August 2010 - moab-dev - mailman3.cels.anl.gov

buildbot failure in Fathom on moab-par
by moab-dev＠mcs.anl.gov 12 Aug '10

12 Aug '10

The Fathomatic has detected a failed build of moab-par on Fathom. Full details are available at: http://gnep.mcs.anl.gov:8010/builders/moab-par/builds/242 Buildbot URL: http://gnep.mcs.anl.gov:8010/ Buildslave for this Build: gnep Build Reason: The Nightly scheduler named 'nightly' triggered this build Build Source Stamp: None Blamelist: BUILD FAILED: failed compile LOG EXCERPT: make[1]: Entering directory `/disk/itaps-buildslave/moab-par/build' cd . && /bin/sh /disk/itaps-buildslave/moab-par/build/config/missing --run automake-1.10 --foreign configure.ac:9: option `silent-rules' not recognized autom4te: closing standard output: Broken pipe sincerely, -The Fathomatic

1 0

buildbot failure in Fathom on moab-serial
by moab-dev＠mcs.anl.gov 12 Aug '10

12 Aug '10

The Fathomatic has detected a failed build of moab-serial on Fathom. Full details are available at: http://gnep.mcs.anl.gov:8010/builders/moab-serial/builds/241 Buildbot URL: http://gnep.mcs.anl.gov:8010/ Buildslave for this Build: gnep Build Reason: The Nightly scheduler named 'nightly' triggered this build Build Source Stamp: None Blamelist: BUILD FAILED: failed compile LOG EXCERPT: make[1]: Entering directory `/disk/itaps-buildslave/moab-serial/build' cd . && /bin/sh /disk/itaps-buildslave/moab-serial/build/config/missing --run automake-1.10 --foreign configure.ac:9: option `silent-rules' not recognized autom4te: closing standard output: Broken pipe sincerely, -The Fathomatic

1 0

r4084 - MOAB/trunk/src/io
by bmsmith6＠wisc.edu 11 Aug '10

11 Aug '10

Author: bmsmith Date: 2010-08-11 17:57:44 -0500 (Wed, 11 Aug 2010) New Revision: 4084 Modified: MOAB/trunk/src/io/ReadNCDF.cpp Log: Print cub nodes that cannot be matched. This is needed because the user likely did not create the geometry, but needs to debug it. Mesh files are too large to fluidly manipulate in Cubit. Modified: MOAB/trunk/src/io/ReadNCDF.cpp =================================================================== --- MOAB/trunk/src/io/ReadNCDF.cpp 2010-08-11 22:19:33 UTC (rev 4083) +++ MOAB/trunk/src/io/ReadNCDF.cpp 2010-08-11 22:57:44 UTC (rev 4084) @@ -1894,7 +1894,7 @@ for(int i=0; i<numberNodes_loading; ++i) { int exo_id = ptr[i]; CartVect exo_coords( orig_coords[0][i], orig_coords[1][i], orig_coords[2][i] ); - EntityHandle cub_vert = -1; + EntityHandle cub_vert = 0; bool found_match = false; // by id @@ -1933,7 +1933,7 @@ } } } - if(-1 != cub_vert) found_match = true; + if(0 != cub_vert) found_match = true; } // If a match is found, update it with the deformed coords from the exo file. @@ -1954,10 +1954,31 @@ average_magnitude += magnitude; } else { ++lost; - std::cout << "cannot match exo vert " << exo_id << " " << exo_coords << std::endl; + std::cout << "cannot match exo node " << exo_id << " " << exo_coords << std::endl; } } + // Exo verts that could not be matched have already been printed. Now print the + // cub verts that could not be matched. + if(matched_cub_vert_id_map.size() < cub_verts.size()) { + Range unmatched_cub_verts = cub_verts; + for(std::map<int,EntityHandle>::const_iterator i=matched_cub_vert_id_map.begin(); + i!=matched_cub_vert_id_map.end(); ++i) { + unmatched_cub_verts.erase( i->second ); + } + for(Range::const_iterator i=unmatched_cub_verts.begin(); i!=unmatched_cub_verts.end(); ++i) { + int cub_id; + rval = mdbImpl->tag_get_data( mGlobalIdTag, &(*i), 1, &cub_id ); + if(MB_SUCCESS != rval) return rval; + CartVect cub_coords; + rval = mdbImpl->get_coords( &(*i), 1, cub_coords.array() ); + if(MB_SUCCESS != rval) return rval; + std::cout << "cannot match cub node " << cub_id << " " << cub_coords << std::endl; + } + std::cout << " " << unmatched_cub_verts.size() + << " nodes from the cub file could not be matched." << std::endl; + }

1 0

r4083 - in MOAB/trunk: . examples itaps/igeom itaps/imesh src test
by kraftche＠cae.wisc.edu 11 Aug '10

11 Aug '10

Author: kraftche Date: 2010-08-11 17:19:33 -0500 (Wed, 11 Aug 2010) New Revision: 4083 Modified: MOAB/trunk/Makefile.am MOAB/trunk/configure.ac MOAB/trunk/examples/Makefile.am MOAB/trunk/itaps/igeom/Makefile.am MOAB/trunk/itaps/imesh/Makefile.am MOAB/trunk/src/Makefile.am MOAB/trunk/test/Makefile.am Log: Enable new features in automake 1.11: silent-rules : If user configures with --enable-silent-rules or builds with 'make V=0", then brief output (e.g. "CXX ReadHDF5.lo") will be printed rather the echoing the entire command (and for actual build commands, the corresponding real command generated by libtool.) This feature is experimental and may be non-portbable, but that shouldn't be a problem because it is disabled by default. Warnings are still printed. color-tests : use color output for test status if supported by temrinal parallel-tests : 'make check' will run tests in parallel (number of jobs indicated by the make '-j' flag), and test output (and stderr) will be redirected to ${testname}.log. I assume that this will not break builds with older versions of automake. There are no significant changes to anything other than the options list to AM_INIT_AUTOMAKE, and automake should ignore unrecognized options. Modified: MOAB/trunk/Makefile.am =================================================================== --- MOAB/trunk/Makefile.am 2010-08-11 19:44:12 UTC (rev 4082) +++ MOAB/trunk/Makefile.am 2010-08-11 22:19:33 UTC (rev 4083) @@ -41,8 +41,8 @@ # 'make prefix=/foo install', we don't know the correct install # directory until we're doing the install. install-data-hook: - echo "MOAB_LIBDIR=${libdir}" >> $(DESTDIR)$(cfgdir)/moab.make - echo "MOAB_INCLUDES=-I${includedir}" >> $(DESTDIR)$(cfgdir)/moab.make + $(AM_V_at)echo "MOAB_LIBDIR=${libdir}" >> $(DESTDIR)$(cfgdir)/moab.make + $(AM_V_at)echo "MOAB_INCLUDES=-I${includedir}" >> $(DESTDIR)$(cfgdir)/moab.make # Generate a file to be installed in $libdir containing the configuration @@ -53,7 +53,7 @@ # Second line: extract just the list of options # Third line: put individual options on separate lines moab.config: config.status Makefile.am - ./config.status --version | \ + $(AM_V_GEN)./config.status --version | \ sed -e 's/.*options "$.*$"/\1/p' -e 'd' | \ sed -e "s/'[[:space:]]'/'\n'/g" >$@ Modified: MOAB/trunk/configure.ac =================================================================== --- MOAB/trunk/configure.ac 2010-08-11 19:44:12 UTC (rev 4082) +++ MOAB/trunk/configure.ac 2010-08-11 22:19:33 UTC (rev 4083) @@ -6,7 +6,7 @@ AC_CONFIG_MACRO_DIR([config]) AC_CONFIG_AUX_DIR([config]) AC_CANONICAL_SYSTEM -AM_INIT_AUTOMAKE +AM_INIT_AUTOMAKE([silent-rules color-tests parallel-tests]) AC_CHECK_PROG( [SED], [sed], [sed], [true] ) Modified: MOAB/trunk/examples/Makefile.am =================================================================== --- MOAB/trunk/examples/Makefile.am 2010-08-11 19:44:12 UTC (rev 4082) +++ MOAB/trunk/examples/Makefile.am 2010-08-11 22:19:33 UTC (rev 4083) @@ -43,15 +43,15 @@ ex_make = $(DESTDIR)$(exampledir)/makefile install-data-hook: - rm -f $(ex_make) - mv $(DESTDIR)$(exampledir)/examples.make $(ex_make) - echo "all: $(check_PROGRAMS)" >>$(ex_make) - rule=' $$(CXX) -o $$@ $$< $$(CXXFLAGS) $$(MOAB_INCLUDES) $$(MOAB_LIBS_LINK)'; \

1 0

r4082 - MOAB/trunk/src/io
by bmsmith6＠wisc.edu 11 Aug '10

11 Aug '10

Author: bmsmith Date: 2010-08-11 14:44:12 -0500 (Wed, 11 Aug 2010) New Revision: 4082 Modified: MOAB/trunk/src/io/ReadNCDF.cpp Log: Rework confusing logic to match nodes If matching by id, use std::map If matching by proximity, use kdtree Modified: MOAB/trunk/src/io/ReadNCDF.cpp =================================================================== --- MOAB/trunk/src/io/ReadNCDF.cpp 2010-08-11 17:50:09 UTC (rev 4081) +++ MOAB/trunk/src/io/ReadNCDF.cpp 2010-08-11 19:44:12 UTC (rev 4082) @@ -1843,92 +1843,103 @@ if( strcmp (op, "set") && strcmp (op, " set")) return MB_NOT_IMPLEMENTED; - // Some accounting - int found = 0; - int lost = 0; + // Two methods of matching nodes (id vs. proximity) + const bool match_node_ids = true; - // Place cub verts in a kdtree. + // Get nodes in cubit file Range cub_verts; rval = mdbImpl->get_entities_by_type( cub_file_set, MBVERTEX, cub_verts ); if(MB_SUCCESS != rval) return rval; std::cout << " cub_file_set contains " << cub_verts.size() << " nodes." + << std::endl; + + // Some accounting + std::cout << " exodus file contains " << numberNodes_loading << " nodes." << std::endl; + double max_magnitude = 0; + double average_magnitude = 0; + int found = 0; + int lost = 0; + std::map<int,EntityHandle> cub_verts_id_map; AdaptiveKDTree kdtree( mdbImpl, true ); EntityHandle root; - AdaptiveKDTree::Settings settings; - settings.maxEntPerLeaf = 1; - settings.candidateSplitsPerDir = 1; - settings.candidatePlaneSet = AdaptiveKDTree::SUBDIVISION; - rval = kdtree.build_tree( cub_verts, root, &settings ); - if(MB_SUCCESS != rval) return rval; - AdaptiveKDTreeIter tree_iter; - rval = kdtree.get_tree_iterator( root, tree_iter ); - if(MB_SUCCESS != rval) return rval; - double max_magnitude = 0; - double average_magnitude = 0; - - // Find the matching cub node by id or proximity - const bool match_node_ids = true; - - // We should not use cub verts unless they have been matched. Place in a map + // Should not use cub verts unless they have been matched. Place in a map // for fast handle_by_id lookup.

1 0

r4072 - MOAB/trunk/tools/mbcoupler
by smithrm＠mcs.anl.gov 04 Aug '10

04 Aug '10

Author: smithrm Date: 2010-08-04 16:05:28 -0500 (Wed, 04 Aug 2010) New Revision: 4072 Added: MOAB/trunk/tools/mbcoupler/ssn_test.cpp Modified: MOAB/trunk/tools/mbcoupler/Coupler.cpp MOAB/trunk/tools/mbcoupler/Coupler.hpp MOAB/trunk/tools/mbcoupler/Makefile.am Log: Initial version of subset normalization. - Skeleton for full normalization function - Subset selection method done The method signatures will be updated before the final version is complete. Modified: MOAB/trunk/tools/mbcoupler/Coupler.cpp =================================================================== --- MOAB/trunk/tools/mbcoupler/Coupler.cpp 2010-08-02 16:25:12 UTC (rev 4071) +++ MOAB/trunk/tools/mbcoupler/Coupler.cpp 2010-08-04 21:05:28 UTC (rev 4072) @@ -4,6 +4,7 @@ #include "moab/GeomUtil.hpp" #include "ElemUtil.hpp" #include "moab/CN.hpp" +#include "iMesh_extensions.h" #include "iostream" extern "C" @@ -19,8 +20,13 @@ #include "assert.h" +#define ERROR(a) {if (iBase_SUCCESS != err) std::cerr << a << std::endl;} +#define ERRORR(a,b) {if (iBase_SUCCESS != err) {std::cerr << a << std::endl; return b;}} + namespace moab { +bool debug = false; + Coupler::Coupler(Interface *impl, ParallelComm *pc, Range &local_elems, @@ -90,7 +96,11 @@ result = myTree->get_tree_box(localRoot, &allBoxes[6*my_rank], &allBoxes[6*my_rank+3]); // now communicate to get all boxes - int mpi_err = MPI_Allgather(&allBoxes[6*my_rank], 6, MPI_DOUBLE, +// int mpi_err = MPI_Allgather(&allBoxes[6*my_rank], 6, MPI_DOUBLE, +// &allBoxes[0], 6, MPI_DOUBLE, +// myPc->proc_config().proc_comm()); + // Change to use "in place" option + int mpi_err = MPI_Allgather(MPI_IN_PLACE, 0, MPI_DATATYPE_NULL, &allBoxes[0], 6, MPI_DOUBLE, myPc->proc_config().proc_comm()); @@ -509,4 +519,449 @@ return MB_SUCCESS; } -} // namespace moab +// Normalize a field over the subset of entities identified by the tags and values passed +int Coupler::normalize_subset(iBase_EntitySetHandle &m1_root_set, + iBase_EntitySetHandle &m2_root_set, + const char *norm_tag, + const char **tag_names, + int num_tags, + const char **tag_values,

1 0

portability issue/question
by Paul Wilson 02 Aug '10

02 Aug '10

Hello, I am integrating MOAB/DAGMC into a c++ code that includes a header that defines index() as a macro, apparently through some combination of system headers. This creates a problem with including moab/Range.hpp because it attempts to define a member function index(). Fortunately it creates a compile time failure because the signature of the Range::index() is different from the macro, otherwise who knows what would happen??? I haven't bothered to track it down yet, but the short solution has been to add the following immediately prior to the include: #ifdef index #undef index #endif I have resisted committing this to the repo because it seems like a little presumptious/overbearing. Any other solutions? Paul -- Paul Wilson -- ------------------------------------------------------------------ -- Paul P.H. Wilson 419 Engineering Research Building wilsonp(a)engr.wisc.edu 1500 Engineering Dr ph/fax: 608/263-0807 Madison, WI 53706 My calendar: http://bit.ly/pphw-calendar Computational Nuclear Engineering Research Group http://cnerg.engr.wisc.edu Associate Professor, Nuclear Engineering Engineering Physics Department http://www.engr.wisc.edu/ep Chair, Energy Analysis & Policy Certificate Nelson Institute for Environmental Studies http://nelson.wisc.edu/eap/ Contributing to the joy and improvement of all those around me

3 3

r4071 - in MOAB/trunk: src/io tools/dagmc
by bmsmith6＠wisc.edu 02 Aug '10

02 Aug '10

Author: bmsmith Date: 2010-08-02 11:25:12 -0500 (Mon, 02 Aug 2010) New Revision: 4071 Modified: MOAB/trunk/src/io/ReadNCDF.cpp MOAB/trunk/tools/dagmc/cub2h5m.cc Log: -Summarize dead elements when reading an exodus file after updating node coordinates. -Enable cub2h5m to create new DAG-MCNP material groups with density altered so that mass is conserved through the deformation. Modified: MOAB/trunk/src/io/ReadNCDF.cpp =================================================================== --- MOAB/trunk/src/io/ReadNCDF.cpp 2010-07-27 21:53:46 UTC (rev 4070) +++ MOAB/trunk/src/io/ReadNCDF.cpp 2010-08-02 16:25:12 UTC (rev 4071) @@ -2053,6 +2053,8 @@ // For each block int first_elem_id_in_block = 0; int block_count = 1; // NetCDF variables start with 1 + int total_elems = 0; + int total_dead_elems = 0; for(std::vector<ReadBlockData>::iterator i=blocksLoading.begin(); i!=blocksLoading.end(); ++i) { @@ -2219,11 +2221,14 @@ // Print some statistics temp_ss.str(""); temp_ss << i->blockId; + total_dead_elems += dead_elem_counter; + total_elems += i->numElements; std::cout << " Block " << temp_ss.str() << " has " << dead_elem_counter << "/" << i->numElements << " dead elements." << std::endl; if(0 != missing_elem_counter) { std::cout << " " << missing_elem_counter - << " dead elements in this block were not found in the cub_file_set. " << std::endl; + << " dead elements in this block were not found in the cub_file_set. " + << std::endl; } // advance the pointers into element ids and block_count. memory cleanup. @@ -2233,6 +2238,9 @@ } + std::cout << " Total: " << total_dead_elems << "/" << total_elems + << " dead elements." << std::endl; + return MB_SUCCESS; } Modified: MOAB/trunk/tools/dagmc/cub2h5m.cc =================================================================== --- MOAB/trunk/tools/dagmc/cub2h5m.cc 2010-07-27 21:53:46 UTC (rev 4070) +++ MOAB/trunk/tools/dagmc/cub2h5m.cc 2010-08-02 16:25:12 UTC (rev 4071) @@ -10,6 +10,7 @@ #include "DagMC.hpp" #include <limits> #include <cstdlib> +#include <sstream> #include <time.h>

1 0

buildbot failure in Fathom on moab-distcheck
by moab-dev＠mcs.anl.gov 02 Aug '10

02 Aug '10

The Fathomatic has detected a failed build of moab-distcheck on Fathom. Full details are available at: http://gnep.mcs.anl.gov:8010/builders/moab-distcheck/builds/224 Buildbot URL: http://gnep.mcs.anl.gov:8010/ Buildslave for this Build: gnep Build Reason: The Nightly scheduler named 'nightly' triggered this build Build Source Stamp: None Blamelist: BUILD FAILED: failed svn LOG EXCERPT: svn: OPTIONS of 'https://svn.mcs.anl.gov/repos/ITAPS/MOAB/trunk': could not connect to server (https://svn.mcs.anl.gov) sincerely, -The Fathomatic

1 0

buildbot failure in Fathom on moab-doc
by moab-dev＠mcs.anl.gov 02 Aug '10

02 Aug '10

The Fathomatic has detected a failed build of moab-doc on Fathom. Full details are available at: http://gnep.mcs.anl.gov:8010/builders/moab-doc/builds/122 Buildbot URL: http://gnep.mcs.anl.gov:8010/ Buildslave for this Build: gnep Build Reason: The Nightly scheduler named 'nightly' triggered this build Build Source Stamp: None Blamelist: BUILD FAILED: failed svn LOG EXCERPT: svn: OPTIONS of 'https://svn.mcs.anl.gov/repos/ITAPS/MOAB/trunk': could not connect to server (https://svn.mcs.anl.gov) sincerely, -The Fathomatic

1 0