pattern search and replace

[moonchild]

Hi,

I have a text file with configuration information gathered from my systems.

I need to search those files for any IP address and delete it or mask it. I might have different IP addresses in the same file. So I am looking for a pattern like xxx.xxx.xxx.xxx to delete or maske.

How can I do that?

thank you in advance

[IT_2007]

you can open the file with vi and do this way
:$s/oldstring/newstring/g

which will replace oldstring to newstring globally in the file.

[A. Clay Stephenson · Accepted solution]

There are many ways to do this. Among them are sed, Perl, and awk. I'll use sed this time:

sed -e '/dog/d' -e '/cat/d' -e 's/cow/Giraffe/g' < infile > outfile

This would read infile and delete every line that contains 'dog' or 'cat' and substitutes "Giraffe" for every instance of "cow" and write the result to outfile.

[James R. Ferguson]

Hi:

If you want to change an address pattern, you could use:

# perl -pe 's/\Q127.0.0.1/127.0.0.2/' file

...and if you wanted to delete lines containing it:

# perl -ne 'print unless m/\Q127.0.0.1/' file

Regards!

...JRF...

[Bill Hassell]

The simplest is to use grep to extract all the IP addresses you want:

grep -e 120.340.056.078 -e 012.012.013.103 -e 020.030.040.050 -e 015.016.170.180 your_filename

Note that this assumes you want just certain addressess and that all addresses have the same format (ie ###.###.###.###) but if the addresses have leading zeros stripped (ie, 12 rather than 012), then you'll have to match exactly what is in the file.

To change a record that matches your search criteria, you'll need to use awk or a shell or Perl script.

[Marvin Strong]

This might not be the best way, but you can strip all IP patterns out of a file pretty easy with perl.

perl -p -e 's/([\d]+)\.([\d]+)\.([\d]+)\.([\d]+)//' file

if you wanted to delete them from the file completely.

perl -pi -e 's/([\d]+)\.([\d]+)\.([\d]+)\.([\d]+)//' file

[moonchild]

Marvin,

I am new to scripting and perl.

How can I put the command you suggested in a shell script so I apply it on all the files in a directory? I am trying the following but it's not working but if I run your command on a specific file from the command line it works fine.

#!/usr/bin/sh

for FILE in `ls *.html`
do
{
perl -p -e 's/([\d]+)\.([\d]+)\.([\d]+)\.([\d]+)//' $FILE > noip-$FILE;
}
done