1
  2
  3
  4
  5
  6
  7
  8
  9
 10
 11
 12
 13
 14
 15
 16
 17
 18
 19
 20
 21
 22
 23
 24
 25
 26
 27
 28
 29
 30
 31
 32
 33
 34
 35
 36
 37
 38
 39
 40
 41
 42
 43
 44
 45
 46
 47
 48
 49
 50
 51
 52
 53
 54
 55
 56
 57
 58
 59
 60
 61
 62
 63
 64
 65
 66
 67
 68
 69
 70
 71
 72
 73
 74
 75
 76
 77
 78
 79
 80
 81
 82
 83
 84
 85
 86
 87
 88
 89
 90
 91
 92
 93
 94
 95
 96
 97
 98
 99
100
#include "moab/ParallelData.hpp"
#include "moab/ParallelComm.hpp"
#include "MBParallelConventions.h"
#include "moab/Interface.hpp"

#include <map>

namespace moab
{

//! return partition sets; if tag_name is input, gets sets with
//! that tag name, otherwise uses PARALLEL_PARTITION tag
ErrorCode ParallelData::get_partition_sets( Range& part_sets, const char* tag_name )<--- The function 'get_partition_sets' is never used.
{
    Tag part_tag = 0;
    ErrorCode result;

    if( NULL != tag_name )
        result = mbImpl->tag_get_handle( tag_name, 1, MB_TYPE_INTEGER, part_tag );
    else
        result = mbImpl->tag_get_handle( PARALLEL_PARTITION_TAG_NAME, 1, MB_TYPE_INTEGER, part_tag );

    if( MB_SUCCESS != result )
        return result;
    else if( 0 == part_tag )
        return MB_TAG_NOT_FOUND;

    result = mbImpl->get_entities_by_type_and_tag( 0, MBENTITYSET, &part_tag, NULL, 1, part_sets, Interface::UNION );
    return result;
}

//! get communication interface sets and the processors with which
//! this processor communicates; sets are sorted by processor
ErrorCode ParallelData::get_interface_sets( std::vector< EntityHandle >& iface_sets, std::vector< int >& iface_procs )
{
#define CONTINUE             \
    {                        \
        result = tmp_result; \
        continue;            \
    }
    iface_sets.clear();
    iface_procs.clear();

    Tag proc_tag = 0, procs_tag = 0;
    ErrorCode result = MB_SUCCESS;
    int my_rank;
    if( parallelComm )
        my_rank = parallelComm->proc_config().proc_rank();
    else
        return MB_FAILURE;

    std::multimap< int, EntityHandle > iface_data;

    for( int i = 0; i < 2; i++ )
    {
        ErrorCode tmp_result;

        if( 0 == i )
            tmp_result = mbImpl->tag_get_handle( PARALLEL_SHARED_PROC_TAG_NAME, 1, MB_TYPE_INTEGER, proc_tag );
        else
            tmp_result =
                mbImpl->tag_get_handle( PARALLEL_SHARED_PROCS_TAG_NAME, MAX_SHARING_PROCS, MB_TYPE_INTEGER, proc_tag );
        if( MB_SUCCESS != tmp_result ) CONTINUE;

        int tsize;
        tmp_result = mbImpl->tag_get_length( proc_tag, tsize );
        if( 0 == tsize || MB_SUCCESS != tmp_result ) CONTINUE;

        Range proc_sets;
        tmp_result =
            mbImpl->get_entities_by_type_and_tag( 0, MBENTITYSET, &proc_tag, NULL, 1, proc_sets, Interface::UNION );
        if( MB_SUCCESS != tmp_result ) CONTINUE;

        if( proc_sets.empty() ) CONTINUE;

        std::vector< int > proc_tags( proc_sets.size() * tsize );
        tmp_result = mbImpl->tag_get_data( procs_tag, proc_sets, &proc_tags[0] );
        if( MB_SUCCESS != tmp_result ) CONTINUE;
        int k;
        Range::iterator rit;

        for( k = 0, rit = proc_sets.begin(); rit != proc_sets.end(); ++rit, k++ )
        {
            for( int j = 0; j < tsize; j++ )
            {
                if( my_rank != proc_tags[2 * k + j] && proc_tags[2 * k + j] >= 0 )
                    iface_data.insert( std::pair< int, EntityHandle >( proc_tags[2 * k + j], *rit ) );
            }
        }
    }

    // now get the results in sorted order
    std::multimap< int, EntityHandle >::iterator mit;
    for( mit = iface_data.begin(); mit != iface_data.end(); ++mit )
        iface_procs.push_back( ( *mit ).first ), iface_sets.push_back( ( *mit ).second );

    return result;
}

}  // namespace moab