Win32 .pc files

[Foster Gareth <gareth foster siemens com>]

I hate to bring this up again, but I still have this issue with the win32
gtk/gtkmm installers. All my .pc files have the path /target/ in them,
rather than c:\gtk.

Shouldn't this be sorted when I install the devel packages by some sort of
script?

I am using the latest installers - any clues?

Cheers,

Gaz


> -----Original Message-----
> From: Roel Vanhout [mailto:roel riks nl] 
> Sent: 19 August 2005 14:16
> To: Ole Laursen
> Cc: gtkmm-list gnome org
> Subject: Re: Libpropc++
> 
> >  > Since template code cannot be 'separated' from the rest 
> of a program
> >  > into a shared library, all the files that depend on a 
> template library
> >  > (such as, in fact, libsigc++) will have to be 
> open-sourced in order to
> >  > comply with the terms of the LGPL. This does indeed 
> implicate that you'd
> >  > have to open-source at least a part of your program in 
> order to be able
> >  > to use libsigc++.
> > But this is your interpretation. Clearly, the authors of gtkmm and
> > libsigc++ didn't intend this interpretation. Especially for 
> libsigc++
> > it simply does not make any sense to license it under LGPL if the
> > template parts weren't covered by the same pattern of use - i.e. as
> > long as you are just using the library, there are no restrictions on
> > your license as long as people can get to the source of the library
> > itself (and do the relinking stuff).
> 
> Well this is where the fun begins. First, "the authors of gtkmm and 
> libsigc++ didn't intend this interpretation" is completely 
> irrelevant. 
> Not the 'intent' of the authors is what gives the license its 
> validity, 
>    a reasonable interpretation of the wording of it does. So 
> then we get 
> to the 'reasonable' part, and yes, in this case it could be 
> argued both 
> ways. I still maintain that the case of gtkmm/libsigc++, a strict 
> interpretation of the LGPL is in order, which would not allow 
> people to 
> use it in closed-source software. Notice that eg wxWidgets explicitly 
> allows static linking (they use the LGPL with this addition, at least 
> the last time I checked). I'd argue that if other projects 
> are aware of 
> this deficiency in the LGPL, then the projects that continue 
> to use it 
> unmodified want a strict interpretation, otherwise they'd use 
> a modified 
> form as well.
> LGPL is legally very muddy water. Even Richard Stallman and 
> Eben Moglen 
> don't agree on the details of it, and they wrote the thing! (I don't 
> have time to dig up references now, I think I mentioned them 
> last time 
> this issue came up on this list).
> 
> > Also, I think 6a) in the LGPL
> > 
> >   http://www.gnu.org/copyleft/lesser.html
> > 
> > covers this with the paranthesis:
> >  
> >   (It is understood that the user who changes the contents of
> >   definitions files in the Library will not necessarily be able to
> >   recompile the application to use the modified definitions.)
> 
> The question here is if template code is still a 'definition' 
> file. I'd 
> argue not, a template class/function necessarily has to be both 
> definition and implementation.
> 
> cheers,
> 
> roel
> 
> _______________________________________________
> gtkmm-list mailing list
> gtkmm-list gnome org
> http://mail.gnome.org/mailman/listinfo/gtkmm-list
>