Source: molmodel
Severity: serious
Version: 3.1.0-2
Hello,
molmodel FTBFS after recent update of gemmi to 0.6.2+ds-1:
/home/merkys/molmodel-3.1.0/src/PDBReader.cpp: In function
'gemmi::Structure getStructureFromFile(const std::string&)':
/home/merkys/molmodel-3.1.0/src/PDBReader.cpp:98:29: error:
'make_structure_from_block' is not a member of 'gemmi::impl'; did you
mean 'gemmi::make_structure_from_block'?
98 | return gemmi::impl::make_structure_from_block (
gemmiDoc.blocks.at(0) );
| ^~~~~~~~~~~~~~~~~~~~~~~~~
In file included from /usr/include/gemmi/mmread.hpp:13,
from
/home/merkys/molmodel-3.1.0/./include/molmodel/internal/Compound.h:48,
from
/home/merkys/molmodel-3.1.0/./include/molmodel/internal/CompoundSystem.h:6,
from
/home/merkys/molmodel-3.1.0/./include/molmodel/internal/PDBReader.h:36,
from /home/merkys/molmodel-3.1.0/src/PDBReader.cpp:33:
/usr/include/gemmi/mmcif.hpp:15:21: note:
'gemmi::make_structure_from_block' declared here
15 | GEMMI_DLL Structure make_structure_from_block(const cif::Block&
block);
| ^~~~~~~~~~~~~~~~~~~~~~~~~
make[3]: *** [CMakeFiles/SimTKmolmodel.dir/build.make:219:
CMakeFiles/SimTKmolmodel.dir/src/PDBReader.cpp.o] Error 1
make[3]: *** Waiting for unfinished jobs....
/home/merkys/molmodel-3.1.0/src/Pdb.cpp: In function 'gemmi::Structure
gemmiStructFromDoc(const gemmi::cif::Document&)':
/home/merkys/molmodel-3.1.0/src/Pdb.cpp:39:25: error:
'make_structure_from_block' is not a member of 'gemmi::impl'; did you
mean 'gemmi::make_structure_from_block'?
39 | return
gemmi::impl::make_structure_from_block(doc.blocks.front());
| ^~~~~~~~~~~~~~~~~~~~~~~~~
In file included from /usr/include/gemmi/mmread.hpp:13,
from
/home/merkys/molmodel-3.1.0/./include/molmodel/internal/Compound.h:48,
from
/home/merkys/molmodel-3.1.0/./include/molmodel/internal/Pdb.h:6,
from /home/merkys/molmodel-3.1.0/src/Pdb.cpp:4:
/usr/include/gemmi/mmcif.hpp:15:21: note:
'gemmi::make_structure_from_block' declared here
15 | GEMMI_DLL Structure make_structure_from_block(const cif::Block&
block);
| ^~~~~~~~~~~~~~~~~~~~~~~~~
The issue is better fixed upstream, I will forward the report. In the
meantime it is OK to remove molmodel from testing to not obstruct the
upcoming openmm transition.
Andrius
