> admittedly checking that generally is a PITA given how name lookup works
> in C++)

Thanks for checking in the bits that didn't break the build.

> patches to remove the tools/fsys trainwreck completely and adapt callers
> to use osl/file.hxx would be appreciated of course :)

Hm, that looks a bit more fun, I'll work on that...

> PS: in case you wanted the free t-shirt as the 400th contributor i'm
> afraid you're probably too late :-P

I didn't know you were even offering such a thing, if I did, I would
have contributed sooner :)

thanks,

greg k-h