Shell Programming and Scripting

#!/bin/bash

# sort -k 1,1 -k 4,4 -o infile impute.data
infil=infile
outfil=outfile
wrkfil=/tmp/impute
touch $wrkfil
touch $outfil
> $wrkfil
> $outfil
fmt="%-10s%-8s%-8s%-14s\n"
typ1=T1
typ2=T2

printf $fmt "Group" "Type" "Value" "Serial_Number" >> $wrkfil
do_dump ()
{
    t1lin=$(grep 'T1 ' $wrkfil)
    t2lin=$(grep ' T2' $wrkfil)
    t1xx="$mix1$mix1"
    t2xx="$mix2$mix2"

    # both T1 and T2 missing
    if [[ ! $t1lin && ! $t2lin ]]; then
        printf $fmt $grp $typ1 $t1xx "impute" >> $outfil
        printf $fmt $grp $typ2 $t2xx "impute" >> $outfil

    # only T2 missing
    elif [[ $t1lin && ! $t2lin ]]; then
        set $t1lin
        [[ ${3:0:1} == $mix2 ]] && t2xx="$mix1$mix1"
        add=$(printf $fmt $grp $typ2 $t2xx "impute")
        sed -i '/T1 /a\'"$add"'' $wrkfil

    # only T1 missing
    elif [[ ! $t1lin && $t2lin ]]; then
        set $t2lin
        [[ ${3:0:1} == $mix1 ]] && t1xx="$mix2$mix2"
        printf $fmt $grp $typ1 $t1xx "impute" >> $outfil
    fi

    # dump to output file
    cat $wrkfil >> $outfil
    > $wrkfil
}

while read -r col1 col2 col3 col4
do
    [[ ${col1:5:1} != [0-9] ]] && continue
    [ ! $grp ] && grp=$col1
    if [[ $col1 == $grp ]]; then
        if [[ ${col3:0:1} != ${col3:1:1} ]]; then
            mix1="${col3:0:1}"
            mix2="${col3:1:1}"
        fi
    else
        do_dump
        grp=$col1
    fi
    printf $fmt $col1 $col2 $col3 $col4 >> $wrkfil
done < $infil
do_dump
cat $outfil

# eof #

# output
# ------
# Group     Type    Value   Serial_Number
# Group1    T1      aa      1             
# Group1    T2      tt      2             
# Group1    T1*T2   tt      3             
# Group1    T1*T2   tt      4             
# Group1    T1*T2   at      5             
# Group4    T1      aa      1             
# Group4    T2      gg      impute        
# Group4    T1*T2   gg      3             
# Group4    T1*T2   aa      4             
# Group4    T1*T2   ag      5             
# Group5    T1      gg      1             
# Group5    T2      tt      impute        
# Group5    T1*T2   gg      3             
# Group5    T1*T2   gt      4             
# Group5    T1*T2   gg      5             
# Group6    T1      aa      impute        
# Group6    T2      gg      2             
# Group6    T1*T2   ag      3             
# Group6    T1*T2   gg      4             
# Group6    T1*T2   ag      5             
# Group6    T1*T2   gg      6             
# Group7    T1      gg      impute        
# Group7    T2      aa      impute        
# Group7    T1*T2   ga      7             
# Group7    T1*T2   gg      8             
# Group7    T1*T2   ga      9             

# real    0m0.034s
# user    0m0.009s
# sys 0m0.016s