KDE4 version, but it seems to have some issues with liborigin (the version bundled isn't in portage, and the one bundled doesn't build)
Attaching a incomplete ebuild anyway for anyone to play with.
Created attachment 211746 [details]
Created attachment 212183 [details]
A bit more baked ebuild
O.K, looks our liborigin is renamed to liborigin2 so renaming in CMakeLists.txt and doing CMAKE_IN_SOURCE_BUILD=1 helps to overcome that problem, but it dies later on hdf5,
/usr/include/H5Gpublic.h:78: error: too few arguments to function ‘hid_t H5Gopen2(hid_t, const char*, hid_t)’
/var/tmp/portage/sci-visualization/labplot-2.0.0_alpha2/work/LabPlot-2.0.0.alpha2/src/filter/FilterHDF5.cc:81: error: at this point in file
(In reply to comment #2)
> Created an attachment (id=212183) [details]
> A bit more baked ebuild
> O.K, looks our liborigin is renamed to liborigin2 so renaming in CMakeLists.txt
> and doing CMAKE_IN_SOURCE_BUILD=1 helps to overcome that problem, but it dies
> later on hdf5,
> /usr/include/H5Gpublic.h:78: error: too few arguments to function ‘hid_t
> H5Gopen2(hid_t, const char*, hid_t)’
> error: at this point in file
see bug 289172. You can always try H5_USE_16_API
(In reply to comment #3)
> see bug 289172. You can always try H5_USE_16_API
That worked, but it's still dying later, this time on gsl,
[ 72%] In file included from /var/tmp/portage/sci-visualization/labplot-2.0.0_alpha2/work/LabPlot-2.0.0.alpha2/src/parser/parser.h:31,
/var/tmp/portage/sci-visualization/labplot-2.0.0_alpha2/work/LabPlot-2.0.0.alpha2/src/parser/constants.h:330: error: ‘GSL_CONST_CGS_BOHR_MAGNETON’ undeclared here (not in a function)
(In reply to comment #4)
> (In reply to comment #3)
> > see bug 289172. You can always try H5_USE_16_API
> That worked, but it's still dying later, this time on gsl,
> [ 72%] In file included from
> from parser.y:4:
> error: ‘GSL_CONST_CGS_BOHR_MAGNETON’ undeclared here (not in a function)
This time see bug 286605 comment #6 and patches therein.
Created attachment 212296 [details]
TODO: Fix main CMakeLists.txt to accept USE flags for gsl, cdf, netcdf and hdf5.
Still dying on emerge,
/var/tmp/portage/sci-visualization/labplot-2.0.0_alpha2/work/LabPlot-2.0.0.alpha2/src/filter/FilterCDF.cc:47: error: prototype for ‘QString FilterCDF::Enc(int)’ does not match any in class ‘FilterCDF’
/var/tmp/portage/sci-visualization/labplot-2.0.0_alpha2/work/LabPlot-2.0.0.alpha2/src/filter/FilterCDF.h:35: error: candidate is: QString FilterCDF::Enc(int) const
Created attachment 212298 [details, diff]
The above compiles, but only if you uninstall sci-libs/cdf (as in not have HAVE_CDF defined),
but it fails at linking even without sci-libs/cdf as following,
Linking CXX executable LabPlot
CMakeFiles/LabPlot.dir/filter/FilterOPJ.o: In function `FilterOPJ::import()':
/var/tmp/portage/sci-visualization/labplot-2.0.0_alpha2/work/LabPlot-2.0.0.alpha2/src/filter/FilterOPJ.cc:19: undefined reference to `OPJFile::OPJFile(char const*)'
/var/tmp/portage/sci-visualization/labplot-2.0.0_alpha2/work/LabPlot-2.0.0.alpha2/src/filter/FilterOPJ.cc:22: undefined reference to `OPJFile::Parse()'
collect2: ld returned 1 exit status
I doubt I'll waste more time on this, feel free to pick up the pieces ;-)
http://www.asinen.org/2009/10/labplot-scidavis/ might be of interest to see whether we still want a labplot
The latest _alpha_ release was in 2008, so I consider upstream dead.
All released versions are kde3-only.
*** Bug 399607 has been marked as a duplicate of this bug. ***
Not remembering the one here, I made a quick-n-dirty live ebuild for labplot-2.0 branch and put it in the sci overlay.
The patch is not needed anymore, and it has no useflags yet.
Anyway, I'm less than impressed by the result. Needs still a LOT of upstream work.