String Removal

[reason1000]

Hello,

Anyone have any example code on how to remove a string from a test file? I would like to be able to remove old uers accounts from my Linux server alias file.

Thanks,
Rod

 

[reason1000]

That should have read "how to remove a string from a text file" not a test file.

 

[BrianAtWork]

Maybe this helps?

testfile

aaa:123:456
bbb:123:456
ccc:123:456

perl -i -ple 's/^bbb:/:/' testfile

now, testfile contains:

aaa:123:456
:123:456
ccc:123:456

That answers your "example code on how to remove a string from a test file".

But your example makes it sound like you want to remove the line that contains a certain user, not just the username?

In that case:

perl -i -ne 'print unless(/^bbb:/)' testfile

testfile contains (from the original):

aaa:123:456
ccc:123:456

All of this, of course, depends on the format of your text file... I just gave an example of a colon delimited file where the userid is the first field. Post back if you have further questions or if you would like to clarify more details.

 

[reason1000]

Thanks for the help! A combination of both command line examples did the trick. I have both lines that begin with a user account name and lines that have the same user account name in the middle or end, that need to be removed.

Here's the line I'm using to do this:

system ("perl -pe 's/$useracct,//g' $file | perl -i -ne 'print unless (/^$useracct/)'");

I then direct standard output to where I want it to go such as a new file. I am still playing with different ideas and might do an edit-in-place, backing up the original file, of course.

Thanks again!

 

[KevinADC]

you might find this resource of interest:

http://www.perl.com/pub/a/2004/10/14/file_editing.html