deleting even numbered lines

bharatbsharma · 11-26-2007, 03:35 AM

In a file how can i delete line no. 2 , 4 , 6 .. etc?

matthewg42 · 11-26-2007, 03:47 AM

Code:

perl -i -ne 'print if ($. % 2);' your_file

Alien_Hominid · 11-26-2007, 03:48 AM

Count (\r)\n occurences. http://en.wikipedia.org/wiki/Newline

radoulov · 11-26-2007, 04:19 AM

Awk:

Code:

zsh-4.3.4-dev-2% print -l {1..10}|awk NR%2
1
3
5
7
9

jschiwal · 11-26-2007, 04:28 AM

sed -n '1~2p' file >newfile

or

sed -i '2~2d' file
to edit the file in place.

radoulov · 11-26-2007, 04:57 AM

Quote:

Originally Posted by jschiwal

sed -n '1~2p' file >newfile

or

sed -i '2~2d' file
to edit the file in place.

Nice one!
(only GNU Sed, I suppose)

jschiwal · 11-26-2007, 05:24 AM

Yes, that is a gnu extension.

Just an FYI, take a look at the respective sized of the utilities used for regex searches:
ls -l /bin/grep /bin/sed /bin/gawk /usr/bin/perl
-rwxr-xr-x 1 root root 295432 Sep 21 14:16 /bin/gawk
-rwxr-xr-x 1 root root 201056 Sep 21 14:20 /bin/grep
-rwxr-xr-x 1 root root 53056 Sep 21 14:15 /bin/sed
-rwxr-xr-x 3 root root 1301600 Nov 7 18:58 /usr/bin/perl

I was surprised that grep was almost as large as gawk. Perl is 25.5 times the size of sed.

ghostdog74 · 11-26-2007, 05:34 AM

Code:

awk 'NR%2==1' file