AWK - Select lines to print according to score - awk

Question

I have a tab-separated file containing a series of lemmas with associated scores.
The file contains 5 columns, the first column is the lemma and the third is the one that contains the score. What I need to do is print the line as it is, when lemma is not repeated and print the line with the highest score when lemma is repeated.
IN
Lemma --- Score --- ---
cserép 06a 55 6 bueno
darázs 05 38 1 bueno
dél 06a 34 1 bueno
dér 06a 29 1 bueno
díj 05 14 89 malo
díj 06a 2 101 malo
díj 06b 2 101 malo
díj 07 90 13 bueno
díj 08a 2 101 malo
díj 08b 2 101 malo
egér 06a 66 5 bueno
fonal 05 12 1 bueno
fonal 07 52 4 bueno
Desired output
Lemma --- Score --- ---
cserép 06a 55 6 bueno
darázs 05 38 1 bueno
dél 06a 34 1 bueno
dér 06a 29 1 bueno
díj 07 90 13 bueno
egér 06a 66 5 bueno
fonal 07 52 4 malo
What I have done. But it only works when the lemma is repeated once.
BEGIN {
OFS=FS="\t";
flag="";
}
{
id=$1;
if (id != flag)
{
if (line != "")
{
sub("^;","",line);
z=split(line,A,";");
if ((A[3] > A[8]) && (A[8] != ""))
{
print A[1]"\t"A[2]"\t"A[3]"\t"A[4]"\t"A[5];
}
else if ((A[8] > A[3]) && (A[8] != ""))
{
print A[6]"\t"A[7]"\t"A[8]"\t"A[9]"\t"A[10]
}
else
{
print A[1]"\t"A[2]"\t"A[3]"\t"A[4]"\t"A[5];
}
}
delete line;
flag=id;
}
line[$1]=line[$1]";"$2";"$3";"$4";"$5;
}
END {
line=line ";"$1";"$2";"$3";"$4";"$5
sub("^;","",line);
z=split(line,A,";");
if ((A[3] > A[8]) && (A[8] != ""))
{
print A[1]"\t"A[2]"\t"A[3]"\t"A[4]"\t"A[5];
}
else if ((A[8] > A[3]) && (A[8] != ""))
{
print A[6]"\t"A[7]"\t"A[8]"\t"A[9]"\t"A[10]
}
else
{
print A[1]"\t"A[2]"\t"A[3]"\t"A[4]"\t"A[5]
}
}

Answer 1

This one doesn't require the file to be sorted by lemma, but, it keeps all the lines to be printed in memory (one for each lemma) so may not be appropriate for a file with millions of different lemmas.
It also does not respect the order of the original file.
Finally, it assumes that all scores are non-negative!
$ cat lemma.awk
BEGIN { FS = OFS = "\t" }
NR == 1 { print }
NR > 1 {
if ($3 > score[$1]) {
score[$1] = $3
line[$1] = $0
}
}
END { for (lemma in line) print line[lemma] }
$ awk -f lemma.awk lemma.txt
Lemma --- Score --- ---
cserép 06a 55 6 bueno
díj 07 90 13 bueno
fonal 07 52 4 bueno
darázs 05 38 1 bueno
egér 06a 66 5 bueno
dél 06a 34 1 bueno
dér 06a 29 1 bueno

Answer 2

Tested with gnu awk:
prevLemma != $1 {
if( prevLemma ) {
print line;
}
prevLemma = $1;
prevScore = $3;
line = $0;
}
prevLemma == $1 { if( prevScore < $3 ) {
prevScore = $3;
line = $0;
}
}
END { print line;}
assumption is: the file is sorted by lemma
when the lemma changes (or at the very beginning when the var is empty) the lemma, score and line are saved
when the lemma changes (or in the END), the line for the previous lemma is printed
when the current line belongs to the same lemma and has a higher score the values are saved again

Answer 3

$ cat tst.awk
$1 != prev { printf "%s", maxLine; maxLine=""; max=$3; prev=$1 }
$3 >= max { max=$3; maxLine=$0 ORS }
END { printf "%s", maxLine }
$ awk -f tst.awk file
Lemma --- Score --- ---
cserép 06a 55 6 bueno
darázs 05 38 1 bueno
dél 06a 34 1 bueno
dér 06a 29 1 bueno
díj 07 90 13 bueno
egér 06a 66 5 bueno
fonal 07 52 4 bueno

Answer 4

Use a script:
if ($1 != $5) print $0
else
{
score($NR) = $3
print $0
}
Actually , this might be better done with perl.