scandir

#!/usr/bin/env bash

####################################################
# Specify your paths here:

from_data_path=/home/lykrysh/Data/march_2015/

from_data_dirtype=1 
#from_data_dirtype=2
#from_data_dirtype=3 

to_csv_path=${from_data_path::-1}_summary/
rundate=201503
filtered=no
num_gr_contaminated=0

####################################################
timestamp=$(date +"%Y%m%d")_$(date +"%H%M")

if [[ ${1} =~ --with-no-* ]]; then
	to_csv_filename=unfiltered_$timestamp
else
	to_csv_filename=filtered_$timestamp
fi

mkdir -p $to_csv_path
sqlite3 $to_csv_path$to_csv_filename.db < sql/mutation_table
sqlite3 $to_csv_path$to_csv_filename.db < sql/group_table
sqlite3 $to_csv_path$to_csv_filename.db < sql/uber_table
smplid=

function scandir() {
	for file in $(ls "$1" | sort -V)
	do
		if [[ -d ${1}/${file} ]]; then
			case "$2" in
			1)
				if [[ ${file} =~ [0-9]+_[0-9]+ ]]; then
					lane=${file##*_}
					mid=${file%_*}
					grps_raw=0
				fi
				;;
			2)
				if [[ ${file} =~ run[0-9]+ ]]; then
					lane=${file##*n}
				fi
				if [[ ${file} =~ [MID]?[0-9]+$ ]]; then
					mid=${file##*D}
					grps_raw=0
				fi
				;;
			3)
				if [[ ${file} =~ lane_[0-9]+ ]]; then
					lane=${file##*_}
				fi
				if [[ ${file} =~ ^[0-9]+$ ]]; then
					mid=${file}
					grps_raw=0
				fi
				;;
			*)	;;
			esac
			scandir "${1}/${file}" "$2"

		elif [[ ${1}/${file} =~ consgroup[0-9]+.fna ]]; then
			thisfile=${1}/${file}
			grpid=${thisfile##*p}
			grpid=${grpid%.*}
			grpconsensus=$(sed "6q;d" ${thisfile});

			if (( "$grpid" == 1 )); then
				ubrconsensus=$(sed "4q;d" ${thisfile});

				sqlite3 $to_csv_path$to_csv_filename.db << EOS
				insert into uber_table values (null, '$rundate', '$(date +"%Y%m%d")', $lane, $mid, '$ubrconsensus');
EOS

			fi

			sqlite3 $to_csv_path$to_csv_filename.db << EOS
			insert into group_table values (null, '$rundate', '$(date +"%Y%m%d")', $lane, $mid, $grpid, '$grpconsensus');
EOS

		elif [[ ${1}/${file} =~ group[0-9]+.fna ]]  && ! [[ ${1}/${file} =~ con* ]]; then
			count=$(echo $(grep '>' ${1}/${file} | wc -l ))
			if (( "$count" >= 5 )); then
				(( grps_raw++ ))
			fi

		elif [[ ${1}/${file} =~ output.txt ]]; then
			temp=${to_csv_path}temp_$timestamp
			echo $rundate > $temp
			echo $(date +"%Y%m%d") >> $temp
			echo $filtered >> $temp
			echo $lane >> $temp
			echo $mid >> $temp
			(( smplid++ ))
			./scan $to_csv_path$to_csv_filename.db ${1}/${file} $temp $grps_raw $num_gr_contaminated >> ${to_csv_path}groups_with_true_mut_subst_only_$timestamp $smplid
			rm -f $temp
		fi
	done
}

scandir $from_data_path $from_data_dirtype
sqlite3 -header -csv $to_csv_path$to_csv_filename.db "select * from mutation_table;" > ${to_csv_path}${to_csv_filename}.csv
sqlite3 -header -csv $to_csv_path$to_csv_filename.db "select * from group_table;" > ${to_csv_path}${to_csv_filename}_group.csv
sqlite3 -header -csv $to_csv_path$to_csv_filename.db "select * from uber_table;" > ${to_csv_path}${to_csv_filename}_uber.csv
#rm -f $to_csv_path$to_csv_filename.db
exit 0