how can i reverse the line order in text files?
(but total number of the lines is not constant )
for example i have a file like this:
line1
line2
line3
.
.
lineN
i wantto make it like this:
lineN
.
.
.
line3
line2
line1
U can try something like this>
#!/usr/bin/awk -f
#< AWK - Reverse order of words in each line of a file
{ cnt = NF;
while ( cnt > 0 ) {
printf( "%s ", $cnt );
--cnt;
}
printf( "\n" );
}
There is a GNU utility called tac, part of coreutils, that does what you want.
If you don't have it, and cannot install it:
awk '{x[NR] = $0}
END { while ( NR > 0 ) print x[NR--] }' [FILENAME ...]
sed '1!G;h;$!d' file
sed -n '1!G;h;$p' file
from sed - one liners
Try this...
cat -n <filename>| sort -r | cut -f2-20 > reversedfile
Experts please comment if anything wrong with this.... 
Though complex , You can have script presented below does the job!!
#!/bin/ksh
Tot=$(sed -n $= FILE )
i=0;
while [[ $i -le $Tot ]] ; do
tail -n"$i" FILE | head -1
((i=i+1))
done
You can extend the script for multiple files. Experts please comment on the approach
This is horrendously slow, as it calls two external commands for every line in the file. Here are timings for a 181-line file (xx.sh is your script):
$ time tac .bashrc > /dev/null
real 0m0.021s
user 0m0.002s
sys 0m0.005s
$ time awk '{x[NR]=$0}END{while (NR) print x[NR--]}' .bashrc > /dev/null
real 0m0.010s
user 0m0.003s
sys 0m0.006s
$ time xx.sh .bashrc > /dev/null
real 0m2.060s
user 0m0.483s
sys 0m1.420s
This might be a bit faster:
nl -ba inputfile | sort -nr | cut -f2-
I think you should use "awk" instead of cut.
cat -n <filename> | sort -r | awk '{print $2}' > reversedfile
tail -r <filename>
would reverse entire file.
The -r option to tail is not standard; it is available on *BSD, but not GNU.
$ tail -r .bashrc
tail: invalid option -- r
Try `tail --help' for more information
reverse.pl
#!/usr/bin/perl
print reverse <>;
reverse.pl filename
use command tac.
tac file1 > file1.tmp
mv file1.tmp file1 
A way to reverse the file and preserve the order. Number the lines, sort, remove the numbers.
cat -n myfile| sort -n -r -k 1 |cut -f2-
How can I reverse lines of the second and the third comlumns?
For example, from file
a 1 11
b 2 12
c 3 13
d 4 14
e 5 15
to like this:
a 5 15
b 4 14
c 3 13
d 2 12
e 1 11
(Sorry my pooring english.)
If you have the tac command:
cut -d ' ' -f1 file > col1
cut -d ' ' -f2- file | tac > col2
paste -d ' ' col1 col2
Thank you very much for the ready answer!
without tac you can do this way...
cat infile | cut -d ' ' -f1 > col1
cat infile | cut -d ' ' -f2 | sort -r > col2
cat infile | cut -d ' ' -f3 | sort -r > col3
paste -d ' ' col1 col2 col3
Another one:
awk 'NR==FNR{a[++i]=$2 FS $3;next}{print $1,a[i--]}' file file
Thanks! What if there are empty columns ahead of table?
For example:
a 1 11
b 2 12
c 3 13
d 4 14
e 5 15
instead of
a 1 11
b 2 12
c 3 13
d 4 14
e 5 15
-----Post Update-----
I found solution
#!/bin/sh
cut -d ' ' -f 1-8 file > col1
cut -d ' ' -f9- file | tac > col2
paste -d ' ' col1 col2