rearrange a file

Shell Programming and Scripting
1

hi!
in awk, i have a file like this:

Trace1: WRIT,Trace2: BLAN,Trace3: BLAN,
-47.2120018005371,,,39815.4809027778
-46.3009986877441,,,39815.4809027778
-46.277000427246,,,39815.4809143519
-46.7389984130859,,,39815.4809259259
-46.3460006713867,,,39815.4809259259
-45.9309997558593,,,39815.4809375
-46.9119987487792,,,39815.4809490741
-46.5379981994628,,,39815.4809606482
-46.9640007019042,,,39815.4809722222
-46.8390007019042,,,39815.4809722222
-47.1399993896484,,,39815.4809837962
-46.1370010375976,,,39815.4809953704
...

with many many lines
my problem is that i want to change this presentation:
skip the 1st line, keep the first hundred lines, and the last colums as the first row and the the first column as the 2nd row.
How can i do that?
in example:
39815.4809027778,39815.4809027778,39815.4809143519,...
-47.2120018005371,-46.3009986877441,-46.277000427246...

thank you!

2 
> cat file141
Trace1: WRIT,Trace2: BLAN,Trace3: BLAN,
-47.2120018005371,,,39815.4809027778
-46.3009986877441,,,39815.4809027778
-46.277000427246,,,39815.4809143519
-46.7389984130859,,,39815.4809259259
-46.3460006713867,,,39815.4809259259
-45.9309997558593,,,39815.4809375
-46.9119987487792,,,39815.4809490741
-46.5379981994628,,,39815.4809606482
-46.9640007019042,,,39815.4809722222
-46.8390007019042,,,39815.4809722222
-47.1399993896484,,,39815.4809837962
-46.1370010375976,,,39815.4809953704

> cat fix141
tail +2 file141 | cut -d"," -f4 | tr "\n" "," >file141.tmp
echo "~" >>file141.tmp
tail +2 file141 | cut -d"," -f1 | tr "\n" "," >>file141.tmp
echo "~" >>file141.tmp
sed "s/,~//g" file141.tmp >file141.out
cat file141.out

> fix141
39815.4809027778,39815.4809027778,39815.4809143519,39815.4809259259,39815.4809259259,39815.4809375,39815.4809490741,39815.4809606482,39815.4809722222,39815.4809722222,39815.4809837962,39815.4809953704
-47.2120018005371,-46.3009986877441,-46.277000427246,-46.7389984130859,-46.3460006713867,-45.9309997558593,-46.9119987487792,-46.5379981994628,-46.9640007019042,-46.8390007019042,-47.1399993896484,-46.1370010375976
3 
BEGIN {
    FS = ","
    tmp1 = "/tmp/" PROCINFO["pid"] ".1"
    tmp2 = "/tmp/" PROCINFO["pid"] ".2"
}

{
    if (another_line) {
        printf "," >>tmp1
        printf "," >>tmp2
    }
    else
        another_line = 1

    printf $NF >>tmp1
    printf $1 >>tmp2
}

END {
    close(tmp1)
    close(tmp2)
    if (another_line) {
        system("cat " tmp1)
        print ""
        system("cat " tmp2)
        print ""
    }

    system("rm -f " tmp1)
    system("rm -f " tmp2)
}
4 
awk -F, 'BEGIN { OFS = FS }
NR == 1 { next }
NR < 100 { print; next }
{
 $0 = $NF "," $0
 $NF = ""
 print
}
' "$FILE"
5

Try this:

tail +2 <myfile> | head -100 | awk 'BEGIN { FS=","; COL1=""; COL2="" } { COL1 = COL1 $1; COL2 = COL2 $4 } END { print COL2 "\n" COL1 }'
6

sorry doesn't work...
fyi i'm using awk95 in a windows xp...

7

What does "doesn't work" mean? What happens? In what way is it not what you want?

Which script "doesn't work"?

8

with panos code: i have this error
awk95: can't open file /tmp/.1
input record number 1, file File_02.csv
source line number 18

with cfajohnson's code : the expected file is the same as the original but without the titles of columns.

9

this is what is expected:
39815.4809027778,39815.4809027778,39815.4809143519,...
-47.2120018005371,-46.3009986877441,-46.277000427246...

thank you!

10

perl might be a easy way to do so.

#!/usr/bin/perl
open FH,"<a.txt";
while(<FH>){
	chomp;
	next if $.==1;
	my @tmp=split(",",$_);
	push @arr,$tmp[3];
	push @brr,$tmp[0];
	last if ($#arr == $#brr) && ($#arr == 5);
}
print join " , ",@arr;
print "\n";
print join " , ",@brr;
11

Or try this:

sed '2,101!d' foo.txt | awk -F, ' { first=first "," $4; second=second "," $1 } END { print substr(first, 2); print substr(second, 2) }'