Search Results

The Perl script will not work either because the numeric part of the header columns moved to the left of the underscore ("_") instead of the right.

So, the index to compare after the "split"...

[edit] I misunderstood the problem. You need the line sorted.

When you say "only first line", what do you mean? Clearly the second line is changed too.

Try also
read LINE < file
echo $LINE | tr '_;' ' '$'\n' | sort -k2n | tr ' ' '_' | awk -F";" '
FNR == NR {T[$0] = NR
MX = NR
next
}
FNR...

In that case:

awk '
function rbefore(STR) { return(substr(STR, 0, RSTART-1)); }# before match
function rall(STR) { return(substr(STR, RSTART, RLENGTH)); }# Entire match
function...

I guess s/he wants the first line sorted, and the columns of the residual lines rearranged to stay with their header.

gnu awk?
awk '
{
for (i in a) delete a[i];
for (i in b) delete b[i];
for (i=1; i<=NF; i++) {
v=$i; sub("^[^0-9]*", "", v);
v=sprintf("%020d", v);
a[v]=v; b[v]=$i;
...

I made assumption that you want longest value in field #4 and that you want to add numbers in fields #5, #6, #7:
awk -F\;...