compile problems

[Peter Suetterlin]

  Hi,

I tried a clean checkout this morning
(cvs -d ":pserver:anonymous at cvs.sourceforge.net:/cvsroot/hugin" co hugin)
and I also tried Brunos nightly tarball (hugin-20030924.tar.gz)
Both show the same error 'no matching function for call to `map_get('
Where is this function supposed to come from?

Pablo d'Angelo wrote:
> On Tue, 23 Sep 2003, Peter Suetterlin wrote:
> 
> >   Hi!
> > 
> > Yesterday evening and this morning I tried to compile an recent cvs
> > version of hugin.  However, I fail to do so, it breaks with an
> > 
> >   g++ -I. -I../include -Wall -O2 -I../  -c -o .obj/Panorama.o Panorama.cpp
> >   Panorama.cpp: In member function `double PT::Panorama::calcHFOV() const':
> >   Panorama.cpp:245: error: no matching function for call to `map_get(const 
> >      std::map<std::string, PT::Variable, std::less<std::string>, 
> >      std::allocator<std::pair<const std::string, PT::Variable> > >&, const 
> >      char[2])'
> > 
> > I get a whole bunch of those, also for other files
> > (PanoToolsInterface.cpp, ImagesList.cpp).  Am I missing something?
> > This is a SuSE 8.2 Linux system (gcc 3.3, wxGTK-2.4.0 (self compiled),
> > glibc 2.3.2, PTLinux-2.6b1)
> 
> Hmm, I don't think so. 
> 
> the new version needs a pano12 directory (with the pano
> tools headers inside) in the include search path inside. Bruno's package
> contains such a directory.

Yes, I found out that one and installed the header files.

> But the error above is something different. I just checked out a fresh
> copy of hugin, using anonymous cvs, and it compiled fine.. I'll try
> gcc-3.3, I'm using gcc 3.2 right now..

Hm, I can't promise, but if I find the time I'll try to downgrade to
3.2 and retry the compile...

> 
> > Dr. Peter "Pit" Suetterlin                 http://www.astro.uu.nl/~suetter
> > Sterrenkundig Instituut Utrecht
> 
> I'm just curios: Are you planing to use hugin/panorama tools for astronomical
> applications?

Sorry, but actually no.  This is 'just for fun' for myself, although
one of my work areas is quite related.  I'm working on image
reconstruction using the so-called speckle masking method, and there I
have to tesselate images into ~1000 subimages, process them and then
stich them together again.  But the requirements for that are quite
different from what hugin/PanoTools do.

Cheers,

  Pit