VectorIO.cpp

Go to the documentation of this file.

#include <mpi.h>
#include <sys/time.h> 
#include <iostream>
#include <functional>
#include <algorithm>
#include <vector>
#include <sstream>
#include "CombBLAS/CombBLAS.h"

using namespace std;
using namespace combblas;



template <class IT, class NT>
class PairReadSaveHandler
{
    public:
        PairReadSaveHandler() {};
        pair<NT,NT> getNoNum(IT index) { return make_pair((NT) 0, (NT) 0); }

        
        template <typename c, typename t>
        pair<NT,NT> read(std::basic_istream<c,t>& is, IT index)
        {
            pair<NT,NT> pp;
            is >> pp.first >> pp.second;
            return pp;
        }
        
    
        template <typename c, typename t>
        void save(std::basic_ostream<c,t>& os, const pair<NT,NT> & pp, IT index)    
        {
            os << pp.first << "\t" << pp.second;
        }
};


int main(int argc, char* argv[])
{
    int nprocs, myrank;
    MPI_Init(&argc, &argv);
    MPI_Comm_size(MPI_COMM_WORLD,&nprocs);
    MPI_Comm_rank(MPI_COMM_WORLD,&myrank);

    if(argc < 5)
    {
        if(myrank == 0)
        {
            cout << "Usage: ./VectorIO <DictionaryVector> <UnPermutedVector> <OutputName> base(0 or 1)" << endl;
            cout << "Example: ./VectorIO vertex_dict.mtx clusters.mtx original_clusters.mtx 0" << endl;
        }
        MPI_Finalize(); 
        return -1;
    }               
    {
        string vec1name(argv[1]);
        string vec2name(argv[2]);
        string voutname(argv[3]);
        int base = atoi(argv[4]);


        FullyDistVec<int64_t,string> vecdict, clusters; // keep cluster ids as "string" for generality
    
        // the "index sets" of vecdict and clusters are the same
        // goal here is to just convert the indices in "clusters" to values in vecdict 
            
        vecdict.ParallelRead(vec1name, base, // keeps lexicographically larger one in case of duplicates (after warning)
                [](string s1, string s2) { cout << "Unexpected duplicate in dictionary" << endl; return std::max<string>(s1, s2); });

        clusters.ParallelRead(vec2name, base, // keeps lexicographically larger one in case of duplicates (after warning)
                [](string s1, string s2) { cout << "Unexpected duplicate in unpermuted vector" << endl; return std::max<string>(s1, s2); });    
        
        vecdict.PrintInfo("dictionary");
            clusters.PrintInfo("unpermuted cluster vector");
              
        // FullyDistVec<IT,NT>::EWiseOut(const FullyDistVec<IT,NT> & rhs, _BinaryOperation __binary_op, FullyDistVec<IT,OUT> & result)
        // Perform __binary_op(*this[i], rhs[i]) for every element in rhs and *this, write the result output vector
        FullyDistVec<int64_t, pair<string, string> > newclusters;
        
        clusters.EWiseOut(vecdict, [] (string s1, string s2) { return make_pair(s1,s2); },  newclusters);
       
            newclusters.ParallelWrite(voutname, base, PairReadSaveHandler<int64_t, string>(), false);   // don't print the index        
        }
    MPI_Finalize();
    return 0;
}