Reading lines from standard input in Python and C++

makebigfile

#!/bin/sh

ECHO=/bin/echo

multiply_file()
{
  source="${1}"
  dest="${2}"
  count="${3}"
  update_frequency="${4}"
  
  while [ "${count}" -gt 0 ]
  do
    cat "${source}" >> "${dest}"
    
    count=$(( ${count} - 1 ))
    
    if [ "$(( ${count} % ${update_frequency} ))" -eq 0 ]
    then
      ${ECHO} -n "${count}..."
    fi
  done
  
  ${ECHO} "finished"
}

multiply_file readlines.cc bigfile1 1000 100
multiply_file bigfile1 bigfile 1000 100
rm bigfile1

makefile

all: readlines.clang readlines.gcc

readlines.clang: readlines.cc
	clang++ -O4 -stdlib=libc++ -o readlines.clang readlines.cc

readlines.gcc: readlines.cc
	g++ -O3 -o readlines.gcc readlines.cc

clean:
	rm -f readlines.clang readlines.gcc bigfile

test: all bigfile
	./readlines.py < bigfile
	./readlines.clang < bigfile
	./readlines.gcc < bigfile

bigfile: readlines.cc
	./makebigfile

.PHONY: clean test bigfile

Reading_lines_in_Python_and_C++.md

Reading Lines in Python and C++

This code snippet compares the speed of reading lines of text from standard input.

Sample output:

> make test  
clang++ -O4 -stdlib=libc++ -o readlines.clang readlines.cc  
g++ -O3 -o readlines.gcc readlines.cc  
./makebigfile  
900...800...700...600...500...400...300...200...100...0...finished  
900...800...700...600...500...400...300...200...100...0...finished  
./readlines.py < bigfile  
Read 29000000 lines in 7 seconds. LPS: 4142857  
./readlines.clang < bigfile  
Read 29000000 lines in 13 seconds. LPS: 2230769  
./readlines.gcc < bigfile  
Read 29000000 lines in 3 seconds. LPS: 9666666  
>

readlines.cc

#include <ctime>
#include <iostream>

using namespace std;

int main()
{
  cin.sync_with_stdio(false);
  
  char buffer[1048576];
  cin.rdbuf()->pubsetbuf(buffer, sizeof(buffer));
  
  int count(0);
  string line;
  time_t start = time(0);
  
  while (getline(cin, line))
    ++count;
  
  time_t seconds = time(0) - start;
  
  if (seconds > 0)
  {
    cout << "Read " << count << " lines in " << seconds << " seconds."
         << " LPS: " << count / seconds
         << endl;
  }
}

readlines.py

#!/usr/bin/env python

import time
import sys

count = 0
start_time = time.time()

for line in sys.stdin:
  count += 1

delta_sec = int(time.time() - start_time)

if delta_sec > 0:
  lines_per_sec = int(round(count/delta_sec))
  print("Read {0:n} lines in {1:n} seconds. LPS: {2:n}".format(count, delta_sec, lines_per_sec))