Hi,
I've just checked out this:
svn co http://svn.salilab.org/imp/branches/stable imp .... Checked out revision 8381.
and tried to compile it, I am using the same compiler that compiles the imp 1.0 version so I don't think it is a compiler problem
Install file: "modules/cgal/include/internal/knn.h" as "build/include/IMP/cgal/internal/knn.h" g++ -o modules/cgal/src/internal/knn.os -c -Os -Wno-long-double -no-cpp-precomp -mno-fused-madd -fno-common -Wall -Wno-deprecated -Woverloaded-virtual -O2 -fno-finite-math-only -ftrapping-math -fsignaling-nans -fno-float-store -frounding-math -fno-strict-aliasing -DCGAL_DISABLE_ROUNDING_MATH_CHECK -fvisibility=hidden -fPIC -DIMPCGAL_EXPORTS -DGCC_VISIBILITY -Ibuild/include -I/sw/include modules/cgal/src/internal/knn.cpp modules/cgal/src/internal/knn.cpp: In member function 'void IMP::cgal::internal::KNNData2::fill_nearest_neighbors_v(const IMP::algebra::VectorD<2u>&, unsigned int, double, IMP::Ints&) const': modules/cgal/src/internal/knn.cpp:122: warning: passing negative value '-0x00000000000000001' for argument 1 to 'IMP::cgal::internal::VectorWithIndex<D>::VectorWithIndex(unsigned int, const IMP::algebra::VectorD<D>&) [with unsigned int D = 2u]' modules/cgal/src/internal/knn.cpp: In member function 'void IMP::cgal::internal::KNNData2::fill_nearest_neighbors_v(const IMP::algebra::VectorD<2u>&, double, double, IMP::Ints&) const': modules/cgal/src/internal/knn.cpp:122: warning: passing negative value '-0x00000000000000001' for argument 1 to 'IMP::cgal::internal::VectorWithIndex<D>::VectorWithIndex(unsigned int, const IMP::algebra::VectorD<D>&) [with unsigned int D = 2u]' modules/cgal/src/internal/knn.cpp: In member function 'void IMP::cgal::internal::KNNData3::fill_nearest_neighbors_v(const IMP::algebra::VectorD<3u>&, unsigned int, double, IMP::Ints&) const': modules/cgal/src/internal/knn.cpp:123: warning: passing negative value '-0x00000000000000001' for argument 1 to 'IMP::cgal::internal::VectorWithIndex<D>::VectorWithIndex(unsigned int, const IMP::algebra::VectorD<D>&) [with unsigned int D = 3u]' modules/cgal/src/internal/knn.cpp: In member function 'void IMP::cgal::internal::KNNData3::fill_nearest_neighbors_v(const IMP::algebra::VectorD<3u>&, double, double, IMP::Ints&) const': modules/cgal/src/internal/knn.cpp:123: warning: passing negative value '-0x00000000000000001' for argument 1 to 'IMP::cgal::internal::VectorWithIndex<D>::VectorWithIndex(unsigned int, const IMP::algebra::VectorD<D>&) [with unsigned int D = 3u]' modules/cgal/src/internal/knn.cpp: In member function 'void IMP::cgal::internal::KNNData4::fill_nearest_neighbors_v(const IMP::algebra::VectorD<4u>&, unsigned int, double, IMP::Ints&) const': modules/cgal/src/internal/knn.cpp:124: warning: passing negative value '-0x00000000000000001' for argument 1 to 'IMP::cgal::internal::VectorWithIndex<D>::VectorWithIndex(unsigned int, const IMP::algebra::VectorD<D>&) [with unsigned int D = 4u]' modules/cgal/src/internal/knn.cpp: In member function 'void IMP::cgal::internal::KNNData4::fill_nearest_neighbors_v(const IMP::algebra::VectorD<4u>&, double, double, IMP::Ints&) const': modules/cgal/src/internal/knn.cpp:124: warning: passing negative value '-0x00000000000000001' for argument 1 to 'IMP::cgal::internal::VectorWithIndex<D>::VectorWithIndex(unsigned int, const IMP::algebra::VectorD<D>&) [with unsigned int D = 4u]' Install file: "modules/cgal/include/internal/polygons.h" as "build/include/IMP/cgal/internal/polygons.h" g++ -o modules/cgal/src/internal/polygons.os -c -Os -Wno-long-double -no-cpp-precomp -mno-fused-madd -fno-common -Wall -Wno-deprecated -Woverloaded-virtual -O2 -fno-finite-math-only -ftrapping-math -fsignaling-nans -fno-float-store -frounding-math -fno-strict-aliasing -DCGAL_DISABLE_ROUNDING_MATH_CHECK -fvisibility=hidden -fPIC -DIMPCGAL_EXPORTS -DGCC_VISIBILITY -Ibuild/include -I/sw/include modules/cgal/src/internal/polygons.cpp Install file: "modules/cgal/include/internal/polyhedrons.h" as "build/include/IMP/cgal/internal/polyhedrons.h" Install file: "modules/algebra/include/Plane3D.h" as "build/include/IMP/algebra/Plane3D.h" g++ -o modules/cgal/src/internal/polyhedrons.os -c -Os -Wno-long-double -no-cpp-precomp -mno-fused-madd -fno-common -Wall -Wno-deprecated -Woverloaded-virtual -O2 -fno-finite-math-only -ftrapping-math -fsignaling-nans -fno-float-store -frounding-math -fno-strict-aliasing -DCGAL_DISABLE_ROUNDING_MATH_CHECK -fvisibility=hidden -fPIC -DIMPCGAL_EXPORTS -DGCC_VISIBILITY -Ibuild/include -I/sw/include modules/cgal/src/internal/polyhedrons.cpp modules/cgal/src/internal/polyhedrons.cpp: In function 'std::vector<std::vector<IMP::algebra::VectorD<3u>, std::allocator<IMP::algebra::VectorD<3u> > >, std::allocator<std::vector<IMP::algebra::VectorD<3u>, std::allocator<IMP::algebra::VectorD<3u> > > > > IMP::cgal::internal::<unnamed>::get_facets(CGAL::Nef_polyhedron_3<K, typename CGAL::Default_items<Kernel>::Items, bool>&)': modules/cgal/src/internal/polyhedrons.cpp:99: internal compiler error: Bus error Please submit a full bug report, with preprocessed source if appropriate. See URL:http://developer.apple.com/bugreporter for instructions. scons: *** [modules/cgal/src/internal/polyhedrons.os] Error 1 scons: building terminated because of errors.
Enabled dependencies: python, NetCDF, FFTW3, CGAL, Boost.FileSystem, Boost.ProgramOptions Disabled/unfound optional dependencies: ANN, ExampleLib, GSL, OpenCV, Bullet, OpenMM
Enabled modules: kernel, cgal, algebra, core, container, atom, display, domino, em, example, misc, saxs, helper, statistics, restrainer, domino2, multifit, benchmark Disabled modules: gsl, modeller, em2d, bullet, openmm Errors building: modules/cgal/src/internal/polyhedrons.os
Maybe it has nothing to do with your problem, but I remarked sometimes the latest svn version does not compile. In these cases, I go back to the latest imp version I could compile (either with "svn update -r…" or more drastically with a full src reinstall "svn co -r"). For me, the v7943 compiled successfully on various linux machines as well as on my macbook. you should also want to use the -k option to scons in order to prevent it to crash and stop when one module cannot compile.
Hope this helps
--Ben
Le 21 janv. 2011 à 15:26, Robert Brynmor Fenwick a écrit :
> > Hi, > > I've just checked out this: > > svn co http://svn.salilab.org/imp/branches/stable imp > .... > Checked out revision 8381. > > and tried to compile it, I am using the same compiler that compiles the imp 1.0 version so I don't think it is a compiler problem > > Install file: "modules/cgal/include/internal/knn.h" as "build/include/IMP/cgal/internal/knn.h" > g++ -o modules/cgal/src/internal/knn.os -c -Os -Wno-long-double -no-cpp-precomp -mno-fused-madd -fno-common -Wall -Wno-deprecated -Woverloaded-virtual -O2 -fno-finite-math-only -ftrapping-math -fsignaling-nans -fno-float-store -frounding-math -fno-strict-aliasing -DCGAL_DISABLE_ROUNDING_MATH_CHECK -fvisibility=hidden -fPIC -DIMPCGAL_EXPORTS -DGCC_VISIBILITY -Ibuild/include -I/sw/include modules/cgal/src/internal/knn.cpp > modules/cgal/src/internal/knn.cpp: In member function 'void IMP::cgal::internal::KNNData2::fill_nearest_neighbors_v(const IMP::algebra::VectorD<2u>&, unsigned int, double, IMP::Ints&) const': > modules/cgal/src/internal/knn.cpp:122: warning: passing negative value '-0x00000000000000001' for argument 1 to 'IMP::cgal::internal::VectorWithIndex<D>::VectorWithIndex(unsigned int, const IMP::algebra::VectorD<D>&) [with unsigned int D = 2u]' > modules/cgal/src/internal/knn.cpp: In member function 'void IMP::cgal::internal::KNNData2::fill_nearest_neighbors_v(const IMP::algebra::VectorD<2u>&, double, double, IMP::Ints&) const': > modules/cgal/src/internal/knn.cpp:122: warning: passing negative value '-0x00000000000000001' for argument 1 to 'IMP::cgal::internal::VectorWithIndex<D>::VectorWithIndex(unsigned int, const IMP::algebra::VectorD<D>&) [with unsigned int D = 2u]' > modules/cgal/src/internal/knn.cpp: In member function 'void IMP::cgal::internal::KNNData3::fill_nearest_neighbors_v(const IMP::algebra::VectorD<3u>&, unsigned int, double, IMP::Ints&) const': > modules/cgal/src/internal/knn.cpp:123: warning: passing negative value '-0x00000000000000001' for argument 1 to 'IMP::cgal::internal::VectorWithIndex<D>::VectorWithIndex(unsigned int, const IMP::algebra::VectorD<D>&) [with unsigned int D = 3u]' > modules/cgal/src/internal/knn.cpp: In member function 'void IMP::cgal::internal::KNNData3::fill_nearest_neighbors_v(const IMP::algebra::VectorD<3u>&, double, double, IMP::Ints&) const': > modules/cgal/src/internal/knn.cpp:123: warning: passing negative value '-0x00000000000000001' for argument 1 to 'IMP::cgal::internal::VectorWithIndex<D>::VectorWithIndex(unsigned int, const IMP::algebra::VectorD<D>&) [with unsigned int D = 3u]' > modules/cgal/src/internal/knn.cpp: In member function 'void IMP::cgal::internal::KNNData4::fill_nearest_neighbors_v(const IMP::algebra::VectorD<4u>&, unsigned int, double, IMP::Ints&) const': > modules/cgal/src/internal/knn.cpp:124: warning: passing negative value '-0x00000000000000001' for argument 1 to 'IMP::cgal::internal::VectorWithIndex<D>::VectorWithIndex(unsigned int, const IMP::algebra::VectorD<D>&) [with unsigned int D = 4u]' > modules/cgal/src/internal/knn.cpp: In member function 'void IMP::cgal::internal::KNNData4::fill_nearest_neighbors_v(const IMP::algebra::VectorD<4u>&, double, double, IMP::Ints&) const': > modules/cgal/src/internal/knn.cpp:124: warning: passing negative value '-0x00000000000000001' for argument 1 to 'IMP::cgal::internal::VectorWithIndex<D>::VectorWithIndex(unsigned int, const IMP::algebra::VectorD<D>&) [with unsigned int D = 4u]' > Install file: "modules/cgal/include/internal/polygons.h" as "build/include/IMP/cgal/internal/polygons.h" > g++ -o modules/cgal/src/internal/polygons.os -c -Os -Wno-long-double -no-cpp-precomp -mno-fused-madd -fno-common -Wall -Wno-deprecated -Woverloaded-virtual -O2 -fno-finite-math-only -ftrapping-math -fsignaling-nans -fno-float-store -frounding-math -fno-strict-aliasing -DCGAL_DISABLE_ROUNDING_MATH_CHECK -fvisibility=hidden -fPIC -DIMPCGAL_EXPORTS -DGCC_VISIBILITY -Ibuild/include -I/sw/include modules/cgal/src/internal/polygons.cpp > Install file: "modules/cgal/include/internal/polyhedrons.h" as "build/include/IMP/cgal/internal/polyhedrons.h" > Install file: "modules/algebra/include/Plane3D.h" as "build/include/IMP/algebra/Plane3D.h" > g++ -o modules/cgal/src/internal/polyhedrons.os -c -Os -Wno-long-double -no-cpp-precomp -mno-fused-madd -fno-common -Wall -Wno-deprecated -Woverloaded-virtual -O2 -fno-finite-math-only -ftrapping-math -fsignaling-nans -fno-float-store -frounding-math -fno-strict-aliasing -DCGAL_DISABLE_ROUNDING_MATH_CHECK -fvisibility=hidden -fPIC -DIMPCGAL_EXPORTS -DGCC_VISIBILITY -Ibuild/include -I/sw/include modules/cgal/src/internal/polyhedrons.cpp > modules/cgal/src/internal/polyhedrons.cpp: In function 'std::vector<std::vector<IMP::algebra::VectorD<3u>, std::allocator<IMP::algebra::VectorD<3u> > >, std::allocator<std::vector<IMP::algebra::VectorD<3u>, std::allocator<IMP::algebra::VectorD<3u> > > > > IMP::cgal::internal::<unnamed>::get_facets(CGAL::Nef_polyhedron_3<K, typename CGAL::Default_items<Kernel>::Items, bool>&)': > modules/cgal/src/internal/polyhedrons.cpp:99: internal compiler error: Bus error > Please submit a full bug report, > with preprocessed source if appropriate. > See URL:http://developer.apple.com/bugreporter for instructions. > scons: *** [modules/cgal/src/internal/polyhedrons.os] Error 1 > scons: building terminated because of errors. > > > Enabled dependencies: python, NetCDF, FFTW3, CGAL, Boost.FileSystem, Boost.ProgramOptions > Disabled/unfound optional dependencies: ANN, ExampleLib, GSL, OpenCV, Bullet, OpenMM > > Enabled modules: kernel, cgal, algebra, core, container, atom, display, domino, em, example, misc, saxs, helper, statistics, restrainer, domino2, multifit, benchmark > Disabled modules: gsl, modeller, em2d, bullet, openmm > Errors building: > modules/cgal/src/internal/polyhedrons.os > > -- > R. Bryn Fenwick > robert.fenwick@irbbarcelona.org > Post-doctoral fellow > Chemistry and Molecular Pharmacology Programme > Institute for Research in Biomedicine (IRB Barcelona) > Parc Científic de Barcelona > Baldiri Reixac 10, 08028 Barcelona, SPAIN > Tel. (+34) 9340 20460 > _______________________________________________ > IMP-users mailing list > IMP-users@salilab.org > https://salilab.org/mailman/listinfo/imp-users
Dr. Benjamin SCHWARZ Biocomputing group Email : schwarz@igbmc.fr Voice : +33 (0)3 68 85 47 30 FAX : +33 (0)3 68 85 47 18
Structural Biology & Genomics Dept. - IGBMC 1 rue Laurent Fries BP 10142 F - 67404 Illkirch CEDEX
On 1/21/11 7:05 AM, Benjamin SCHWARZ wrote: > Maybe it has nothing to do with your problem, but I remarked sometimes > the latest svn version does not compile.
Robert is using the stable branch, which compiles on all of our test machines and passes all of the unit tests. It is, of course, possible that your setup or configuration differs enough from ours that you will detect a bug that we don't cover in our tests though.
>> modules/cgal/src/internal/polyhedrons.cpp:99: internal compiler error: >> Bus error >> Please submit a full bug report, >> with preprocessed source if appropriate.
An internal compiler error is always a problem with the compiler, not the code. I suggest you report this to Apple as suggested. You may be able to work around it by making minor changes to that C++ file though (e.g. removing or modifying the function where the crash occurs, reducing the compiler optimization level, etc.) Alternatively, if you don't use CGAL, you could simply disable the IMP.cgal module.
Ben
Can you give me a hint on how to disable the cgal module, I am not sure I understand how to do this.
Bryn
On 1/21/11 8:16 AM, R. Bryn Fenwick wrote: > Can you give me a hint on how to disable the cgal module, I am not sure > I understand how to do this.
Add "cgal=no" or "disabledmodules=cgal" to your scons invocation. (The first disables the detection of the external CGAL library, which the IMP.cgal module needs to compile; the second disables the IMP.cgal module itself.)
Ben
As an aside,
I have noticed that even if I run the command scons -c the installation directories are not cleaned. I have recently discovered that one set of code that I was running was functioning via links that were generated by a previous version. This had the affect of artificially giving me a functioning code. I am sure that the nightly builds are made in a clean directory, however other users might run into a similar problem as I was previously was.
Bryn
On Jan 21, 2011, at 8:22 AM, R. Bryn Fenwick wrote:
> > As an aside, > > I have noticed that even if I run the command scons -c the installation directories are not cleaned. I have recently discovered that one set of code that I was running was functioning via links that were generated by a previous version. This had the affect of artificially giving me a functioning code. I am sure that the nightly builds are made in a clean directory, however other users might run into a similar problem as I was previously was.
With most build tools (scons, make etc), clean just cleans up the build directories rather than uninstalls the package. The normal pattern (when installing) is something like scons # put it somewhere easy to access, perhaps, using sudo to put it somewhere not normally writeable (sudo) scons install # clean up the intermediate files to save disk space scons -c all
We have tried to be careful so that an installed IMP setup will not get mixed in with a local build when running tests and using imppy.sh (assuming that the build doesn't return an error). How did you get the problem? And with what versions if IMP? Thanks.
Daniel
> > Bryn > > -- > R. Bryn Fenwick > Sent with Sparrow > On Friday, 21 January 2011 at 17:16, R. Bryn Fenwick wrote: > >> >> Can you give me a hint on how to disable the cgal module, I am not sure I understand how to do this. >> >> Bryn >> >> -- >> R. Bryn Fenwick >> Sent with Sparrow >> On Friday, 21 January 2011 at 17:13, Ben Webb wrote: >> >>> don't use CGAL, you could simply disable the IMP.cgal module. >> > > _______________________________________________ > IMP-users mailing list > IMP-users@salilab.org > https://salilab.org/mailman/listinfo/imp-users
My guess is that you checked things in between my checking in patches last night and found IMP in an inconsistent state. Do you still get this behavior on the head version? eg "svn co http://svn.salilab.org/imp/trunk imp" Thanks. Daniel
On Jan 21, 2011, at 6:26 AM, Robert Brynmor Fenwick wrote:
> > Hi, > > I've just checked out this: > > svn co http://svn.salilab.org/imp/branches/stable imp > .... > Checked out revision 8381. > > and tried to compile it, I am using the same compiler that compiles the imp 1.0 version so I don't think it is a compiler problem > > Install file: "modules/cgal/include/internal/knn.h" as "build/include/IMP/cgal/internal/knn.h" > g++ -o modules/cgal/src/internal/knn.os -c -Os -Wno-long-double -no-cpp-precomp -mno-fused-madd -fno-common -Wall -Wno-deprecated -Woverloaded-virtual -O2 -fno-finite-math-only -ftrapping-math -fsignaling-nans -fno-float-store -frounding-math -fno-strict-aliasing -DCGAL_DISABLE_ROUNDING_MATH_CHECK -fvisibility=hidden -fPIC -DIMPCGAL_EXPORTS -DGCC_VISIBILITY -Ibuild/include -I/sw/include modules/cgal/src/internal/knn.cpp > modules/cgal/src/internal/knn.cpp: In member function 'void IMP::cgal::internal::KNNData2::fill_nearest_neighbors_v(const IMP::algebra::VectorD<2u>&, unsigned int, double, IMP::Ints&) const': > modules/cgal/src/internal/knn.cpp:122: warning: passing negative value '-0x00000000000000001' for argument 1 to 'IMP::cgal::internal::VectorWithIndex<D>::VectorWithIndex(unsigned int, const IMP::algebra::VectorD<D>&) [with unsigned int D = 2u]' > modules/cgal/src/internal/knn.cpp: In member function 'void IMP::cgal::internal::KNNData2::fill_nearest_neighbors_v(const IMP::algebra::VectorD<2u>&, double, double, IMP::Ints&) const': > modules/cgal/src/internal/knn.cpp:122: warning: passing negative value '-0x00000000000000001' for argument 1 to 'IMP::cgal::internal::VectorWithIndex<D>::VectorWithIndex(unsigned int, const IMP::algebra::VectorD<D>&) [with unsigned int D = 2u]' > modules/cgal/src/internal/knn.cpp: In member function 'void IMP::cgal::internal::KNNData3::fill_nearest_neighbors_v(const IMP::algebra::VectorD<3u>&, unsigned int, double, IMP::Ints&) const': > modules/cgal/src/internal/knn.cpp:123: warning: passing negative value '-0x00000000000000001' for argument 1 to 'IMP::cgal::internal::VectorWithIndex<D>::VectorWithIndex(unsigned int, const IMP::algebra::VectorD<D>&) [with unsigned int D = 3u]' > modules/cgal/src/internal/knn.cpp: In member function 'void IMP::cgal::internal::KNNData3::fill_nearest_neighbors_v(const IMP::algebra::VectorD<3u>&, double, double, IMP::Ints&) const': > modules/cgal/src/internal/knn.cpp:123: warning: passing negative value '-0x00000000000000001' for argument 1 to 'IMP::cgal::internal::VectorWithIndex<D>::VectorWithIndex(unsigned int, const IMP::algebra::VectorD<D>&) [with unsigned int D = 3u]' > modules/cgal/src/internal/knn.cpp: In member function 'void IMP::cgal::internal::KNNData4::fill_nearest_neighbors_v(const IMP::algebra::VectorD<4u>&, unsigned int, double, IMP::Ints&) const': > modules/cgal/src/internal/knn.cpp:124: warning: passing negative value '-0x00000000000000001' for argument 1 to 'IMP::cgal::internal::VectorWithIndex<D>::VectorWithIndex(unsigned int, const IMP::algebra::VectorD<D>&) [with unsigned int D = 4u]' > modules/cgal/src/internal/knn.cpp: In member function 'void IMP::cgal::internal::KNNData4::fill_nearest_neighbors_v(const IMP::algebra::VectorD<4u>&, double, double, IMP::Ints&) const': > modules/cgal/src/internal/knn.cpp:124: warning: passing negative value '-0x00000000000000001' for argument 1 to 'IMP::cgal::internal::VectorWithIndex<D>::VectorWithIndex(unsigned int, const IMP::algebra::VectorD<D>&) [with unsigned int D = 4u]' > Install file: "modules/cgal/include/internal/polygons.h" as "build/include/IMP/cgal/internal/polygons.h" > g++ -o modules/cgal/src/internal/polygons.os -c -Os -Wno-long-double -no-cpp-precomp -mno-fused-madd -fno-common -Wall -Wno-deprecated -Woverloaded-virtual -O2 -fno-finite-math-only -ftrapping-math -fsignaling-nans -fno-float-store -frounding-math -fno-strict-aliasing -DCGAL_DISABLE_ROUNDING_MATH_CHECK -fvisibility=hidden -fPIC -DIMPCGAL_EXPORTS -DGCC_VISIBILITY -Ibuild/include -I/sw/include modules/cgal/src/internal/polygons.cpp > Install file: "modules/cgal/include/internal/polyhedrons.h" as "build/include/IMP/cgal/internal/polyhedrons.h" > Install file: "modules/algebra/include/Plane3D.h" as "build/include/IMP/algebra/Plane3D.h" > g++ -o modules/cgal/src/internal/polyhedrons.os -c -Os -Wno-long-double -no-cpp-precomp -mno-fused-madd -fno-common -Wall -Wno-deprecated -Woverloaded-virtual -O2 -fno-finite-math-only -ftrapping-math -fsignaling-nans -fno-float-store -frounding-math -fno-strict-aliasing -DCGAL_DISABLE_ROUNDING_MATH_CHECK -fvisibility=hidden -fPIC -DIMPCGAL_EXPORTS -DGCC_VISIBILITY -Ibuild/include -I/sw/include modules/cgal/src/internal/polyhedrons.cpp > modules/cgal/src/internal/polyhedrons.cpp: In function 'std::vector<std::vector<IMP::algebra::VectorD<3u>, std::allocator<IMP::algebra::VectorD<3u> > >, std::allocator<std::vector<IMP::algebra::VectorD<3u>, std::allocator<IMP::algebra::VectorD<3u> > > > > IMP::cgal::internal::<unnamed>::get_facets(CGAL::Nef_polyhedron_3<K, typename CGAL::Default_items<Kernel>::Items, bool>&)': > modules/cgal/src/internal/polyhedrons.cpp:99: internal compiler error: Bus error > Please submit a full bug report, > with preprocessed source if appropriate. > See URL:http://developer.apple.com/bugreporter for instructions. > scons: *** [modules/cgal/src/internal/polyhedrons.os] Error 1 > scons: building terminated because of errors. > > > Enabled dependencies: python, NetCDF, FFTW3, CGAL, Boost.FileSystem, Boost.ProgramOptions > Disabled/unfound optional dependencies: ANN, ExampleLib, GSL, OpenCV, Bullet, OpenMM > > Enabled modules: kernel, cgal, algebra, core, container, atom, display, domino, em, example, misc, saxs, helper, statistics, restrainer, domino2, multifit, benchmark > Disabled modules: gsl, modeller, em2d, bullet, openmm > Errors building: > modules/cgal/src/internal/polyhedrons.os > > -- > R. Bryn Fenwick > robert.fenwick@irbbarcelona.org > Post-doctoral fellow > Chemistry and Molecular Pharmacology Programme > Institute for Research in Biomedicine (IRB Barcelona) > Parc Científic de Barcelona > Baldiri Reixac 10, 08028 Barcelona, SPAIN > Tel. (+34) 9340 20460 > _______________________________________________ > IMP-users mailing list > IMP-users@salilab.org > https://salilab.org/mailman/listinfo/imp-users
On 1/21/11 9:35 AM, Daniel Russel wrote: > My guess is that you checked things in between my checking in patches > last night and found IMP in an inconsistent state.
No, this is the stable branch. It's from the last time the nightly build succeeded, which was in late October. It won't be in an inconsistent state.
Ben
OK, I have managed to get the code to compile by removing the use of the cgal, I have however found another error that seems to be a problem with the saxs module:
Traceback (most recent call last): File "foxs_sax_fitting_6.py", line 39, in <module> IMP.core.XYZR(particles[i]).set_radius(radius) File "/Users/fenwick/opt/imp_stable/lib/python2.5/site-packages/IMP/core/__init__.py", line 579, in set_radius def set_radius(self, *args): return _IMP_core.XYZR_set_radius(self, *args) _IMP.UsageException: Cannot set value "radius" from particle Atom N as it is not there.
On 21 January 2011 18:40, Ben Webb ben@salilab.org wrote:
> On 1/21/11 9:35 AM, Daniel Russel wrote: > > My guess is that you checked things in between my checking in patches > > last night and found IMP in an inconsistent state. > > No, this is the stable branch. It's from the last time the nightly build > succeeded, which was in late October. It won't be in an inconsistent state. > > Ben > -- > ben@salilab.org http://salilab.org/~ben/ > "It is a capital mistake to theorize before one has data." > - Sir Arthur Conan Doyle > _______________________________________________ > IMP-users mailing list > IMP-users@salilab.org > https://salilab.org/mailman/listinfo/imp-users >
I would recommend updating to svn head (check out svn.salilab.org/imp/trunk). We probably should remove mentions of the stable branch from the manual (or figure out a better scheme for updating it). --Daniel
On Jan 21, 2011, at 10:51 AM, Robert Brynmor Fenwick wrote:
> > OK, I have managed to get the code to compile by removing the use of the cgal, I have however found another error that seems to be a problem with the saxs module: > > Traceback (most recent call last): > File "foxs_sax_fitting_6.py", line 39, in <module> > IMP.core.XYZR(particles[i]).set_radius(radius) > File "/Users/fenwick/opt/imp_stable/lib/python2.5/site-packages/IMP/core/__init__.py", line 579, in set_radius > def set_radius(self, *args): return _IMP_core.XYZR_set_radius(self, *args) > _IMP.UsageException: Cannot set value "radius" from particle Atom N as it is not there. > > > On 21 January 2011 18:40, Ben Webb ben@salilab.org wrote: > On 1/21/11 9:35 AM, Daniel Russel wrote: > > My guess is that you checked things in between my checking in patches > > last night and found IMP in an inconsistent state. > > No, this is the stable branch. It's from the last time the nightly build > succeeded, which was in late October. It won't be in an inconsistent state. > > Ben > -- > ben@salilab.org http://salilab.org/~ben/ > "It is a capital mistake to theorize before one has data." > - Sir Arthur Conan Doyle > _______________________________________________ > IMP-users mailing list > IMP-users@salilab.org > https://salilab.org/mailman/listinfo/imp-users > > > > -- > R. Bryn Fenwick > robert.fenwick@irbbarcelona.org > Post-doctoral fellow > Chemistry and Molecular Pharmacology Programme > Institute for Research in Biomedicine (IRB Barcelona) > Parc Científic de Barcelona > Baldiri Reixac 10, 08028 Barcelona, SPAIN > Tel. (+34) 9340 20460 > _______________________________________________ > IMP-users mailing list > IMP-users@salilab.org > https://salilab.org/mailman/listinfo/imp-users
Ok so in theory the trunk should always be stable?
Bryn
Sent from my iPod
On 21/01/2011, at 21:04, Daniel Russel drussel@gmail.com wrote:
I would recommend updating to svn head (check out svn.salilab.org/imp/trunk). We probably should remove mentions of the stable branch from the manual (or figure out a better scheme for updating it). --Daniel
On Jan 21, 2011, at 10:51 AM, Robert Brynmor Fenwick wrote:
OK, I have managed to get the code to compile by removing the use of the cgal, I have however found another error that seems to be a problem with the saxs module:
Traceback (most recent call last): File "foxs_sax_fitting_6.py", line 39, in <module> IMP.core.XYZR(particles[i]).set_radius(radius) File "/Users/fenwick/opt/imp_stable/lib/python2.5/site-packages/IMP/core/__init__.py", line 579, in set_radius def set_radius(self, *args): return _IMP_core.XYZR_set_radius(self, *args) _IMP.UsageException: Cannot set value "radius" from particle Atom N as it is not there.
On 21 January 2011 18:40, Ben Webb ben@salilab.org wrote:
> On 1/21/11 9:35 AM, Daniel Russel wrote: > > My guess is that you checked things in between my checking in patches > > last night and found IMP in an inconsistent state. > > No, this is the stable branch. It's from the last time the nightly build > succeeded, which was in late October. It won't be in an inconsistent state. > > Ben > -- > ben@salilab.org http://salilab.org/~ben/ > "It is a capital mistake to theorize before one has data." > - Sir Arthur Conan Doyle > _______________________________________________ > IMP-users mailing list > IMP-users@salilab.org > https://salilab.org/mailman/listinfo/imp-users >
On 1/22/11 3:00 AM, Robert Brynmor Fenwick wrote: > Ok so in theory the trunk should always be stable?
In practice this is not the case, no. See the nightly builds page to see what is broken: http://salilab.org/imp/nightly/tests.html
For example, today there are no red boxes on that page (indicating something that failed to compile) but there are quite a few orange boxes (indicating failed unit tests). But if you're not using one of those modules, or the functionality you want isn't one of the failing unit tests, you should be OK. But living on the cutting edge is always risky!
Ben
Yes ok i understand. I want to use the python interface which I think is quite fragile. For that reason I was trying to compile the stable branch. Is it known when 1.1 will be released?
Sent from my iPod
On 22/01/2011, at 18:15, Ben Webb ben@salilab.org wrote:
> On 1/22/11 3:00 AM, Robert Brynmor Fenwick wrote: >> Ok so in theory the trunk should always be stable? > > In practice this is not the case, no. See the nightly builds page to see > what is broken: > http://salilab.org/imp/nightly/tests.html > > For example, today there are no red boxes on that page (indicating > something that failed to compile) but there are quite a few orange boxes > (indicating failed unit tests). But if you're not using one of those > modules, or the functionality you want isn't one of the failing unit > tests, you should be OK. But living on the cutting edge is always risky! > > Ben > -- > ben@salilab.org http://salilab.org/~ben/ > "It is a capital mistake to theorize before one has data." > - Sir Arthur Conan Doyle > _______________________________________________ > IMP-users mailing list > IMP-users@salilab.org > https://salilab.org/mailman/listinfo/imp-users
On 1/23/11 12:18 AM, Robert Brynmor Fenwick wrote: > Yes ok i understand. I want to use the python interface which I think > is quite fragile.
No, the Python interface should be just as stable as the C++ code. It is an integral part of our build and test procedure, so if it breaks, we fix it ASAP.
> Is it known when 1.1 will be released?
Likely to be within the next few months, dependent on the full integration of the MultiFit module.
Ben
OK I have returned to compiling the trunk and it seems that everything is working.
Thank you to Ben, Benjamin and David for all the help.
Bryn
On 23 January 2011 09:42, Ben Webb ben@salilab.org wrote:
> On 1/23/11 12:18 AM, Robert Brynmor Fenwick wrote: > > Yes ok i understand. I want to use the python interface which I think > > is quite fragile. > > No, the Python interface should be just as stable as the C++ code. It is > an integral part of our build and test procedure, so if it breaks, we > fix it ASAP. > > > Is it known when 1.1 will be released? > > Likely to be within the next few months, dependent on the full > integration of the MultiFit module. > > Ben > -- > ben@salilab.org http://salilab.org/~ben/ > "It is a capital mistake to theorize before one has data." > - Sir Arthur Conan Doyle > _______________________________________________ > IMP-users mailing list > IMP-users@salilab.org > https://salilab.org/mailman/listinfo/imp-users >
-
Ben Webb -
Benjamin SCHWARZ -
Daniel Russel -
R. Bryn Fenwick -
Robert Brynmor Fenwick