Bachelors_Thesis_Code/porat-porat.cpp

/* #define NDEBUG */

#include <stdlib.h>
#include <string>
#include <vector>
#include <iostream>
#include <map>
#include <thread>
#include <math.h>

#include <string_view>

#include "hash_function_library.hpp"

// Initialization of constants
int p = 7919; // the prime for our hash function
int r = 11; // random int in \in F_p
char T[] = "abcabcabcdabc"; // text
/* char P[] = "abcabc"; // pattern */
char P[] = "abcabcabcabcabcabcddddddddddddddddddabc"; // pattern
std::vector<long> prehashed_values;
std::vector<int> shortest_periods;
std::map<int, int> prehashed_indices;

// https://stackoverflow.com/questions/18620942/find-the-smallest-period-of-input-string-in-on
// BEGIN stolen code
std::vector<int> calculateLPS(char * pat, int m) {
    /* int[] lps = new int[pat.length()]; */
    int len = 0;
    int i = 1;
    std::vector<int> lps = {0};
    lps.resize(m);

    while (i < m) {
        if (pat[i] == pat[len]) {
            len++;
            lps[i] = len;
            i++;
        }
        else {
            if (len != 0) {
                len = lps[len - 1];
            }
            else {
                lps[i] = len;
                i++;
            }
        }
    }
    return lps;
}

// calculates the length of the shortest period
int len_of_shortest_period (char * pattern, int m) {
    std::vector<int> lps = calculateLPS(pattern, m);
    //start at the end of the string
    int i = lps.size()-1;
    while (lps[i] != 0) {
        //shift back
        i -= lps[i];
    }
    return i+1;
}
// END

class porat_process {
    // TODO: use a different hash function. This one is BAD
    public:
        // we use the polynomial fingerprint
        void increment_hash (char c) {
            prev_pow = prev_pow*r % p;
            hash = (hash + c*prev_pow) % p;
            l++;
        }

        void subtract_hash (long pre_fingerprint, int i) {
            // i is the number of removed elements
            // pre_fingerprint is the fingerprint of those previous elements
            hash = (hash - pre_fingerprint)/(long)pow(r, i); // we are guaranteed that integer division will return a whole number
            prev_pow /= (long)pow(r, i);
            l -= i;
            // TODO: untested, especially prev_pow
        }

        bool should_spawn_child() {
            if (l == next_i_squared) {
                next_i_squared <<= 1;
                return true;
            }
            else
                return false;
        }

        long get_fingerprint() {
            return hash;
        }

    private:
        long prev_pow = 1;
        int l = 0;
        int next_i_squared = 1;
        long hash = 0;
};

void print_map(std::string_view comment, const std::map<int, int>& m)
{
    std::cout << comment;
    for (const auto& [key, value] : m) {
        std::cout << key << " = " << value << "; ";
    }
    std::cout << "\n";
}

void print_vector(std::string_view comment, const std::vector<int>& m)
{
    std::cout << comment << "[";
    for (const auto& a : m) {
        std::cout << a << ", ";
    }
    std::cout << "]\n";
}

void print_vector_long(std::string_view comment, const std::vector<long>& m)
{
    std::cout << comment << "[";
    for (const auto& a : m) {
        std::cout << a << ", ";
    }
    std::cout << "]\n";
}

int main() {
    int n = sizeof(T)/sizeof(char) - 1;
    int m = sizeof(P)/sizeof(char) - 1;

    {
        int i = 1;
        while (i < m) {
            // calculate shortest period length
            int period = len_of_shortest_period(P, i);
            shortest_periods.push_back(period);

            // calculate fingerprint of period
            if (!prehashed_indices.contains(period)) {
                porat_process process;
                for (int ii = 0; ii < period; ii++){
                    std::cout << P[ii];
                    process.increment_hash(P[ii]);
                }
                /* prehashed_indices[period] = process.get_fingerprint(); */
                prehashed_values.push_back(process.get_fingerprint());
                std::cout << period << " " << prehashed_values.size() << std::endl;
                prehashed_indices[period] = prehashed_values.size()-1;
            }

            i <<= 1;
        }
        if (i != m) { // so i>m, which means we skipped exactly m
            // calculate shortest period length
            int period = len_of_shortest_period(P, m);
            shortest_periods.push_back(period);

            // calculate fingerprint of period
            if (!prehashed_indices.contains(period)) {
                porat_process process;
                std::cout << "[";
                for (int ii = 0; ii < period; ii++) {
                    std::cout << P[ii];
                    process.increment_hash(P[ii]);
                }
                std::cout << "]\n";
                prehashed_values.push_back(process.get_fingerprint());
                std::cout << period << " " << prehashed_values.size() << std::endl;
                prehashed_indices[period] = prehashed_values.size()-1;
            }
            /* // calculate fingerprint of phi(P_{2^i}) */
            /* while (ii < m) { */
            /*     process.increment_hash(P[ii]); */
            /*     ii++; */
            /* } */
            /* prehashed_values.push_back(process.get_fingerprint()); */
        }
    }

    {
        std::cout << P << std::endl;
        int i = 0;
        while ((1 << i) < m) {
            std::cout << "pattern: ";
            for (int ii = 0; ii < (1 << i); ii++)
                std::cout << P[ii];
            std::cout << std::endl;
            std::cout << "period:  ";
            for (int ii = 0; ii < shortest_periods[i]; ii++)
                std::cout << P[ii];
            std::cout << std::endl;
            std::cout << "|prefix_{P_" << (1 << i) << "}| = " << shortest_periods[i] << std::endl;
            std::cout << prehashed_values[prehashed_indices[shortest_periods[i]]] << std::endl;
            i++;
        }
        if ((1 << i) != m) { // so i>m, which means we skipped exactly m
            std::cout << "pattern: ";
            for (int ii = 0; ii < m; ii++)
                std::cout << P[ii];
            std::cout << std::endl;
            std::cout << "period:  ";
            for (int ii = 0; ii < shortest_periods[i]; ii++)
                std::cout << P[ii];
            std::cout << std::endl;
            std::cout << "|prefix_{P_" << m << "}| = " << shortest_periods[i] << std::endl;
            std::cout << prehashed_values[prehashed_indices[shortest_periods[i]]] << std::endl;
            std::cout << prehashed_values[0] << std::endl;
            std::cout << prehashed_values[1] << std::endl;
            std::cout << prehashed_values[2] << std::endl;
            std::cout << prehashed_values[3] << std::endl;
            std::cout << prehashed_values[4] << std::endl;
        }
    }

    print_map("Indices map: ", prehashed_indices);
    print_vector_long("Values vector: ", prehashed_values);
    print_vector("Periods vector: ", shortest_periods);

    return EXIT_SUCCESS;
}