Mac OS and hugin scriping/plugin interface: help needed

[kfj]

Hi all!

As you may have gleaned from the traffic on this list recently, work
has been done on the python scripting and plugin interface for hugin.
There is a branch for this project, called python_scripting, and the
python interface and the plugin interface have been made to run on
Kubuntu and Windows. So far, there hasn't been any echo from the Mac
side of things. What would be required is a trial build of the
python_scripting branch under Mac OS, to see if it compiles and runs
there as well. Trying this should require some, but not excessive
effort:

- The system has to have Python - I don't know what the Mac standard
is, but I'd even think it possible Macs already have it per default.

- The system should have SWIG, the 'Simplified Wrapper and Interface
Generator', version 2.0.1, which should be compiled from source.

after that it should be a matter of running an ordinary cmake cycle on
the python_scripting branch with BUILD_HSI=ON; the cmake code should
deal with the system-specific details. What's missing from the cmake
code is install targets, since neither T. Modes nor I knew where to
install the Python modules on Mac OS.

So it should be easy - theoretically - but if problems ocur, it'd be
great to know of them now rather than after putting more work into the
code on the platforms it runs on. If any of you have already managed
to compile the branch on Mac OS, please report as well!

If you want to just read up on the project and don't have access to
the code, the README is also available by itself from my bazaar repo,
and it should give you an idea about what's required to get going:

http://bazaar.launchpad.net/~kfj/+junk/script/view/head:/main/README.hsi

Kay

[Harry van der Wolf]

Hi Kay,

I have been a litle distracted the past weeks. I have a new Android tablet and to be able to tune it to my wishes I started creating rooted custom roms (as there wasn't any for my device) and already released a few (others like them as well).
Currently, I'm rather satisfied so other topics get renewed attention.

2011/2/7 kfj <_k...@yahoo.com>

Hi all!

As you may have gleaned from the traffic on this list recently, work
has been done on the python scripting and plugin interface for hugin.
There is a branch for this project, called python_scripting, and the
python interface and the plugin interface have been made to run on
Kubuntu and Windows. So far, there hasn't been any echo from the Mac
side of things. What would be required is a trial build of the
python_scripting branch under Mac OS, to see if it compiles and runs
there as well. Trying this should require some, but not excessive
effort:

- The system has to have Python - I don't know what the Mac standard
is, but I'd even think it possible Macs already have it per default.

Snow Leopard comes standard with python 2.3, 2.4, 2.5 and 2.6 where it defaults to 2.6. You can set the default python selection with python_select .

 

- The system should have SWIG, the 'Simplified Wrapper and Interface
Generator', version 2.0.1, which should be compiled from source.

Swig 2.0.1 is part of MacPorts and so is swig-python and can be easily compiled from macports. If compiled from macports the macports version of python will be installed which takes over from Apples python. No problem at all. Python for macports ican be selected from s 2.3-2.7 + 3.2.
As python 2.6 is mentioned as prerequisite it would require on OSX (extremely short howto):
- install macports from macports.org

in terminal:
sudo port install python26
sudo port install python-swig (this will automatically install swig 2.0.1 as dependency)
sudo port install hugin-app  (this will install hugin 2010.4 and all it's dependencies).
sudo port deactivate hugin-app (this is a sort of "make uninstall"  step. It will remove hugin from it's installation folders, but leave all dependencies intact).
sudo port deactivate libpano13 (this will also remove libpano from it's installation paths


- somehow get and compile panotools 2.9.18 on your system. Use the default /usr/local install
- compile hugin with python extension as described in Kay's link mentioned below, also to the default /usr/local

Later on, you can remove the panotools 2.9.18 and huginpython from /usr/local if you want your "clean" system back.

 

after that it should be a matter of running an ordinary cmake cycle on
the python_scripting branch with BUILD_HSI=ON; the cmake code should
deal with the system-specific details. What's missing from the cmake
code is install targets, since neither T. Modes nor I knew where to
install the Python modules on Mac OS.

As swig is part of macports, the easy install would be via macports. The install target will also be the macports repository which is /opt/local and which will be automatically taken care off.
The install target for the python things will be /opt/local as well.
I have some fears with regard to the bundle integration.
Otherwise installing to /user/local is no problem either as the several environment settings will take care of that.

 

So it should be easy - theoretically - but if problems ocur, it'd be
great to know of them now rather than after putting more work into the
code on the platforms it runs on. If any of you have already managed
to compile the branch on Mac OS, please report as well!

If you want to just read up on the project and don't have access to
the code, the README is also available by itself from my bazaar repo,
and it should give you an idea about what's required to get going:

http://bazaar.launchpad.net/~kfj/+junk/script/view/head:/main/README.hsi

Kay

This mail is more or less meant as an "info" mail to you and others.
I will try to compile it and report back as soon as possible. The current trunk is based on panotools 2.9.18 which is at this moment in a normal cmake environment a complexing factor (see above) as 2.9.17 is the default now on macports (I took care of that myself as it is neccessary for Hugin 2010.4).
I have to twist my system a little.

I'll be back.

Harry

[kfj]

On 7 Feb., 18:45, Harry van der Wolf <hvdw...@gmail.com> wrote:

> I have been a litle distracted the past weeks. I have a new Android tablet
> and to be able to tune it to my wishes I started creating rooted custom roms

> ...

sounds exciting. I found myself ogling tablets recently...

> Python for macports ican be selected from s 2.3-2.7 + 3.2.
> As python 2.6 is mentioned as prerequisite it would require on OSX

> ...

Thank you, Harry. This makes me hope that it is all quite feasible.
I'm glad Python 2.6 which I have used on my system AND Swig 2.0.1 are
readily available for Macs - SWIG 2.0.1 wasn't in the standard packets
for Ubuntu and I had to compile it myself - which was easy, though. I
just don't have a Mac to try out your suggestions myself, but with all
the good advice maybe we'll have a successful build soon!
T. Modes has made the effort to also support Python 3.X and has got it
to run with 3.X on Windows, so far I haven't done so on Ubuntu, but it
should be quite possible as well. He's using 2.7 on windows, and I
think any of the 2.Xs should work.

> This mail is more or less meant as an "info" mail to you and others.
> I will try to compile it and report back as soon as possible. The current
> trunk is based on panotools 2.9.18 which is at this moment in a normal cmake
> environment a complexing factor (see above) as 2.9.17 is the default now on
> macports (I took care of that myself as it is neccessary for Hugin 2010.4).
> I have to twist my system a little.

If you have a look at the code you will see that the python interface
stands pretty much by itself codewise, so if worse comes to worse and
it should turn out that there's a problem using 2.9.18 I'm quite
confident that it could be backported into an older version and
compile with 2.9.17 without too much ado - the dependencies are with
hugin headers, there's very little direct contact with panotools - but
I'm hopeful you'll twist it just the right way - and I hope my code
stands up to some twisting, too. The main thing right now is to see it
compiles and does what it's supposed to do on Mac OS with no
fundamental problems, so we have all platforms covered and can start
doing something with it.

Kay

[Harry van der Wolf]

I deactivated my libpano 2.9.17 and compiled 2.9.18.

Then I compiled hugin using the "-DBUILD_HSI=ON" parameter.
The compilation stopped with a lot of errors. See below.

Harry


Scanning dependencies of target _hsi
[ 53%] Building CXX object src/hugin_script_interface/CMakeFiles/_hsi.dir/hsiPYTHON_wrap.cxx.o
In file included from /Users/Shared/development/hugin_related/huginpython/src/hugin_base/nona/Stitcher.h:44,
                 from /Users/Shared/development/hugin_related/huginpython/src/hugin_base/algorithms/nona/NonaImageStitcher.h:36,
                 from /Users/Shared/development/hugin_related/huginpython/build/src/hugin_script_interface/hsiPYTHON_wrap.cxx:3040:
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/impexalpha.hxx:295: error: expected unqualified-id before numeric constant
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/impexalpha.hxx:299: error: wrong number of template arguments (1, should be 2)
/Users/Shared/development/hugin_related/huginpython/src/foreign/vigra/vigra/accessor.hxx:953: error: provided for ‘template<class T, int SIZE> class vigra::TinyVector’
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/impexalpha.hxx:300: error: ‘value_type’ is not a class or namespace
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/impexalpha.hxx: In member function ‘int vigra::MultiImageVectorMaskAccessor4<Iter1, Acc1, Iter2, Acc2>::operator()(const DIFFERENCE&) const’:
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/impexalpha.hxx:315: error: expected unqualified-id before numeric constant
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/impexalpha.hxx:316: error: ‘v1’ was not declared in this scope
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/impexalpha.hxx: In member function ‘int vigra::MultiImageVectorMaskAccessor4<Iter1, Acc1, Iter2, Acc2>::operator()(DIFFERENCE1, const DIFFERENCE2&) const’:
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/impexalpha.hxx:328: error: expected unqualified-id before numeric constant
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/impexalpha.hxx:329: error: ‘v1’ was not declared in this scope
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/impexalpha.hxx: In member function ‘int vigra::MultiImageVectorMaskAccessor4<Iter1, Acc1, Iter2, Acc2>::set(const int&, const DIFFERENCE&) const’:
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/impexalpha.hxx:342: error: invalid types ‘const int[int]’ for array subscript
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/impexalpha.hxx:343: error: invalid types ‘const int[int]’ for array subscript
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/impexalpha.hxx:344: error: invalid types ‘const int[int]’ for array subscript
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/impexalpha.hxx:345: error: invalid types ‘const int[int]’ for array subscript
In file included from /Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/ROIImage.h:30,
                 from /Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/blend.h:33,
                 from /Users/Shared/development/hugin_related/huginpython/src/hugin_base/nona/Stitcher.h:46,
                 from /Users/Shared/development/hugin_related/huginpython/src/hugin_base/algorithms/nona/NonaImageStitcher.h:36,
                 from /Users/Shared/development/hugin_related/huginpython/build/src/hugin_script_interface/hsiPYTHON_wrap.cxx:3040:
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/lut.h: At global scope:
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/lut.h:120: error: expected unqualified-id before numeric constant
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/lut.h:149: error: expected identifier before numeric constant
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/lut.h:149: error: expected ‘,’ or ‘...’ before numeric constant
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/lut.h:188: error: type/value mismatch at argument 1 in template parameter list for ‘template<class T, unsigned int RED_IDX, unsigned int GREEN_IDX, unsigned int BLUE_IDX> class vigra::RGBValue’
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/lut.h:188: error:   expected a type, got ‘65536’
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/lut.h:198: error: type/value mismatch at argument 1 in template parameter list for ‘template<class T, unsigned int RED_IDX, unsigned int GREEN_IDX, unsigned int BLUE_IDX> class vigra::RGBValue’
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/lut.h:198: error:   expected a type, got ‘65536’
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/lut.h:209: error: expected identifier before numeric constant
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/lut.h:209: error: expected ‘,’ or ‘...’ before numeric constant
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/lut.h:215: error: expected identifier before numeric constant
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/lut.h:215: error: expected ‘,’ or ‘...’ before numeric constant
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/lut.h:220: error: expected identifier before numeric constant
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/lut.h:220: error: expected ‘,’ or ‘...’ before numeric constant
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/lut.h:226: error: type/value mismatch at argument 1 in template parameter list for ‘template<class T, unsigned int RED_IDX, unsigned int GREEN_IDX, unsigned int BLUE_IDX> class vigra::RGBValue’
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/lut.h:226: error:   expected a type, got ‘65536’
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/lut.h: In member function ‘typename LUT::value_type vigra_ext::LUTFunctor<VTIn, LUT>::applyLutInteger() const’:
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/lut.h:153: error: type/value mismatch at argument 1 in template parameter list for ‘template<class T1> struct vigra_ext::LUTTraits’
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/lut.h:153: error:   expected a type, got ‘65536’
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/lut.h:154: error: ‘v’ was not declared in this scope
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/lut.h:157: error: type/value mismatch at argument 1 in template parameter list for ‘template<class T1> struct vigra_ext::LUTTraits’
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/lut.h:157: error:   expected a type, got ‘65536’
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/lut.h:158: error: ‘v’ was not declared in this scope
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/lut.h: In member function ‘vigra::RGBValue<typename LUT::value_type, 0u, 1u, 2u> vigra_ext::LUTFunctor<VTIn, LUT>::applyVector(int, vigra::VigraFalseType) const’:
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/lut.h:190: error: type/value mismatch at argument 1 in template parameter list for ‘template<class T, unsigned int RED_IDX, unsigned int GREEN_IDX, unsigned int BLUE_IDX> class vigra::RGBValue’
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/lut.h:190: error:   expected a type, got ‘65536’
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/lut.h:190: error: invalid type in declaration before ‘;’ token
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/lut.h:191: error: request for member ‘size’ in ‘v’, which is of non-class type ‘int’
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/lut.h:192: error: invalid types ‘int[size_t]’ for array subscript
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/lut.h:192: error: invalid types ‘int[size_t]’ for array subscript
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/lut.h: In member function ‘vigra::RGBValue<typename LUT::value_type, 0u, 1u, 2u> vigra_ext::LUTFunctor<VTIn, LUT>::applyVector(int, vigra::VigraTrueType) const’:
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/lut.h:202: error: request for member ‘size’ in ‘v’, which is of non-class type ‘int’
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/lut.h:203: error: invalid types ‘int[size_t]’ for array subscript
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/lut.h: In member function ‘typename LUT::value_type vigra_ext::LUTFunctor<VTIn, LUT>::applyScalar() const’:
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/lut.h:211: error: ‘v’ was not declared in this scope
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/lut.h: In member function ‘typename LUT::value_type vigra_ext::LUTFunctor<VTIn, LUT>::applyScalar() const’:
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/lut.h:217: error: ‘v’ was not declared in this scope
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/lut.h: In member function ‘typename LUT::value_type vigra_ext::LUTFunctor<VTIn, LUT>::apply() const’:
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/lut.h:222: error: type/value mismatch at argument 1 in template parameter list for ‘template<class A> struct vigra::NumericTraits’
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/lut.h:222: error:   expected a type, got ‘65536’
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/lut.h:222: error: invalid type in declaration before ‘;’ token
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/lut.h:223: error: ‘v’ was not declared in this scope
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/lut.h: In member function ‘vigra::RGBValue<typename LUT::value_type, 0u, 1u, 2u> vigra_ext::LUTFunctor<VTIn, LUT>::apply(int, vigra::VigraFalseType) const’:
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/lut.h:228: error: type/value mismatch at argument 1 in template parameter list for ‘template<class A> struct vigra::NumericTraits’
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/lut.h:228: error:   expected a type, got ‘65536’
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/lut.h:228: error: invalid type in declaration before ‘;’ token
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/lut.h: At global scope:
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/lut.h:251: error: expected identifier before numeric constant
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/lut.h:251: error: expected ‘>’ before numeric constant
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/lut.h:253: error: default argument for template parameter for class enclosing ‘struct vigra_ext::InvLUTFunctor<<declaration error> >’
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/lut.h:258: error: ‘LUT’ has not been declared
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/lut.h:269: error: expected `)' before ‘&’ token
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/lut.h:370: error: ‘LUT’ does not name a type
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/lut.h: In member function ‘int vigra_ext::InvLUTFunctor<<declaration error> >::applyLutFloat(int) const’:
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/lut.h:277: error: ‘m_lut’ was not declared in this scope
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/lut.h:282: error: ‘LUT’ has not been declared
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/lut.h:282: error: expected initializer before ‘p’
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/lut.h:285: error: ‘p’ was not declared in this scope
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/lut.h: In member function ‘int vigra_ext::InvLUTFunctor<<declaration error> >::applyLutInteger(T) const’:
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/lut.h:310: error: invalid use of incomplete type ‘const struct vigra_ext::InvLUTFunctor<<declaration error> >’
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/lut.h:253: error: declaration of ‘const struct vigra_ext::InvLUTFunctor<<declaration error> >’
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/lut.h: In member function ‘vigra::RGBValue<int, 0u, 1u, 2u> vigra_ext::InvLUTFunctor<<declaration error> >::applyVector(vigra::RGBValue<T, 0u, 1u, 2u>, vigra::VigraFalseType) const’:
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/lut.h:318: error: type/value mismatch at argument 1 in template parameter list for ‘template<class T, unsigned int RED_IDX, unsigned int GREEN_IDX, unsigned int BLUE_IDX> class vigra::RGBValue’
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/lut.h:318: error:   expected a type, got ‘65536’
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/lut.h:318: error: invalid type in declaration before ‘;’ token
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/lut.h:320: error: invalid types ‘int[size_t]’ for array subscript
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/lut.h:320: error: invalid use of incomplete type ‘const struct vigra_ext::InvLUTFunctor<<declaration error> >’
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/lut.h:253: error: declaration of ‘const struct vigra_ext::InvLUTFunctor<<declaration error> >’
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/lut.h: In member function ‘vigra::RGBValue<int, 0u, 1u, 2u> vigra_ext::InvLUTFunctor<<declaration error> >::applyVector(vigra::RGBValue<T, 0u, 1u, 2u>, vigra::VigraTrueType) const’:
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/lut.h:330: error: invalid use of incomplete type ‘const struct vigra_ext::InvLUTFunctor<<declaration error> >’
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/lut.h:253: error: declaration of ‘const struct vigra_ext::InvLUTFunctor<<declaration error> >’
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/lut.h: In member function ‘int vigra_ext::InvLUTFunctor<<declaration error> >::applyScalar(T, vigra::VigraFalseType) const’:
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/lut.h:339: error: invalid use of incomplete type ‘const struct vigra_ext::InvLUTFunctor<<declaration error> >’
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/lut.h:253: error: declaration of ‘const struct vigra_ext::InvLUTFunctor<<declaration error> >’
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/lut.h: In member function ‘int vigra_ext::InvLUTFunctor<<declaration error> >::applyScalar(T, vigra::VigraTrueType) const’:
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/lut.h:346: error: invalid use of incomplete type ‘const struct vigra_ext::InvLUTFunctor<<declaration error> >’
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/lut.h:253: error: declaration of ‘const struct vigra_ext::InvLUTFunctor<<declaration error> >’
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/lut.h: In member function ‘int vigra_ext::InvLUTFunctor<<declaration error> >::apply(T, vigra::VigraTrueType) const’:
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/lut.h:353: error: invalid use of incomplete type ‘const struct vigra_ext::InvLUTFunctor<<declaration error> >’
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/lut.h:253: error: declaration of ‘const struct vigra_ext::InvLUTFunctor<<declaration error> >’
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/lut.h: In member function ‘vigra::RGBValue<int, 0u, 1u, 2u> vigra_ext::InvLUTFunctor<<declaration error> >::apply(vigra::RGBValue<T, 0u, 1u, 2u>, vigra::VigraFalseType) const’:
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/lut.h:360: error: invalid use of incomplete type ‘const struct vigra_ext::InvLUTFunctor<<declaration error> >’
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/lut.h:253: error: declaration of ‘const struct vigra_ext::InvLUTFunctor<<declaration error> >’
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/lut.h: In member function ‘typename vigra::NumericTraits<T>::RealPromote vigra_ext::InvLUTFunctor<<declaration error> >::operator()(T) const’:
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/lut.h:367: error: invalid use of incomplete type ‘const struct vigra_ext::InvLUTFunctor<<declaration error> >’
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/lut.h:253: error: declaration of ‘const struct vigra_ext::InvLUTFunctor<<declaration error> >’
In file included from /Users/Shared/development/hugin_related/huginpython/src/hugin_base/nona/RemappedPanoImage.h:197,
                 from /Users/Shared/development/hugin_related/huginpython/src/hugin_base/nona/Stitcher.h:53,
                 from /Users/Shared/development/hugin_related/huginpython/src/hugin_base/algorithms/nona/NonaImageStitcher.h:36,
                 from /Users/Shared/development/hugin_related/huginpython/build/src/hugin_script_interface/hsiPYTHON_wrap.cxx:3040:
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/photometric/ResponseTransform.h: At global scope:
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/photometric/ResponseTransform.h:55: error: expected unqualified-id before numeric constant
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/photometric/ResponseTransform.h:90: error: type/value mismatch at argument 1 in template parameter list for ‘template<class A> struct vigra::NumericTraits’
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/photometric/ResponseTransform.h:90: error:   expected a type, got ‘65536’
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/photometric/ResponseTransform.h:91: error: expected identifier before numeric constant
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/photometric/ResponseTransform.h:91: error: expected ‘,’ or ‘...’ before numeric constant
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/photometric/ResponseTransform.h:94: error: type/value mismatch at argument 1 in template parameter list for ‘template<class A> struct vigra::NumericTraits’
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/photometric/ResponseTransform.h:94: error:   expected a type, got ‘65536’
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/photometric/ResponseTransform.h:95: error: expected identifier before numeric constant
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/photometric/ResponseTransform.h:95: error: expected ‘,’ or ‘...’ before numeric constant
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/photometric/ResponseTransform.h:98: error: type/value mismatch at argument 1 in template parameter list for ‘template<class T, unsigned int RED_IDX, unsigned int GREEN_IDX, unsigned int BLUE_IDX> class vigra::RGBValue’
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/photometric/ResponseTransform.h:98: error:   expected a type, got ‘65536’
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/photometric/ResponseTransform.h:98: error: template argument 1 is invalid
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/photometric/ResponseTransform.h:99: error: type/value mismatch at argument 1 in template parameter list for ‘template<class T, unsigned int RED_IDX, unsigned int GREEN_IDX, unsigned int BLUE_IDX> class vigra::RGBValue’
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/photometric/ResponseTransform.h:99: error:   expected a type, got ‘65536’
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/photometric/ResponseTransform.h:102: error: type/value mismatch at argument 1 in template parameter list for ‘template<class T, unsigned int RED_IDX, unsigned int GREEN_IDX, unsigned int BLUE_IDX> class vigra::RGBValue’
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/photometric/ResponseTransform.h:102: error:   expected a type, got ‘65536’
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/photometric/ResponseTransform.h:102: error: template argument 1 is invalid
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/photometric/ResponseTransform.h:103: error: type/value mismatch at argument 1 in template parameter list for ‘template<class T, unsigned int RED_IDX, unsigned int GREEN_IDX, unsigned int BLUE_IDX> class vigra::RGBValue’
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/photometric/ResponseTransform.h:103: error:   expected a type, got ‘65536’
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/photometric/ResponseTransform.h:117: error: type/value mismatch at argument 1 in template parameter list for ‘template<class VTIn, class LUT> struct vigra_ext::LUTFunctor’
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/photometric/ResponseTransform.h:117: error:   expected a type, got ‘65536’
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/photometric/ResponseTransform.h:142: error: expected unqualified-id before numeric constant
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/photometric/ResponseTransform.h:143: error: type/value mismatch at argument 1 in template parameter list for ‘template<class A> struct vigra::NumericTraits’
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/photometric/ResponseTransform.h:143: error:   expected a type, got ‘65536’
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/photometric/ResponseTransform.h:191: error: expected identifier before numeric constant
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/photometric/ResponseTransform.h:191: error: expected ‘,’ or ‘...’ before numeric constant
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/photometric/ResponseTransform.h:195: error: expected identifier before numeric constant
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/photometric/ResponseTransform.h:195: error: expected ‘,’ or ‘...’ before numeric constant
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/photometric/ResponseTransform.h:198: error: type/value mismatch at argument 1 in template parameter list for ‘template<class T, unsigned int RED_IDX, unsigned int GREEN_IDX, unsigned int BLUE_IDX> class vigra::RGBValue’
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/photometric/ResponseTransform.h:198: error:   expected a type, got ‘65536’
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/photometric/ResponseTransform.h:198: error: template argument 1 is invalid
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/photometric/ResponseTransform.h:199: error: type/value mismatch at argument 1 in template parameter list for ‘template<class T, unsigned int RED_IDX, unsigned int GREEN_IDX, unsigned int BLUE_IDX> class vigra::RGBValue’
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/photometric/ResponseTransform.h:199: error:   expected a type, got ‘65536’
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/photometric/ResponseTransform.h:202: error: type/value mismatch at argument 1 in template parameter list for ‘template<class T, unsigned int RED_IDX, unsigned int GREEN_IDX, unsigned int BLUE_IDX> class vigra::RGBValue’
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/photometric/ResponseTransform.h:202: error:   expected a type, got ‘65536’
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/photometric/ResponseTransform.h:202: error: template argument 1 is invalid
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/photometric/ResponseTransform.h:203: error: type/value mismatch at argument 1 in template parameter list for ‘template<class T, unsigned int RED_IDX, unsigned int GREEN_IDX, unsigned int BLUE_IDX> class vigra::RGBValue’
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/photometric/ResponseTransform.h:203: error:   expected a type, got ‘65536’
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/photometric/ResponseTransform.h:229: error: wrong number of template arguments (2, should be 1)
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/lut.h:252: error: provided for ‘template<<declaration error> > struct vigra_ext::InvLUTFunctor’
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/photometric/ResponseTransform.h: In member function ‘void HuginBase::Photometric::InvResponseTransform<VTIn, VTOut>::enforceMonotonicity()’:
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/photometric/ResponseTransform.h:177: error: wrong number of template arguments (2, should be 1)
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/lut.h:252: error: provided for ‘template<<declaration error> > struct vigra_ext::InvLUTFunctor’
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/photometric/ResponseTransform.h: In member function ‘void HuginBase::Photometric::ResponseTransform<VTIn>::initWithSrcImg(const HuginBase::SrcPanoImage&)’:
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/photometric/ResponseTransform.h:284: error: type/value mismatch at argument 1 in template parameter list for ‘template<class T1> struct vigra_ext::LUTTraits’
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/photometric/ResponseTransform.h:284: error:   expected a type, got ‘65536’
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/photometric/ResponseTransform.h:317: error: type/value mismatch at argument 1 in template parameter list for ‘template<class VTIn, class LUT> struct vigra_ext::LUTFunctor’
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/photometric/ResponseTransform.h:317: error:   expected a type, got ‘65536’
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/photometric/ResponseTransform.h: At global scope:
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/photometric/ResponseTransform.h:353: error: template argument 1 is invalid
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/photometric/ResponseTransform.h:354: error: expected initializer before ‘ResponseTransform’
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/photometric/ResponseTransform.h:368: error: template argument 1 is invalid
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/photometric/ResponseTransform.h:368: error: template argument 1 is invalid
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/photometric/ResponseTransform.h:369: error: template argument 1 is invalid
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/photometric/ResponseTransform.h: In member function ‘int HuginBase::Photometric::ResponseTransform<VTIn>::apply(int, const hugin_utils::FDiff2D&, vigra::VigraFalseType) const’:
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/photometric/ResponseTransform.h:371: error: type/value mismatch at argument 1 in template parameter list for ‘template<class T, unsigned int RED_IDX, unsigned int GREEN_IDX, unsigned int BLUE_IDX> class vigra::RGBValue’
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/photometric/ResponseTransform.h:371: error:   expected a type, got ‘65536’
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/photometric/ResponseTransform.h:371: error: template argument 1 is invalid
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/photometric/ResponseTransform.h:371: error: invalid type in declaration before ‘=’ token
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/photometric/ResponseTransform.h:376: error: request for member ‘red’ in ‘ret’, which is of non-class type ‘int’
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/photometric/ResponseTransform.h:376: error: request for member ‘red’ in ‘ret’, which is of non-class type ‘int’
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/photometric/ResponseTransform.h:377: error: request for member ‘blue’ in ‘ret’, which is of non-class type ‘int’
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/photometric/ResponseTransform.h:377: error: request for member ‘blue’ in ‘ret’, which is of non-class type ‘int’
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/photometric/ResponseTransform.h: At global scope:
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/photometric/ResponseTransform.h:387: error: template argument 1 is invalid
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/photometric/ResponseTransform.h:388: error: expected initializer before ‘ResponseTransform’
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/photometric/ResponseTransform.h:395: error: template argument 1 is invalid
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/photometric/ResponseTransform.h:395: error: template argument 1 is invalid
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/photometric/ResponseTransform.h:396: error: template argument 1 is invalid
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/photometric/ResponseTransform.h: In member function ‘int HuginBase::Photometric::ResponseTransform<VTIn>::apply(int, const hugin_utils::FDiff2D&) const’:
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/photometric/ResponseTransform.h:398: error: type/value mismatch at argument 1 in template parameter list for ‘template<class T, unsigned int RED_IDX, unsigned int GREEN_IDX, unsigned int BLUE_IDX> class vigra::RGBValue’
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/photometric/ResponseTransform.h:398: error:   expected a type, got ‘65536’
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/photometric/ResponseTransform.h:398: error: template argument 1 is invalid
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/photometric/ResponseTransform.h:398: error: invalid type in declaration before ‘;’ token
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/photometric/ResponseTransform.h: In constructor ‘HuginBase::Photometric::InvResponseTransform<VTIn, VTOut>::InvResponseTransform(const HuginBase::SrcPanoImage&)’:
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/photometric/ResponseTransform.h:418: error: wrong number of template arguments (2, should be 1)
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/lut.h:252: error: provided for ‘template<<declaration error> > struct vigra_ext::InvLUTFunctor’
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/photometric/ResponseTransform.h: In member function ‘void HuginBase::Photometric::InvResponseTransform<VTIn, VTOut>::init(const HuginBase::SrcPanoImage&)’:
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/photometric/ResponseTransform.h:430: error: wrong number of template arguments (2, should be 1)
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/lut.h:252: error: provided for ‘template<<declaration error> > struct vigra_ext::InvLUTFunctor’
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/photometric/ResponseTransform.h: At global scope:
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/photometric/ResponseTransform.h:478: error: ‘typename vigra::NumericTraits<typename HuginBase::Photometric::InvResponseTransform<VTIn, VTOut>::dest_type>::RealPromote HuginBase::Photometric::InvResponseTransform<VTIn, VTOut>::apply’ is not a static member of ‘class HuginBase::Photometric::InvResponseTransform<VTIn, VTOut>’
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/photometric/ResponseTransform.h:478: error: template definition of non-template ‘typename vigra::NumericTraits<typename HuginBase::Photometric::InvResponseTransform<VTIn, VTOut>::dest_type>::RealPromote HuginBase::Photometric::InvResponseTransform<VTIn, VTOut>::apply’
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/photometric/ResponseTransform.h:478: error: expected `)' before ‘v’
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/photometric/ResponseTransform.h:478: error: expected primary-expression before ‘const’
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/photometric/ResponseTransform.h:478: error: expected primary-expression before ‘)’ token
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/photometric/ResponseTransform.h:478: error: expected `;' before ‘const’
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/photometric/ResponseTransform.h:502: error: template argument 1 is invalid
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/photometric/ResponseTransform.h:502: error: template argument 1 is invalid
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/photometric/ResponseTransform.h:503: error: type/value mismatch at argument 1 in template parameter list for ‘template<class T, unsigned int RED_IDX, unsigned int GREEN_IDX, unsigned int BLUE_IDX> class vigra::RGBValue’
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/photometric/ResponseTransform.h:503: error:   expected a type, got ‘65536’
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/photometric/ResponseTransform.h: In member function ‘int HuginBase::Photometric::InvResponseTransform<VTIn, VTOut>::apply(int, const hugin_utils::FDiff2D&, vigra::VigraFalseType) const’:
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/photometric/ResponseTransform.h:505: error: type/value mismatch at argument 1 in template parameter list for ‘template<class T, unsigned int RED_IDX, unsigned int GREEN_IDX, unsigned int BLUE_IDX> class vigra::RGBValue’
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/photometric/ResponseTransform.h:505: error:   expected a type, got ‘65536’
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/photometric/ResponseTransform.h:505: error: template argument 1 is invalid
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/photometric/ResponseTransform.h:505: error: invalid type in declaration before ‘(’ token
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/photometric/ResponseTransform.h:509: error: type/value mismatch at argument 1 in template parameter list for ‘template<class T1> struct vigra_ext::LUTTraits’
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/photometric/ResponseTransform.h:509: error:   expected a type, got ‘65536’
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/photometric/ResponseTransform.h:514: error: request for member ‘red’ in ‘ret’, which is of non-class type ‘int’
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/photometric/ResponseTransform.h:515: error: request for member ‘blue’ in ‘ret’, which is of non-class type ‘int’
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/photometric/ResponseTransform.h:523: error: invalid types ‘int[size_t]’ for array subscript
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/photometric/ResponseTransform.h:523: error: invalid types ‘int[size_t]’ for array subscript
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/photometric/ResponseTransform.h: At global scope:
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/photometric/ResponseTransform.h:532: error: ‘typename vigra::NumericTraits<typename HuginBase::Photometric::InvResponseTransform<VTIn, VTOut>::dest_type>::RealPromote HuginBase::Photometric::InvResponseTransform<VTIn, VTOut>::apply’ is not a static member of ‘class HuginBase::Photometric::InvResponseTransform<VTIn, VTOut>’
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/photometric/ResponseTransform.h:532: error: template definition of non-template ‘typename vigra::NumericTraits<typename HuginBase::Photometric::InvResponseTransform<VTIn, VTOut>::dest_type>::RealPromote HuginBase::Photometric::InvResponseTransform<VTIn, VTOut>::apply’
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/photometric/ResponseTransform.h:532: error: expected `)' before ‘v’
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/photometric/ResponseTransform.h:532: error: expected primary-expression before ‘const’
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/photometric/ResponseTransform.h:532: error: expected `;' before ‘const’
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/photometric/ResponseTransform.h:539: error: template argument 1 is invalid
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/photometric/ResponseTransform.h:539: error: template argument 1 is invalid
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/photometric/ResponseTransform.h:540: error: type/value mismatch at argument 1 in template parameter list for ‘template<class T, unsigned int RED_IDX, unsigned int GREEN_IDX, unsigned int BLUE_IDX> class vigra::RGBValue’
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/photometric/ResponseTransform.h:540: error:   expected a type, got ‘65536’
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/photometric/ResponseTransform.h: In member function ‘int HuginBase::Photometric::InvResponseTransform<VTIn, VTOut>::apply(int, const hugin_utils::FDiff2D&) const’:
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/photometric/ResponseTransform.h:542: error: type/value mismatch at argument 1 in template parameter list for ‘template<class T, unsigned int RED_IDX, unsigned int GREEN_IDX, unsigned int BLUE_IDX> class vigra::RGBValue’
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/photometric/ResponseTransform.h:542: error:   expected a type, got ‘65536’
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/photometric/ResponseTransform.h:542: error: template argument 1 is invalid
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/photometric/ResponseTransform.h:542: error: invalid type in declaration before ‘;’ token
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/photometric/ResponseTransform.h: In member function ‘void HuginBase::Photometric::InvResponseTransform<VTIn, VTOut>::emitGLSL(std::ostringstream&, std::vector<double, std::allocator<double> >&, std::vector<double, std::allocator<double> >&) const’:
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/photometric/ResponseTransform.h:568: error: type/value mismatch at argument 1 in template parameter list for ‘template<class T1> struct vigra_ext::LUTTraits’
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/photometric/ResponseTransform.h:568: error:   expected a type, got ‘65536’
In file included from /Users/Shared/development/hugin_related/huginpython/src/hugin_base/nona/ImageRemapper.h:152,
                 from /Users/Shared/development/hugin_related/huginpython/src/hugin_base/nona/Stitcher.h:54,
                 from /Users/Shared/development/hugin_related/huginpython/src/hugin_base/algorithms/nona/NonaImageStitcher.h:36,
                 from /Users/Shared/development/hugin_related/huginpython/build/src/hugin_script_interface/hsiPYTHON_wrap.cxx:3040:
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/VignettingCorrection.h: At global scope:
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/VignettingCorrection.h:45: error: expected identifier before numeric constant
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/VignettingCorrection.h:45: error: expected ‘>’ before numeric constant
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/VignettingCorrection.h:47: error: default argument for template parameter for class enclosing ‘class vigra_ext::VigCorrFlatDivFunctor<<declaration error> >’
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/VignettingCorrection.h:51: error: expected unqualified-id before numeric constant
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/VignettingCorrection.h:55: error: ‘VT2’ does not name a type
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/VignettingCorrection.h:59: error: type/value mismatch at argument 1 in template parameter list for ‘template<class A> struct vigra::NumericTraits’
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/VignettingCorrection.h:59: error:   expected a type, got ‘65536’
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/VignettingCorrection.h:62: error: ‘VT2’ was not declared in this scope
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/VignettingCorrection.h:62: error: template argument 1 is invalid
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/VignettingCorrection.h:64: error: expected ‘,’ or ‘...’ before ‘&’ token
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/VignettingCorrection.h:64: error: ISO C++ forbids declaration of ‘InvResp’ with no type
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/VignettingCorrection.h:68: error: ‘InvResp’ does not name a type
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/VignettingCorrection.h:69: error: ‘Adjust’ does not name a type
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/VignettingCorrection.h:74: error: ‘first_argument_type’ has not been declared
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/VignettingCorrection.h:74: error: ‘second_argument_type’ has not been declared
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/VignettingCorrection.h: In constructor ‘vigra_ext::VigCorrFlatDivFunctor<<declaration error> >::VigCorrFlatDivFunctor(int, int)’:
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/VignettingCorrection.h:65: error: class ‘vigra_ext::VigCorrFlatDivFunctor<<declaration error> >’ does not have any field named ‘m_InvResp’
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/VignettingCorrection.h:65: error: ‘fr’ was not declared in this scope
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/VignettingCorrection.h:65: error: class ‘vigra_ext::VigCorrFlatDivFunctor<<declaration error> >’ does not have any field named ‘m_Adjust’
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/VignettingCorrection.h:65: error: ‘adj’ was not declared in this scope
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/VignettingCorrection.h: In member function ‘int vigra_ext::VigCorrFlatDivFunctor<<declaration error> >::operator()(const int&, const int&) const’:
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/VignettingCorrection.h:78: error: there are no arguments to ‘m_InvResp’ that depend on a template parameter, so a declaration of ‘m_InvResp’ must be available
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/VignettingCorrection.h:78: error: (if you use ‘-fpermissive’, G++ will accept your code, but allowing the use of an undeclared name is deprecated)
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/VignettingCorrection.h:81: error: there are no arguments to ‘m_Adjust’ that depend on a template parameter, so a declaration of ‘m_Adjust’ must be available
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/VignettingCorrection.h: At global scope:
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/VignettingCorrection.h:85: error: expected identifier before numeric constant
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/VignettingCorrection.h:85: error: expected ‘>’ before numeric constant
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/VignettingCorrection.h:87: error: default argument for template parameter for class enclosing ‘class vigra_ext::VigCorrDivFunctor<<declaration error> >’
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/VignettingCorrection.h:91: error: expected unqualified-id before numeric constant
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/VignettingCorrection.h:95: error: type/value mismatch at argument 1 in template parameter list for ‘template<class A> struct vigra::NumericTraits’
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/VignettingCorrection.h:95: error:   expected a type, got ‘65536’
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/VignettingCorrection.h:98: error: expected ‘,’ or ‘...’ before ‘&’ token
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/VignettingCorrection.h:98: error: ISO C++ forbids declaration of ‘InvResp’ with no type
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/VignettingCorrection.h:102: error: ‘InvResp’ does not name a type
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/VignettingCorrection.h:103: error: ‘VigFunc’ does not name a type
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/VignettingCorrection.h:104: error: ‘Adjust’ does not name a type
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/VignettingCorrection.h:108: error: ‘first_argument_type’ has not been declared
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/VignettingCorrection.h: In constructor ‘vigra_ext::VigCorrDivFunctor<<declaration error> >::VigCorrDivFunctor(int)’:
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/VignettingCorrection.h:99: error: class ‘vigra_ext::VigCorrDivFunctor<<declaration error> >’ does not have any field named ‘m_InvResp’
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/VignettingCorrection.h:99: error: ‘fr’ was not declared in this scope
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/VignettingCorrection.h:99: error: class ‘vigra_ext::VigCorrDivFunctor<<declaration error> >’ does not have any field named ‘m_VigFunc’
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/VignettingCorrection.h:99: error: ‘vf’ was not declared in this scope
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/VignettingCorrection.h:99: error: class ‘vigra_ext::VigCorrDivFunctor<<declaration error> >’ does not have any field named ‘m_Adjust’
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/VignettingCorrection.h:99: error: ‘adj’ was not declared in this scope
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/VignettingCorrection.h: In member function ‘int vigra_ext::VigCorrDivFunctor<<declaration error> >::operator()(const int&, float, float) const’:
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/VignettingCorrection.h:112: error: there are no arguments to ‘m_InvResp’ that depend on a template parameter, so a declaration of ‘m_InvResp’ must be available
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/VignettingCorrection.h:112: error: there are no arguments to ‘m_VigFunc’ that depend on a template parameter, so a declaration of ‘m_VigFunc’ must be available
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/VignettingCorrection.h: At global scope:
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/VignettingCorrection.h:118: error: expected identifier before numeric constant
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/VignettingCorrection.h:118: error: expected ‘>’ before numeric constant
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/VignettingCorrection.h:120: error: default argument for template parameter for class enclosing ‘class vigra_ext::VigCorrFlatAddFunctor<<declaration error> >’
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/VignettingCorrection.h:124: error: expected unqualified-id before numeric constant
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/VignettingCorrection.h:128: error: ‘VT2’ does not name a type
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/VignettingCorrection.h:132: error: type/value mismatch at argument 1 in template parameter list for ‘template<class A> struct vigra::NumericTraits’
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/VignettingCorrection.h:132: error:   expected a type, got ‘65536’
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/VignettingCorrection.h:135: error: ‘VT2’ was not declared in this scope
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/VignettingCorrection.h:135: error: template argument 1 is invalid
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/VignettingCorrection.h:137: error: expected ‘,’ or ‘...’ before ‘&’ token
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/VignettingCorrection.h:137: error: ISO C++ forbids declaration of ‘InvResp’ with no type
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/VignettingCorrection.h:141: error: ‘InvResp’ does not name a type
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/VignettingCorrection.h:142: error: ‘Adjust’ does not name a type
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/VignettingCorrection.h:146: error: ‘first_argument_type’ has not been declared
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/VignettingCorrection.h:146: error: ‘second_argument_type’ has not been declared
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/VignettingCorrection.h: In constructor ‘vigra_ext::VigCorrFlatAddFunctor<<declaration error> >::VigCorrFlatAddFunctor(int)’:
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/VignettingCorrection.h:138: error: class ‘vigra_ext::VigCorrFlatAddFunctor<<declaration error> >’ does not have any field named ‘m_InvResp’
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/VignettingCorrection.h:138: error: ‘fr’ was not declared in this scope
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/VignettingCorrection.h:138: error: class ‘vigra_ext::VigCorrFlatAddFunctor<<declaration error> >’ does not have any field named ‘m_Adjust’
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/VignettingCorrection.h:138: error: ‘adj’ was not declared in this scope
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/VignettingCorrection.h: In member function ‘int vigra_ext::VigCorrFlatAddFunctor<<declaration error> >::operator()(const int&, const int&) const’:
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/VignettingCorrection.h:150: error: there are no arguments to ‘m_InvResp’ that depend on a template parameter, so a declaration of ‘m_InvResp’ must be available
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/VignettingCorrection.h: At global scope:
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/VignettingCorrection.h:154: error: expected identifier before numeric constant
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/VignettingCorrection.h:154: error: expected ‘>’ before numeric constant
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/VignettingCorrection.h:156: error: default argument for template parameter for class enclosing ‘class vigra_ext::VigCorrAddFunctor<<declaration error> >’
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/VignettingCorrection.h:160: error: expected unqualified-id before numeric constant
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/VignettingCorrection.h:164: error: type/value mismatch at argument 1 in template parameter list for ‘template<class A> struct vigra::NumericTraits’
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/VignettingCorrection.h:164: error:   expected a type, got ‘65536’
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/VignettingCorrection.h:168: error: expected ‘,’ or ‘...’ before ‘&’ token
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/VignettingCorrection.h:168: error: ISO C++ forbids declaration of ‘InvResp’ with no type
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/VignettingCorrection.h:172: error: ‘InvResp’ does not name a type
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/VignettingCorrection.h:173: error: ‘VigFunc’ does not name a type
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/VignettingCorrection.h:174: error: ‘Adjust’ does not name a type
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/VignettingCorrection.h:178: error: ‘first_argument_type’ has not been declared
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/VignettingCorrection.h: In constructor ‘vigra_ext::VigCorrAddFunctor<<declaration error> >::VigCorrAddFunctor(int)’:
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/VignettingCorrection.h:169: error: class ‘vigra_ext::VigCorrAddFunctor<<declaration error> >’ does not have any field named ‘m_InvResp’
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/VignettingCorrection.h:169: error: ‘fr’ was not declared in this scope
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/VignettingCorrection.h:169: error: class ‘vigra_ext::VigCorrAddFunctor<<declaration error> >’ does not have any field named ‘m_VigFunc’
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/VignettingCorrection.h:169: error: ‘vf’ was not declared in this scope
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/VignettingCorrection.h:169: error: class ‘vigra_ext::VigCorrAddFunctor<<declaration error> >’ does not have any field named ‘m_Adjust’
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/VignettingCorrection.h:169: error: ‘adj’ was not declared in this scope
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/VignettingCorrection.h: In member function ‘int vigra_ext::VigCorrAddFunctor<<declaration error> >::operator()(const int&, float, float) const’:
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/VignettingCorrection.h:182: error: there are no arguments to ‘m_InvResp’ that depend on a template parameter, so a declaration of ‘m_InvResp’ must be available
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/VignettingCorrection.h:182: error: there are no arguments to ‘m_VigFunc’ that depend on a template parameter, so a declaration of ‘m_VigFunc’ must be available
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/VignettingCorrection.h: In function ‘void vigra_ext::flatfieldVigCorrection(vigra::triple<Iterator, Iterator, Accessor>, std::pair<MaskIterator, MaskAccessor>, std::pair<MaskImageIterator, MaskAccessor>, double, double, bool, typename vigra::NumericTraits<typename SrcAccessor::value_type>::RealPromote, typename vigra::NumericTraits<typename SrcAccessor::value_type>::RealPromote, bool)’:
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/VignettingCorrection.h:446: error: wrong number of template arguments (4, should be 1)
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/VignettingCorrection.h:46: error: provided for ‘template<<declaration error> > class vigra_ext::VigCorrFlatDivFunctor’
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/VignettingCorrection.h:450: error: wrong number of template arguments (4, should be 1)
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/VignettingCorrection.h:119: error: provided for ‘template<<declaration error> > class vigra_ext::VigCorrFlatAddFunctor’
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/VignettingCorrection.h:457: error: wrong number of template arguments (4, should be 1)
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/VignettingCorrection.h:46: error: provided for ‘template<<declaration error> > class vigra_ext::VigCorrFlatDivFunctor’
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/VignettingCorrection.h:461: error: wrong number of template arguments (4, should be 1)
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/VignettingCorrection.h:119: error: provided for ‘template<<declaration error> > class vigra_ext::VigCorrFlatAddFunctor’
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/VignettingCorrection.h: In function ‘void vigra_ext::radialVigCorrection(vigra::triple<Iterator, Iterator, Accessor>, std::pair<MaskIterator, MaskAccessor>, double, double, const std::vector<double, std::allocator<double> >&, hugin_utils::FDiff2D, bool, typename vigra::NumericTraits<typename SrcAccessor::value_type>::RealPromote, typename vigra::NumericTraits<typename SrcAccessor::value_type>::RealPromote, bool)’:
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/VignettingCorrection.h:493: error: wrong number of template arguments (4, should be 1)
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/VignettingCorrection.h:86: error: provided for ‘template<<declaration error> > class vigra_ext::VigCorrDivFunctor’
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/VignettingCorrection.h:497: error: wrong number of template arguments (4, should be 1)
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/VignettingCorrection.h:155: error: provided for ‘template<<declaration error> > class vigra_ext::VigCorrAddFunctor’
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/VignettingCorrection.h:504: error: wrong number of template arguments (4, should be 1)
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/VignettingCorrection.h:86: error: provided for ‘template<<declaration error> > class vigra_ext::VigCorrDivFunctor’
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/VignettingCorrection.h:508: error: wrong number of template arguments (4, should be 1)
/Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/VignettingCorrection.h:155: error: provided for ‘template<<declaration error> > class vigra_ext::VigCorrAddFunctor’
make[2]: *** [src/hugin_script_interface/CMakeFiles/_hsi.dir/hsiPYTHON_wrap.cxx.o] Error 1
make[1]: *** [src/hugin_script_interface/CMakeFiles/_hsi.dir/all] Error 2
make: *** [all] Error 2

[kfj]

On 8 Feb., 16:27, Harry van der Wolf <hvdw...@gmail.com> wrote:
> I deactivated my libpano 2.9.17 and compiled 2.9.18.
>
> Then I compiled hugin using the "-DBUILD_HSI=ON" parameter.
> The compilation stopped with a lot of errors. See below.

Hi Harry!

first of all - thanks for trying. The file producing all the errors is
obviously

/Users/Shared/development/hugin_related/huginpython/src/hugin_base/
algorithms/nona/NonaImageStitcher.h

Since this isn't really crucial, just to go forward you may want to
exclude it from being wrapped by SWIG. This is fairly simple: The file
is referenced twice, first in the block bracketed in %{ and %}, then
further down, where the actual %include statement pulls it into the
interface. Just comment out both occurences - I've done so with
another file which failed to wrap here (ImageGraph.h, just below).
Then just try again, and if you're lucky you'll get a compiling
wrapper, just without the NonaImageStitcher bit, which is really a
side issue and only has to be dealt with if someone really needs it -
and for special cases, dedicated interface code can be writen.

The problem reminds me of what happened here when I tried to build the
interface with distutils, instead of letting cmake handle it all.
There must be a subtle difference between flags passed to the compiler
or a similar incompatibility, I haven't figured it out yet. even if
the compile succeeds, you're not necessarily there yet - but if the
module is built and you can import it from a python interpreter
prompt, you can be quite confident things have worked out.

Sorry I can't be more helpful, since I don't have Mac to try this
myself. Stay in touch!

Kay

[Harry van der Wolf]

Hi,

2011/2/9 kfj <_k...@yahoo.com>

On 8 Feb., 16:27, Harry van der Wolf <hvdw...@gmail.com> wrote:
> I deactivated my libpano 2.9.17 and compiled 2.9.18.
>
> Then I compiled hugin using the "-DBUILD_HSI=ON" parameter.
> The compilation stopped with a lot of errors. See below.

Hi Harry!

first of all - thanks for trying. The file producing all the errors is
obviously

/Users/Shared/development/hugin_related/huginpython/src/hugin_base/
algorithms/nona/NonaImageStitcher.h

Since this isn't really crucial, just to go forward you may want to
exclude it from being wrapped by SWIG. This is fairly simple: The file
is referenced twice, first in the block bracketed in %{ and %}, then
further down, where the actual %include statement pulls it into the
interface. Just comment out both occurences

I did a quick check but couldn't find it.
From which file do I have to uncomment this?

Harry

[kfj]

On 9 Feb., 14:17, Harry van der Wolf <hvdw...@gmail.com> wrote:

> I did a quick check but couldn't find it.
> From which file do I have to uncomment this?

sorry, Harry, I didn't mention it - I'm so in the code it's obvious to
me. It's in the interface definition file:

<source>/src/hugin_script_interface/hsi.i

This is the file which SWIG translates into

<build>/src/hugin_script_interface/CMakeFiles/_hsi.dir/
hsiPYTHON_wrap.cxx

Which is where your compiler found the errors. The lines with the
#include, and the %include statement, respectively, and which I
recomennd to try and comment out, are

51 and 269

once the i-file is modified, cmake should take care of propagating the
change through the cxx to the object file and, if all goes well, into
the module.

Kay

[Harry van der Wolf]

Hi

2011/2/9 kfj <_k...@yahoo.com>

It's in the interface definition file:

<source>/src/hugin_script_interface/hsi.i

recomennd to try and comment out, are

51 and 269

did that, got past that, got new errors: see below.

I checked already but VignettingCorrection.h is not in the hsi.i file.

Next hint please :)

Harry


[ 52%] Building CXX object src/hugin_script_interface/CMakeFiles/_hsi.dir/hsiPYTHON_wrap.cxx.o
In file included from /Users/Shared/development/hugin_related/huginpython/src/hugin_base/algorithms/optimizer/PhotometricOptimizer.h:34,
                 from /Users/Shared/development/hugin_related/huginpython/build/src/hugin_script_interface/hsiPYTHON_wrap.cxx:3036:

[kfj]

On 9 Feb., 17:00, Harry van der Wolf <hvdw...@gmail.com> wrote:

> > recomennd to try and comment out, are
>
> > 51 and 269
>
> did that, got past that, got new errors: see below.
>
> I checked already but VignettingCorrection.h is not in the hsi.i file.
>
> Next hint please :)
>

It's not VignettingCorrection.h, but PhotometricOptimizer.h. a bit
down from the lines you commented out. Try throw that out for good
measure, so to be specific, comment out lines 53 and 272.

> [ 52%] Building CXX object
> src/hugin_script_interface/CMakeFiles/_hsi.dir/hsiPYTHON_wrap.cxx.o
> In file included from

> /Users/Shared/development/hugin_related/huginpython/src/hugin_base/algorithms/optimizer/PhotometricOptimizer.h:34, <<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<< likely problem

I'll have a look to see if I can make more sense from the error output
as well.

Kay

[kfj]

On 9 Feb., 17:00, Harry van der Wolf <hvdw...@gmail.com> wrote:

> ....

> /Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/VignettingCorrection.h:87:
> error: default argument for template parameter for class enclosing ‘class
> vigra_ext::VigCorrDivFunctor<<declaration error> >’
> /Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/VignettingCorrection.h:91:
> error: expected unqualified-id before numeric constant
> /Users/Shared/development/hugin_related/huginpython/src/hugin_base/vigra_ext/VignettingCorrection.h:95:
> error: type/value mismatch at argument 1 in template parameter list for
> ‘template<class A> struct vigra::NumericTraits’

> ...
etc. etc.

Well, I had a look at the vigra_ext code, and it's full of evil C++
trickery, which somehow the compiler on my system can palate, but
your's seems to choke on it. I can't give you better advice than
trying to avoid the offending headers, like the last two which we
commented out. To truly debug that sort of problem, it'd be advisable
to understand what the code does. There doesn't seem to be any
documentation about the goings-on, and so ultimately it's 'ask Pablo',
since he's listed as the author. Maybe he has an idea. But as I've
said before, this part of the wrap isn't really crucial - more nice-to-
have, so if we'd get away for now with just throwing a few headers out
of the i-file, this would be great. Do stay in touch and let me know
if you run into more problems.

Kay

[Harry van der Wolf]

Hi,

2011/2/9 kfj <_k...@yahoo.com>

> I checked already but VignettingCorrection.h is not in the hsi.i file.
>
> Next hint please :)
>

It's not VignettingCorrection.h, but PhotometricOptimizer.h. a bit
down from the lines you commented out. Try throw that out for good
measure, so to be specific, comment out lines 53 and 272.

I removed src/hugin_base/algorithms/optimizer/PhotometricOptimizer.h and that fixed the compilation.

There is a target _hsi and a shared module _hsi.so is created. I assume this a dynamic library for a linux system. If it is a shared library, the name should be _hsi.dylib as that's the dynamic name/extension on macosx. If I'm wrong, nice. If I'm right the name needs to be changed.

Hugin starts, but I have no idea what to do with it.

Do you have a sample script and image set? This would also enable simpler and uniform testing on multiple OSes.

Harry

[Harry van der Wolf]

2011/2/9 Harry van der Wolf <hvd...@gmail.com>

Do you have a sample script and image set? This would also enable simpler and uniform testing on multiple OSes.

Harry

Sorry, I now see that the available scripts work on any image set.

 

Harry

[kfj]

On 10 Feb., 09:23, Harry van der Wolf <hvdw...@gmail.com> wrote:

> There is a target _hsi and a shared module _hsi.so is created. I assume this
> a dynamic library for a linux system. If it is a shared library, the name
> should be _hsi.dylib as that's the dynamic name/extension on macosx. If I'm
> wrong, nice. If I'm right the name needs to be changed.

I'm not sure what extension a shared library has to have on Mac OS.
I'll explain what these files are to you, and if you want a slightly
more in-depth explanation about the files and how they come to be,
there is a README.hsi file in the source directory.

hsi.i (the interface definition file), is converted by SWIG into the
files hsi.py (the high-level-interface) and hsiPYTHON_wrap.cxx (the
low-level-interface). The latter is compiled into _hsi.so. When you
import hsi.py into a python session, it will try and load _hsi.so. So
I'd ask you to do the following:

- Open a shell prompt
- set your PYTHONPATH environment variable to the target directory
where cmake has placed hsi.py and _hsi.so
(like, type export PYTHONPATH=<whatever that path is>)
- start a python interpreter session by typing 'python'
- at the python prompt, type 'import hsi'

after that, if all goes well, the module is loaded, and you can try
stuff with it, like

help(hsi)

If that works, we have hsi runing, and I'd like you to report back.
If it doesn't work, chances are that all that's needed is a name
change of _hsi.so to whatever is appropriate for python modules on Mac
OS, but I'd ask you to try with the unchanged name first.

> Sorry, I now see that the available scripts work on any image set.

Does this actually mean that you could execute one of the plugins from
hugin, or does it mean that you had a look at the scripts and saw that
they would work on any image set?

If you got the plugins to load and execute from hugin, I'd love to
hear that because it'd be another milestone.

Kay

[kfj]

On 10 Feb., 11:12, kfj <_...@yahoo.com> wrote:

> If you got the plugins to load and execute from hugin, I'd love to
> hear that because it'd be another milestone.

And I haven't heard anything since. Anybody out there? Please! This
seems to be like 1 1/2 steps before we've got the last platform
covered, so it's really frustrating to come to a grinding halt here.

Kay

[Harry van der Wolf]

Hi,

2011/2/13 kfj <_k...@yahoo.com>

I understand your feeling, but I've been busy thursday/friday with 2 big upgrades at work. And normally in the weekend I have more time. Not this weekend. I have just returned.

I tried with a set both manually and via the assistant to see whether that makes a difference. Manually I tried before and after optimizing.
I tried the python script top_five.py but I get an error:

  Result
  Script return -2

The script doesn't reduce the number of CPs (and there are far more than 5 CPs per image pair).
The other scripts don't work either. I also tried the xx3 scripts (python 3 I suppose?)
Can I find more info about what goes wrong? It's not in the system.log

Harry

[kfj]

On 13 Feb., 20:32, Harry van der Wolf <hvdw...@gmail.com> wrote:


> The other scripts don't work either. I also tried the xx3 scripts (python 3
> I suppose?)
> Can I find more info about what goes wrong? It's not in the system.log

It's not the scripts which aren't working, but the interface module
not being loaded, because hugin doesn't know where to find it. Merely
clicking on 'execute Python script' and hoping the relevant files will
be found won't get you anywhere. Four days ago I posted a message
explaining what I'd like you to do and why, so here it is again:

So, again, once you've established you can import hsi into a python
session, we can proceed to the next step and call it from hugin. But
first things first.

Kay

[Harry van der Wolf]

Hi

2011/2/14 kfj <_k...@yahoo.com>

- Open a shell prompt
- set your PYTHONPATH environment variable to the target directory
where cmake has placed hsi.py and _hsi.so
 (like, type export PYTHONPATH=<whatever that path is>)
- start a python interpreter session by typing 'python'
- at the python prompt, type 'import hsi'

cmake hadn't installed it. So I had to set the python path to the cmake build folder.

I got the following error upon importing:
$ python
Python 2.6.6 (r266:84292, Dec 27 2010, 17:46:10)
[GCC 4.2.1 (Apple Inc. build 5664)] on darwin
Type "help", "copyright", "credits" or "license" for more information.
>>> import hsi
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
  File "hsi.py", line 25, in <module>
    _hsi = swig_import_helper()
  File "hsi.py", line 21, in swig_import_helper
    _mod = imp.load_module('_hsi', fp, pathname, description)
ImportError: dlopen(./_hsi.so, 2): no suitable image found.  Did find:
    ./_hsi.so: mach-o, but wrong architecture
>>>


The "./_hsi.so: mach-o, but wrong architecture" is is a weird error. The _hsi.so is an i386 library like all the rest.

Harry

[kfj]

On 14 Feb., 19:41, Harry van der Wolf <hvdw...@gmail.com> wrote:

> cmake hadn't installed it. So I had to set the python path to the cmake
> build folder.

You did the right thing for now. That is the current status quo on the
Mac. If you look at the CMakeLists.txt in the python_scripting
directory, you will see that the installation paths for, amongst other
things, Mac OS aren't set, because we didn't know what these paths
should be for python modules on Mac OS. This is where the Mac OS
wizards come in and set these paths to what they know to be
appropriate ;-)

> I got the following error upon importing:

> ...

> ImportError: dlopen(./_hsi.so, 2): no suitable image found.  Did find:
>     ./_hsi.so: mach-o, but wrong architecture
>
> The "./_hsi.so: mach-o, but wrong architecture" is is a weird error. The
> _hsi.so is an i386 library like all the rest.

This is indeed a weird error, but with this one I can't help you
beyond guesswork. Did you make an attempt to rename _hsi.so to what
you think is the accepted name for shared libraries, just on the off
chance that's the problem? And are you sure they're all either 32 bit
or 64 bit?

I've googled your error, and it seems to be a common thing, often
occuring when there are 32/64 bit problems. Often the tool lipo is
mentioned to have a look at the libraries in question.

The next possibility is this: When importing a python module which has
been compiled from C or C++, if I'm not mistaken, the module should
ideally be compiled with the same compiler that the python interpreter
was compiled with. So if the python you have on your mac was compiled
with apple's compiler and your module is made by something else,

things might go wrong. Your Python interpreter says:

[GCC 4.2.1 (Apple Inc. build 5664)] on darwin

What gcc is being used to build hugin, and therefore the python
module? On my system gcc is 4.4.5, so maybe there's a problem here. It
may be enough to get a recent build of python, the best for
compatibilty's sake would be to compile pyhon from source with the
self-same compiler you're using otherwise.

If that doesn't help, I'd kindly ask you to try and find some help
from someone who knows the ins and outs of building and using SWIG
modules on the Mac - you'd be much better suited for this task than I,
because I've never had a Mac, let alone programmed one. SWIG modules
are a very common thing after all, many large libraries use SWIG to
create their python bindings, so it should be a very well-known thing.
If you can't find help, please let me know as well, because then we
have a genuine problem at hand and have to figure out a way of dealing
with it.

Kay

[Harry van der Wolf]

2011/2/15 kfj <_k...@yahoo.com>

On 14 Feb., 19:41, Harry van der Wolf <hvdw...@gmail.com> wrote:

> cmake hadn't installed it. So I had to set the python path to the cmake
> build folder.

You did the right thing for now. That is the current status quo on the
Mac. If you look at the CMakeLists.txt in the python_scripting
directory, you will see that the installation paths for, amongst other
things, Mac OS aren't set, because we didn't know what these paths
should be for python modules on Mac OS. This is where the Mac OS
wizards come in and set these paths to what they know to be
appropriate ;-)

> I got the following error upon importing:

> ...

> ImportError: dlopen(./_hsi.so, 2): no suitable image found.  Did find:
>     ./_hsi.so: mach-o, but wrong architecture
>
> The "./_hsi.so: mach-o, but wrong architecture" is is a weird error. The
> _hsi.so is an i386 library like all the rest.

This is indeed a weird error, but with this one I can't help you
beyond guesswork. Did you make an attempt to rename _hsi.so to what
you think is the accepted name for shared libraries, just on the off
chance that's the problem? And are you sure they're all either 32 bit
or 64 bit?

Absolutely, that's exactly why I mention it as OSX is (AFAIK) the only platform where I can define for which architecture to build, without cross compiling.

 

I've googled your error, and it seems to be a common thing, often
occuring when there are 32/64 bit problems. Often the tool lipo is
mentioned to have a look at the libraries in question.

Yes, that's how I checked that everything is i386.
 

The next possibility is this: When importing a python module which has
been compiled from C or C++, if I'm not mistaken, the module should
ideally be compiled with the same compiler that the python interpreter
was compiled with. So if the python you have on your mac was compiled
with apple's compiler and your module is made by something else,

things might go wrong. Your Python interpreter says:

[GCC 4.2.1 (Apple Inc. build 5664)] on darwin

What gcc is being used to build hugin, and therefore the python
module? On my system gcc is 4.4.5, so maybe there's a problem here. It
may be enough to get a recent build of python, the best for
compatibilty's sake would be to compile pyhon from source with the
self-same compiler you're using otherwise.

The gcc used is 4.2.1. The SDK used is 10.5 to be able to create i386 libs/binaries.
I need to compile Hugin et al very strictly as i386 as the current wxwindows can't be compiled as x86_64 which is the default on Snow Leopard on intel.
In my first attempt I didn't think of it and hugin et al were compiled as x86_64 and couldn't be linked against wxwindows.

My python 2.6 is both i386/x86_64 as most libraries/binaries/frameworks on OSX.

I renamed the .so to .dylib, but that doesn't make a difference, but I can't find where to change that it will be named a .dylib.
I think that's also where the weird error comes from. When the system encounters an .so I assume it is expecting a "real" .so and not some incorrectly named .dylib.
I removed the remarks in front of the install targets for the bottom ELSE(win32) in CMakeLists.txt and the _hsi.so got installed in /usr/local/lib which is correct.
The hsi.py and hpi.py is also installed in /usr/local/lib where I think they should have been installed in /usr/local/bin (where the other .py files reside that I installed as a check).

I'll see how I can get it compiled as a .dylib. All other libs are .dylib including libhugin_python_interface.dylib

Harry

[kfj]

On 15 Feb., 22:02, Harry van der Wolf <hvdw...@gmail.com> wrote:

> I need to compile Hugin et al very strictly as i386 as the current wxwindows
> can't be compiled as x86_64 which is the default on Snow Leopard on intel.
> In my first attempt I didn't think of it and hugin et al were compiled as
> x86_64 and couldn't be linked against wxwindows.

Are you sure this is still the case? I've done some superficial
googling on the topic, and it looks to me as if there might be
possibilities to compile wxWindows for x86_64. Have a look at

http://wiki.wxwidgets.org/Development:_wxMac

> My python 2.6 is both i386/x86_64 as most libraries/binaries/frameworks on
> OSX.

The difference sounds problematic. Does that actually mean that hugin
and python are compiled to different standards? This might explain the
'mach-o, but wrong architecture' error.

> I renamed the .so to .dylib, but that doesn't make a difference, but I can't
> find where to change that it will be named a .dylib.
> I think that's also where the weird error comes from. When the system
> encounters an .so I assume it is expecting a "real" .so and not some
> incorrectly named .dylib.

How can you tell the difference between the two? Really, cmake is a
very capable build tool, and if it produces an .so as a SWIG module, I
suspect it has good reason to do so - unless, of course, it's been
ported sloppily to the Mac. On the other platforms, we rely on cmake
to generate aptly named files, and manually interfering is generally
not a good idea. The different extension might even reflect the 'mach-
o but wrong architecture' error. Really, extensions shouldn't matter -
it was just a wild shot in case they did, even though they shouldn't.

> I removed the remarks in front of the install targets for the bottom
> ELSE(win32) in CMakeLists.txt and the _hsi.so got installed in
> /usr/local/lib which is correct.
> The hsi.py and hpi.py is also installed in /usr/local/lib where I think they
> should have been installed in /usr/local/bin (where the other .py files
> reside that I installed as a check).

Fine. So you've worked out working target directories for the Mac. /
usr/local/lib and /usr/local/bin don't look like typical Python module
paths to me - the modules may be found, since they are part of the
system path, but usually Python also checks in directories like /usr/
local/lib/python2.6/dist-packages or /usr/local/lib/python2.6/site-
packages for local python modules. It may be cleaner to use one of
those, but for now, since we haven't even got the stuff to run yet,
you might as well stick with what works for you.

> I'll see how I can get it compiled as a .dylib. All other libs are .dylib
> including libhugin_python_interface.dylib

Thank you for your efforts. I hope it will work out eventually. Keep
me updated on your progress!

Kay

[Harry van der Wolf]

2011/2/16 kfj <_k...@yahoo.com>

On 15 Feb., 22:02, Harry van der Wolf <hvdw...@gmail.com> wrote:

> I need to compile Hugin et al very strictly as i386 as the current wxwindows
> can't be compiled as x86_64 which is the default on Snow Leopard on intel.
> In my first attempt I didn't think of it and hugin et al were compiled as
> x86_64 and couldn't be linked against wxwindows.

Are you sure this is still the case? I've done some superficial
googling on the topic, and it looks to me as if there might be
possibilities to compile wxWindows for x86_64. Have a look at

http://wiki.wxwidgets.org/Development:_wxMac

That's the development version 2.9. We have issues and work-arounds already for the stable 2.8.11 version (and previous). I have two separate repositories on my macbook: a macports repository and a hugin repositories with a huge load of redundant libs and so on and both have wxmac 2.8.11. I'm not going to compile 2.9 right now.
 

> My python 2.6 is both i386/x86_64 as most libraries/binaries/frameworks on
> OSX.

The difference sounds problematic. Does that actually mean that hugin
and python are compiled to different standards? This might explain the
'mach-o, but wrong architecture' error.

No, it's not. That's exactly how OSX works. My bundles are i386/ppc/x86_64 architecture bundles. That's not 3 libraries or 3 binaries for each architecture, but one library or binary containing all architectures.
You get used to that if you build for OSX ;)
 

> I renamed the .so to .dylib, but that doesn't make a difference, but I can't
> find where to change that it will be named a .dylib.
> I think that's also where the weird error comes from. When the system
> encounters an .so I assume it is expecting a "real" .so and not some
> incorrectly named .dylib.

How can you tell the difference between the two? Really, cmake is a
very capable build tool, and if it produces an .so as a SWIG module, I
suspect it has good reason to do so - unless, of course, it's been
ported sloppily to the Mac. On the other platforms, we rely on cmake
to generate aptly named files, and manually interfering is generally
not a good idea. The different extension might even reflect the 'mach-
o but wrong architecture' error. Really, extensions shouldn't matter -
it was just a wild shot in case they did, even though they shouldn't.

As the tools tell that's is a mach-o library, it MUST be a dylib as that's the OSX linker standard for building libraries. An .so is not a mach-o library but generally an ELF library. If cmake/swig builds an .so for OSX then it's doing something wrong. I assume that some header/meta data or wahtever confuses the otool and lipo and link/loader tools.
I did some googling and I have seen .dylibs created from swig.

Does swig build an .so on windows or does it build a .dll?

 

> I removed the remarks in front of the install targets for the bottom
> ELSE(win32) in CMakeLists.txt and the _hsi.so got installed in
> /usr/local/lib which is correct.
> The hsi.py and hpi.py is also installed in /usr/local/lib where I think they
> should have been installed in /usr/local/bin (where the other .py files
> reside that I installed as a check).

Fine. So you've worked out working target directories for the Mac. /
usr/local/lib and /usr/local/bin don't look like typical Python module
paths to me - the modules may be found, since they are part of the
system path, but usually Python also checks in directories like /usr/
local/lib/python2.6/dist-packages or /usr/local/lib/python2.6/site-
packages for local python modules. It may be cleaner to use one of
those, but for now, since we haven't even got the stuff to run yet,
you might as well stick with what works for you.

I think it depends on the package. I also see /usr/local/share/<package>/{lib,tool,contrib}.
I assume python gives you a lot of "freedom".


I'll try some more.

Harry

[kfj]

On 16 Feb., 13:47, Harry van der Wolf <hvdw...@gmail.com> wrote:

> That's the development version 2.9. We have issues and work-arounds already
> for the stable 2.8.11 version (and previous). I have two separate
> repositories on my macbook: a macports repository and a hugin repositories
> with a huge load of redundant libs and so on and both have wxmac 2.8.11. I'm
> not going to compile 2.9 right now.

Let's hope we get it to work with 2.8.11 then.

> > > I renamed the .so to .dylib, but that doesn't make a difference, but I
> > > can't find where to change that it will be named a .dylib.

In the cmake documentation it says that you can pass compiler flags to
SWIG by:

SET_SOURCE_FILES_PROPERTIES( ${swig_generated_file_fullname}
PROPERTIES COMPILE_FLAGS "-bla")

Maybe that's where you can put in the flag to build a .dylib instead
of an .so

> > > I think that's also where the weird error comes from. When the system
> > > encounters an .so I assume it is expecting a "real" .so and not some
> > > incorrectly named .dylib.

> As the tools tell that's is a mach-o library, it MUST be a dylib as that's
> the OSX linker standard for building libraries. An .so is not a mach-o
> library but generally an ELF library. If cmake/swig builds an .so for OSX
> then it's doing something wrong. I assume that some header/meta data or
> wahtever confuses the otool and lipo and link/loader tools.
> I did some googling and I have seen .dylibs created from swig.
>
> Does swig build an .so on windows or does it build a .dll?

I suppose it's not an .so, but no dll either. I think on Windows they
have a .pyd extension, but you'd have to ask Thomas to be sure, since
he managed to get it to run on Windows.

> > Fine. So you've worked out working target directories for the Mac. /
> > usr/local/lib and /usr/local/bin don't look like typical Python module
> > paths to me - the modules may be found, since they are part of the
> > system path, but usually Python also checks in directories like /usr/
> > local/lib/python2.6/dist-packages or /usr/local/lib/python2.6/site-
> > packages for local python modules. It may be cleaner to use one of
> > those, but for now, since we haven't even got the stuff to run yet,
> > you might as well stick with what works for you.
>
> I think it depends on the package. I also see
> /usr/local/share/<package>/{lib,tool,contrib}.
> I assume python gives you a lot of "freedom".

I chose the one I'm using because it was in Python's module path
already. That way I had it out of the system path and still didn't
have to change anything to make Python find it.

> I'll try some more.

That's about the only good news on the topic today ;-)

Kay

[Harry van der Wolf]

I need something like
SET(CMAKE_SHARED_MODULE_CREATE_C_FLAGS
    "${CMAKE_SHARED_MODULE_CREATE_C_FLAGS}  -bundle")

to compile _hsi.so as _hsi.dylib

In older versions it might be
SET(CMAKE_SHARED_MODULE_CREATE_C_FLAGS
    "${CMAKE_SHARED_MODULE_CREATE_C_FLAGS}  -bundle -flat_namespace -undefined suppress")

I tried it on a couple of places but I stil get a _hsi.so.

Any hints?

[T. Modes]

> In the cmake documentation it says that you can pass compiler flags to
> SWIG by:
>
> SET_SOURCE_FILES_PROPERTIES( ${swig_generated_file_fullname}
>         PROPERTIES COMPILE_FLAGS "-bla")
>
> Maybe that's where you can put in the flag to build a .dylib instead
> of an .so

In our case it should be:
SET_TARGET_PROPERTIES( ${SWIG_MODULE_hsi_REAL_NAME} PROPERTIES
COMPILE_FLAGS "-bundle" )

> > Does swig build an .so on windows or does it build a .dll?
>
> I suppose it's not an .so, but no dll either. I think on Windows they
> have a .pyd extension, but you'd have to ask Thomas to be sure, since
> he managed to get it to run on Windows.

On windows it is a dll with .pyd extension (even if hugin_base is a
static lib, as default on windows)

Thomas

[Harry van der Wolf]

2011/2/16 T. Modes <Thomas...@gmx.de>

In our case it should be:
SET_TARGET_PROPERTIES( ${SWIG_MODULE_hsi_REAL_NAME} PROPERTIES
COMPILE_FLAGS "-bundle" )

I tried that but always keep the _hsi.so, so I assume that that's the correct format after all or at least what swig and cmake make up together.
I tried multiple times with a shipload of other cmake settings with regard to OSX SDKs and so on but it simply doesn't make a difference. And I always keep the same error message.

As I will go on wintersports holiday early tomorrow morning I advice to move on with it and hopefully OSX can pick up later. Maybe that I have a fresh mind next week and see the light.

Harry

[kfj]

On 18 Feb., 13:57, Harry van der Wolf <hvdw...@gmail.com> wrote:

> I tried that but always keep the _hsi.so, so I assume that that's the
> correct format after all or at least what swig and cmake make up together.
> I tried multiple times with a shipload of other cmake settings with regard
> to OSX SDKs and so on but it simply doesn't make a difference. And I always
> keep the same error message.

bad news :(

> As I will go on wintersports holiday early tomorrow morning I advice to move
> on with it and hopefully OSX can pick up later. Maybe that I have a fresh
> mind next week and see the light.

Happy skiing! Is there noone else you could ask to look at our
problem? It might help to have a few more people look at at; I can't
imagine we're the first ones to stumble upon the issue.

Be that as it may, I'll gladly do the moving on - I'm already relying
happily on hsi for my own scripting, and so far it does precisely what
it's supposed to do. I wish there were more people using it so we
could get an idea if it works just as well for others. Then we could
move it from alpha to beta and maybe even consider designing a proper
plugin interface that goes beyond unconditionally calling a plugin
with the current panorama as the only parameter and no feedback apart
from a measly integer return code.

Thank you for keeping the faith and trying again and again. At least
you have achieved the first two steps: got SWIG to convert the
interface definition and modified the interface definition so that the
resulting code compiles without errors. I'm confident the next steps
will be sorted out eventually.

Kay

[Harry van der Wolf]

2011/2/18 kfj <_k...@yahoo.com>

Happy skiing! Is there noone else you could ask to look at our
problem? It might help to have a few more people look at at; I can't
imagine we're the first ones to stumble upon the issue.

There are a few others who could do it, but I "see" them very infrequently on hugin-ptx. I imagine that if they are following this post and want to pick up, they will. I too would like to see a blooming OSX community of builders/developers, but unfortunately, it is not that way.

And as info to them: Currently it's only a cmake build. I didn't touch the OSX project yet. 
The "Hugin Compiling OSX"[1] at wiki.panotools is up-to-date.

The only change to the cmake command is the addition of the "-DBUILD_HSI=ON" option like "cmake BUILD_HSI=ON ..".

Harry


1: <http://wiki.panotools.org/Hugin_Compiling_OSX>

[Harry van der Wolf]

2011/2/18 Harry van der Wolf <hvd...@gmail.com>

I gave it another shot after so many changes in the week I was away. However, I still keep the same errors upon loading hsi into python.

I then tried to build it 64bit as the libhugin_python_interface and _hsi.so are built before the wxmac stuff (which can't be built 64bit on OSX for 2.8.x). The issue was that hsi couldn't be loaded into python. Even without being able to built Hugin 64bit it might be possible to load the hsi into python.
Unfortunately I can't build the libhugin_python_interface and _hsi.so as 64 bit.

Currently the 2011.0 release cycle is more important but I will keep trying.

However, now that 2.9.1 has been officially released, be it a development release, it has also been ported to macports. That will allow for a cmake build of "python" hugin as 64 bit. In that case I will also trying to investigate why the 64bit doesn't build currently.

I will also built 2.9.1 in my own repository for the XCode build, but after 2011.0 has been released as that one has been based on 2.8.11 until now. I don't want to introduce a new variable just before release. It will also require a huge change in the XCode project as all wxwidgets (wxmac) related stuff is forced to be built 32bit. That is also a massive change I don't want to do when being in a release cycle.

Note though that the macports/cmake repository is different from the XCode repository giving me the option to push forward with the cmake build of "python"  hugin.


Harry

[kfj]

On 2 Mrz., 22:04, Harry van der Wolf <hvdw...@gmail.com> wrote:

> I gave it another shot after so many changes in the week I was away.
> However, I still keep the same errors upon loading hsi into python.
>

> ...

Thank you for trying, Harry. I hope that someone else can make sense
of the problems you describe - since I'm blissfully ignorant of all
things Mac, I don't really understand what the issue is. I thought
Windows programming was bad, but it seems that it's even worse your
end.

Anyway, I think there is some support for the python_scripting branch
being merged back into trunk and available to be built with a compile-
time switch, just as it is now in it's separate branch. The python-
enabled Mac version will just be a non-issue until someone figures out
the way through the version- bit- bundle- and port-jungle, but it's
absence won't be a show-stopper, and the matter isn't urgent (or if it
is for anyone they can sort it out, then ;-)

Kay