Replace a pattern-matching field with field's value from previous row - awk

Question

maybe the title is not a good description but, essentially, this is the problem I am trying to solve:
I have a text file with n rows and m space-separater fields in each row
if field j of row i matches a pattern, replace it with field j from row i - 1
I am not forced to use AWK (GAWK in this instance), but it seemed a good choice for this operation. This is the sript I wrote and it work as expected, but I was wondering if there is a more time-efficient way to solve the problem
{
if ($0!~/NoData/) {
split($0, data, " ");
print $0
} else {
split($0, row, " ", seps);
for(i in row) {if (row[i]~/NoData/) row[i]=data[i]; else data[i]=row[i]; printf "%s%s", row[i], seps[i];}
printf "\n"
}
}
As a sample, the script, running on this input file
0.8147 0.2785 0.9572 0.7922 0.6787 0.7060
0.9058 0.5469 0.4854 0.9595 0.7577 0.0318
0.1270 0.9575 0.8003 0.6557 0.7431 0.2769
0.9134 0.9649 NoData 0.0357 0.3922 0.0462
0.6324 0.1576 NoData NoData 0.6555 0.0971
0.0975 0.9706 NoData NoData 0.1712 0.8235
should produce this result
0.8147 0.2785 0.9572 0.7922 0.6787 0.7060
0.9058 0.5469 0.4854 0.9595 0.7577 0.0318
0.1270 0.9575 0.8003 0.6557 0.7431 0.2769
0.9134 0.9649 0.8003 0.0357 0.3922 0.0462
0.6324 0.1576 0.8003 0.0357 0.6555 0.0971
0.0975 0.9706 0.8003 0.0357 0.1712 0.8235

Answer 1

awk '{for(i=1;i<=NF;i++){ if($i~/NoData/){ $i=last[i]; } last[i]=$i } }1' file
If you want to preserve original formatting, you may use below, if you have gawk 4th argument for split may be utilized.
awk '{
split($0,D,/[^[:space:]]*/);
s = "";
for(i=1;i<=NF;i++){
if($i~/NoData/){ $i = last[i]; }
last[i]=$i ;
s = s sprintf("%s%s",D[i],$i)
}
print s
}' file
OR by setting OFS="" or OFS=
awk -v OFS= '{
split($0,D,/[^[:space:]]*/);
for(i=1;i<=NF;i++){
if($i~/NoData/){ $i = last[i]; }
last[i]=$i ;
$i = sprintf("%s%s",D[i],$i)
}
}1' file
Example - 1 ( Preserve Formatting )
$ cat file
0.8147 0.2785 0.9572 0.7922 0.6787 0.7060
0.9058 0.5469 0.4854 0.9595 0.7577 0.0318
0.1270 0.9575 0.8003 0.6557 0.7431 0.2769
0.9134 0.9649 NoData 0.0357 0.3922 0.0462
0.6324 0.1576 NoData NoData 0.6555 0.0971
0.0975 0.9706 NoData NoData 0.1712 0.8235
$ awk '{
split($0,D,/[^[:space:]]*/);
s = "";
for(i=1;i<=NF;i++){
if($i~/NoData/){ $i = last[i]; }
last[i]=$i ;
s = s sprintf("%s%s",D[i],$i)
}
print s
}' file
0.8147 0.2785 0.9572 0.7922 0.6787 0.7060
0.9058 0.5469 0.4854 0.9595 0.7577 0.0318
0.1270 0.9575 0.8003 0.6557 0.7431 0.2769
0.9134 0.9649 0.8003 0.0357 0.3922 0.0462
0.6324 0.1576 0.8003 0.0357 0.6555 0.0971
0.0975 0.9706 0.8003 0.0357 0.1712 0.8235
Example - 2 ( Without Preserving Source Formatting )
It takes single space as output separator by default, in case if you set OFS it will override default value.
$ cat file
0.8147 0.2785 0.9572 0.7922 0.6787 0.7060
0.9058 0.5469 0.4854 0.9595 0.7577 0.0318
0.1270 0.9575 0.8003 0.6557 0.7431 0.2769
0.9134 0.9649 NoData 0.0357 0.3922 0.0462
0.6324 0.1576 NoData NoData 0.6555 0.0971
0.0975 0.9706 NoData NoData 0.1712 0.8235
$ awk '{for(i=1;i<=NF;i++){ if($i~/NoData/){ $i=last[i]; } last[i]=$i } }1' file
0.8147 0.2785 0.9572 0.7922 0.6787 0.7060
0.9058 0.5469 0.4854 0.9595 0.7577 0.0318
0.1270 0.9575 0.8003 0.6557 0.7431 0.2769
0.9134 0.9649 0.8003 0.0357 0.3922 0.0462
0.6324 0.1576 0.8003 0.0357 0.6555 0.0971
0.0975 0.9706 0.8003 0.0357 0.1712 0.8235