benchmarks/csv/csv_bench.sh

   1 #!/bin/bash
   2 # This file is part of NIT ( http://www.nitlanguage.org ).
   3 #
   4 # Licensed under the Apache License, Version 2.0 (the "License");
   5 # you may not use this file except in compliance with the License.
   6 # You may obtain a copy of the License at
   7 #
   8 #     http://www.apache.org/licenses/LICENSE-2.0
   9 #
  10 # Unless required by applicable law or agreed to in writing, software
  11 # distributed under the License is distributed on an "AS IS" BASIS,
  12 # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
  13 # See the License for the specific language governing permissions and
  14 # limitations under the License.
  15
  16 # Shell script to bench json parsers over different documents
  17
  18 source ../bench_common.sh
  19 source ../bench_plot.sh
  20
  21 ## CONFIGURATION OPTIONS ##
  22
  23 # Default number of times a command must be run with bench_command
  24 # Can be overrided with 'the option -n'
  25 count=5
  26
  27 ## HANDLE OPTIONS ##
  28
  29 function init_repo()
  30 {
  31         mkdir -p inputs
  32         nitc --semi-global scripts/csv_gen.nit -o scripts/csv_gen
  33         echo "Generating 1000 lines documents"
  34         ./scripts/csv_gen 10 1000 inputs/1000_l.csv
  35         ./scripts/csv_gen 10 1000 inputs/1000_uni_l.csv --unicode
  36         echo "Generating 10000 lines documents"
  37         ./scripts/csv_gen 10 10000 inputs/10000_l.csv
  38         ./scripts/csv_gen 10 10000 inputs/10000_uni_l.csv --unicode
  39         echo "Generating 100000 lines documents"
  40         ./scripts/csv_gen 10 100000 inputs/100000_l.csv
  41         ./scripts/csv_gen 10 100000 inputs/100000_uni_l.csv --unicode
  42         echo "Generating 1000000 lines documents"
  43         ./scripts/csv_gen 10 1000000 inputs/1000000_l.csv
  44         ./scripts/csv_gen 10 1000000 inputs/1000000_uni_l.csv --unicode
  45 }
  46
  47 function usage()
  48 {
  49         echo "run_bench: ./csv_bench.sh [options]"
  50         echo "  -v: verbose mode"
  51         echo "  -n count: number of execution for each bar (default: $count)"
  52         echo "  -h: this help"
  53 }
  54
  55 stop=false
  56 fast=false
  57 while [ "$stop" = false ]; do
  58         case "$1" in
  59                 -v) verbose=true; shift;;
  60                 --fast) fast=true; shift;;
  61                 -h) usage; exit;;
  62                 -n) count="$2"; shift; shift;;
  63                 *) stop=true
  64         esac
  65 done
  66
  67 if [ -z "$fast" ]; then
  68         init_repo
  69 fi
  70
  71 mkdir -p out
  72
  73 echo "Compiling engines"
  74
  75 echo "Java Parser"
  76
  77 javac -cp './scripts/commons-csv-1.3.jar' scripts/JavaCSV.java
  78
  79 echo "Go parser"
  80
  81 go build -o scripts/go_csv scripts/go_csv.go
  82
  83 echo "Nit/Ad-Hoc Parser"
  84
  85 nitc --semi-global scripts/nit_csv.nit -o scripts/nit_csv
  86
  87 declare -a script_names=('Python 3 - Pandas' 'Python 2 - Pandas' 'Go' 'Nit' 'Python 3 - Standard' 'Python 2 - Standard' 'Java - Apache commons' 'Ruby')
  88 declare -a script_cmds=('python3 scripts/python_csv.py' 'python2 scripts/python_csv.py' './scripts/go_csv' './scripts/nit_csv' 'python3 scripts/python_stdcsv.py' 'python2 scripts/python_stdcsv.py' "java -cp /usr/share/java/commons-csv.jar:. scripts.JavaCSV" 'ruby scripts/ruby_csv.rb')
  89
  90 for script in `seq 1 ${#script_cmds[@]}`; do
  91         echo "Preparing res for ${script_names[$script - 1]}"
  92         prepare_res "./out/${script_names[$script - 1]}.dat" "${script_names[$script - 1]}" "${script_names[$script - 1]}"
  93         for file in inputs/*.csv; do
  94                 fname=`basename $file .csv`
  95                 bench_command $file "Benching file $file using ${script_cmds[$script - 1]} parser" ${script_cmds[$script - 1]} $file
  96         done;
  97 done;
  98
  99 rm scripts/nit_csv
 100 rm scripts/JavaCSV.class
 101 rm scripts/go_csv
 102
 103 plot out/bench_csv.gnu