remove part of a word

B-Boy · 08-08-2008, 12:35 AM

Hi I would like to remove part of a word but with cut i cannot put no space for the delim

Code:

cat /home/xxx/Desktop/xxx/file |grep lin|cut -d " " -f 1|cut -d"" -f 1
cut: the delimiter must be a single character
Try `cut --help' for more information.

what i am trying in the above code is for example "I search for the word linux0001 but i only want the middle character "ux0"

please help and thanks in advanced

Mr. C. · 08-08-2008, 12:59 AM

Are you saying you just want the first 3 characters after anything that contains "lin"?

If that is all you need:

sed d -n 's/.*\lin\(...\).*/\1/p' /home/xxx/Desktop/xxx/file

Beware, you probably want to be more specific, so that things like /usr/linux/foo does not match. Clarify as much as you can the pattern that is necessary to correctly determine your needs.

B-Boy · 08-08-2008, 01:13 AM

ok what i am trying to do is from the word usa12312301 which in my case means usa is country and 123123 is branch and 01 wkst i just want the branch number...so far i have managed to get an output of the numbers only 12312301 but now i want to take out the last two numbers 01 so that i end up with just the branch num 123123

Mr. C. · 08-08-2008, 01:20 AM

You want to use regular expressions here if the parts of your string vary in length. If they are all fixed length, or have clear delimiters, you can use cut or other tools for cutting operations.

So, are you branches always 6 digits, and branch numbers 2 digits?

chrism01 · 08-08-2008, 01:27 AM

As Mr C said, if its a fixed format you could use cut:

echo usa12312301|cut -c4-9
123123

B-Boy · 08-08-2008, 01:28 AM

yes the branch is always 6 digit and the PCs are two digits ....the region isa always 3 digits

B-Boy · 08-08-2008, 01:43 AM

Quote:

Originally Posted by chrism01

As Mr C said, if its a fixed format you could use cut:

echo usa12312301|cut -c4-9
123123

thanks this is exactly what i am looking for :-)

B-Boy · 08-08-2008, 02:17 AM

last thing the problem i am getting now is that it cuts the character correctly but only on the first string .... usa12312301 after running echo usa12312301|cut -c4-9 i get 123123 but my file has many strings usa12312301 usa12355502 usa1236673 now it only displays the first one sting 123123 i want the cmd to work on each sting in a file and echo all

chrism01 · 08-08-2008, 02:35 AM

Now you tell me ... just put it in a loop. ie assuming the values are space or newline separated:

Code:

for rec in `cat file`
do
    echo $rec|cut -c4-9
done

That should get you started.
See
http://tldp.org/LDP/Bash-Beginners-G...tml/index.html
http://www.tldp.org/LDP/abs/html/

ghostdog74 · 08-08-2008, 02:38 AM

then use awk

Code:

awk '{
 for ( i=1 ; i<=NF; i++ ) {
  print substr($i,4,5)
 }
}
' file

B-Boy · 08-08-2008, 03:47 AM

Quote:

Originally Posted by chrism01

Now you tell me ... just put it in a loop. ie assuming the values are space or newline separated:

Code:

for rec in `cat file`
do
    echo $rec|cut -c4-9
done

That should get you started.
See
http://tldp.org/LDP/Bash-Beginners-G...tml/index.html
http://www.tldp.org/LDP/abs/html/

thanks guys...that helped