$ NUM_THREADS=1  ./rtoa-pgatram-allinone2b t1.txt t1.txt t1.txt t1.txt | cksum
do_it_all time    :    0.498 secs
737201628 75552000

$ NUM_THREADS=4  ./rtoa-pgatram-allinone2b t1.txt t1.txt t1.txt t1.txt | cksum
do_it_all time    :    0.176 secs
737201628 75552000

$ NUM_THREADS=8  ./rtoa-pgatram-allinone2b t1.txt t1.txt t1.txt t1.txt | cksum
do_it_all time    :    0.124 secs
737201628 75552000

$ NUM_THREADS=16 ./rtoa-pgatram-allinone2b t1.txt t1.txt t1.txt t1.txt | cksum
do_it_all time    :    0.096 secs
737201628 75552000

##</code><code>##

$ NUM_THREADS=1  ./rtoa-pgatram-allinone2c t1.txt t1.txt t1.txt t1.txt | cksum
do_it_all time    :    0.488 secs
737201628 75552000

$ NUM_THREADS=4  ./rtoa-pgatram-allinone2c t1.txt t1.txt t1.txt t1.txt | cksum
do_it_all time    :    0.143 secs
737201628 75552000

$ NUM_THREADS=8  ./rtoa-pgatram-allinone2c t1.txt t1.txt t1.txt t1.txt | cksum
do_it_all time    :    0.091 secs
737201628 75552000

$ NUM_THREADS=16 ./rtoa-pgatram-allinone2c t1.txt t1.txt t1.txt t1.txt | cksum
do_it_all time    :    0.065 secs
737201628 75552000

##</code><code>##

// rtoa-pgatram-allinone2c.cpp. Crude allinone version.
//   based on rtoa-pgatram-allinone2.cpp https://perlmonks.org/?node_id=11152186
//
// Obtain the fast_io library (required dependency):
//   git clone --depth=1 https://github.com/cppfastio/fast_io
//
// Obtain the Portable Memory Mapping C++ Class (required dependency):
//   git clone --depth=1 https://github.com/stbrumme/portable-memory-mapping
// Copy two files to your source directory: *.cpp and *.h
//   cp portable-memory-mapping/MemoryMapped.* .
//
// Compile with g++ or clang++:
//   clang++ -o rtoa-pgatram-allinone2c -std=c++20 -fopenmp -Wall -O3 -I fast_io/include MemoryMapped.cpp rtoa-pgatram-allinone2c.cpp
//
// OpenMP Little Book:
//   https://nanxiao.gitbooks.io/openmp-little-book/content/

#include <chrono>
#include <cstring>
#include <string>
#include <string_view>
#include <numeric>
#include <thread>

#ifdef _OPENMP
#include <omp.h>
#endif

#include <iostream>
#include <iomanip>

// See [id://11149504] for more info on the fast_io C++ library
#include <fast_io.h>
#include <fast_io_legacy.h>

#include "MemoryMapped.h"

// ---------------------------------------------------------------

typedef std::chrono::high_resolution_clock high_resolution_clock;
typedef std::chrono::high_resolution_clock::time_point time_point;
typedef std::chrono::milliseconds milliseconds;

double elaspe_time(
   time_point cend,
   time_point cstart)
{
   return double (
      std::chrono::duration_cast<milliseconds>(cend - cstart).count()
   ) * 1e-3;
}

// ---------------------------------------------------------------

// Though there are less than 256 initializers in this ascii table,
// the others are guaranteed by ANSI C to be initialized to zero.
static const int romtab[256] = {
    0,0,0,0,0,0,   0,   0,   0,   0,  //  00- 09
    0,0,0,0,0,0,   0,   0,   0,   0,  //  10- 19
    0,0,0,0,0,0,   0,   0,   0,   0,  //  20- 29
    0,0,0,0,0,0,   0,   0,   0,   0,  //  30- 39
    0,0,0,0,0,0,   0,   0,   0,   0,  //  40- 49
    0,0,0,0,0,0,   0,   0,   0,   0,  //  50- 59
    0,0,0,0,0,0,   0, 100, 500,   0,  //  60- 69
    0,0,0,1,0,0,  50,1000,   0,   0,  //  70- 79
    0,0,0,0,0,0,   5,   0,  10,   0,  //  80- 89
    0,0,0,0,0,0,   0,   0,   0, 100,  //  90- 99
  500,0,0,0,0,1,   0,   0,  50,1000,  // 100-109
    0,0,0,0,0,0,   0,   0,   5,   0,  // 110-119
   10,0,0,0,0,0,   0,   0,   0,   0   // 120-129
};

// Return the arabic number for a roman letter c.
// Return zero if the roman letter c is invalid.
inline int urtoa(int c) { return romtab[c]; }

inline int accfn(int t, char c) {
   return t + urtoa(c) - t % urtoa(c) * 2;
}

inline int roman_to_dec(std::string_view st) {
   return std::accumulate(st.begin(), st.end(), 0, accfn);
}

// ---------------------------------------------------------------

inline constexpr auto MIN_CHUNK_SIZE = 1048576;

// Helper funtion to get the next up value for integer division.
static std::size_t divide_up(std::size_t dividend, std::size_t divisor)
{
   if (dividend % divisor)
      return dividend / divisor + 1;
   else
      return dividend / divisor;
}

// Read an input file of Roman Numerals and do it all.
static void do_it_all(
   std::string_view fname,      //  in: file name containing a list of Roman Numerals
   int nthds                    //  in: number of threads
)
{
   // Map file to memory.
   MemoryMapped loader(fname.data(), MemoryMapped::WholeFile, MemoryMapped::SequentialScan);
   if (!loader.isValid()) {
      fast_io::io::perrln("Error opening '", fname, "' : ", std::errc{errno});
      return;
   }

   // Get raw pointer to mapped memory.
   char* data = (char*) loader.getData();
   std::size_t filesize = loader.size();

   // Loop contiguous file container. Each thread process max 1 chunk.
   std::size_t chunk_size = divide_up(filesize, nthds * 1);
   if (chunk_size < MIN_CHUNK_SIZE) chunk_size = MIN_CHUNK_SIZE;
   std::size_t num_chunks = divide_up(filesize, chunk_size);

   char const *end{data + filesize - 1};
   fast_io::out_buf_type obf{fast_io::out()};

   #pragma omp parallel for ordered schedule(static, 1)
   for (std::size_t chunk_id = 0; chunk_id < num_chunks; ++chunk_id) {
      char const *last{data + filesize - 1};
      char const *first{data};
      std::string buf;

      if (chunk_id < num_chunks - 1) {
         last = first + (chunk_size * (chunk_id + 1) - 1);
         if (*last != '\n') last = fast_io::find_lf(last, end);
      }
      if (chunk_id > 0) {
         first += (chunk_size * chunk_id - 1);
         if (*first != '\n') first = fast_io::find_lf(first, end);

         ++first;
      }

      while (first <= last) {
         auto start_ptr{first}; first = fast_io::find_lf(first, last);
         auto end_ptr{first};

         // chunk ids greater than 0 store locally for later output
         int dec = roman_to_dec(std::string_view(start_ptr, end_ptr - start_ptr));
         if (chunk_id)
            buf.append(fast_io::concatln(dec));
         else
            fast_io::io::println(obf, dec);

         ++first;
      }

      #pragma omp ordered
      {
         if (chunk_id)
            fast_io::io::print(obf, buf);
      }
   }
}

int main(int argc, char* argv[])
{
   if (argc < 2) {
      if (argc > 0)
         std::cerr << "usage: rtoa-pgatram-allinone2c file... >out.txt\n";
      return 1;
   }

   // Get the list of input files from the command line
   int    nfiles = argc - 1;
   char** fname  = &argv[1];

   std::cerr << std::setprecision(3) << std::setiosflags(std::ios::fixed);
   time_point cstartall, cendall;
   cstartall = high_resolution_clock::now();

#ifdef _OPENMP
   // Determine the number of threads.
   int nthds = std::thread::hardware_concurrency();
   const char* env_nthds1 = std::getenv("OMP_NUM_THREADS");
   const char* env_nthds2 = std::getenv("NUM_THREADS");

   if (env_nthds1 && strlen(env_nthds1))
      nthds = ::atoi(env_nthds1);
   else if (env_nthds2 && strlen(env_nthds2))
      nthds = ::atoi(env_nthds2);

   omp_set_dynamic(false);
   omp_set_num_threads(nthds);
#else
   int nthds = 1;
#endif

   for (int i = 0; i < nfiles; ++i)
      do_it_all(fname[i], nthds);

   cendall = high_resolution_clock::now();
   double ctakenall = elaspe_time(cendall, cstartall);
   std::cerr << "do_it_all time    : " << std::setw(8) << ctakenall << " secs\n";

   return 0;
}