#! /bin/bash

#Datasets="../think/cocomin_experiment_1-15-07.csv ../think/cocomin_experiment_horizon16.csv"
#Datasets="../think/cocomin_experiment_horizon16.csv"
#Datasets="../think/cocomin_compare1.csv"

Datapath="../think/separated_data/"
Datasets=$(ls $Datapath)

Subsets="coc81_all nasa93_all coc81_kind_min coc81_lang_ftn coc81_mode_e coc81_kind_max coc81_mode_org coc81_lang_mol nasa93_project_gro nasa93_cat2_missionplanning nasa93_cat2_avionicsmonitoring nasa93_mode_semidetached nasa93_project_sts nasa93_forg_g nasa93_center_5 nasa93_year_1975 nasa93_year_1980 nasa93_mode_embedded nasa93_center_2"

for Data in $Datasets
do

Data=$Datapath$Data

  DataName=$(basename $Data .csv)

  if [ -d $DataName ]; then
	echo -n ""
  else
	mkdir $DataName
cp aggregate_copy $DataName"/aggregate"
  fi

  for Subset in $Subsets
  do
	RawFile=$DataName"/"$Subset"_Raw.txt"
	RanksFile=$DataName"/"$Subset"_Ranks.txt"

	grep $Subset $Data | 
	  gawk 'BEGIN{FS=OFS=","}{print $3"_"$4"_"$5"_"$6"_"$7,$8,$9}' | 
	  ../../nonparametric/nonParametricAnalysisWMW RawResultsFileName=$RawFile RanksFileName=$RanksFile
  done
  
  cd $DataName
  ./aggregate
  cd ..
  
done