I have a number of files in a common directory (/home/test) with a common name:
ABC_1_20110508.out
ABC_1_20110509.out
ABC_1_20110510.out
..
Each text file has one record that looks like this:
(count, 553076)
I would like to strip out the numbers and just list them out in a file one at a time.
553076
1005
7778000
...
Can someone show me how to do this using perl?
use this regex:
/\(\w+, (\d+)\)/
you can also use the magic diamond operator to iterate over all of the files at once:
while (<>) {
# extract the number
/\(\w+, (\d+)\)/;
# print it out
print $1, "\n";
}
And if your perl script is called myscript.pl, the you can call it like this:
$ myscript.pl /home/test/ABC_1_*.out
can't open, no such file... Somehow it doesn't like that /home/test/ABC_1_*.out when I pass it in as you suggested. I can certainly ls -l ABC_1_* without no problems
cding to that directory and using myscript.pl *
Sounds like a one-liner to me:
$ perl -wne '/(\d+)/ && print "$1\n"' *.out > out.txt
perl -wne '/(\d+)/ && ($i += $1); print qq($i\n);' *.txt Or if you prefer just to receive the sum from out.txt: perl -we 'while(<ARGV>) { $i += $_; } print $i' out.txtEasiest way is to use the <> operator. When invoking a perl program without arguments, <> acts just like <STDIN>. If you call it arguments, <> will give you the contents of every file in @ARGV without you having to manually manage the filehandles.
Ex: ./your_script.pl /home/test/ABC_1_????????.out or cat /home/test/ABC_1_????????.out | ./your_script.pl. These would have the same effect.
