Shell script with awk to extract substring
I am modifying my .profile on an AIX system so it will capture my IP Address into a variable and then export it as my env variable DISPLAY.
Below is code that works but the last octet of the IP address can be one, two or three digits long. If it is less than three digits, I want to suppress the trailing space before I concatenate additional stuff.
The value of the variables are as follows:
SSH_CLIENT is an ip address and additional information as follows:
SSH_CLIENT=192.168.110.31 1234 56
with the code below, I am able to extract into my_ipadr as follows:
my_ipadr=192.168.110.31 :0.0 (with a trailing space after the 31 and before the:0.0)
When I concatenate the last part, the space becomes a problem. WHen it is assigned to the DISPLAY variable it looks like this:
DISPLAY=192.168.110.31 :0.0
complete with the space between 31 and :0.0
How can I "trim" the space off prior to concatenation? I can't change the substr command so it only picks up two numbers because it is possible for it to be three numbers. and if it is only one number then I will have two spaces.
I have toyed with the # (sharp) character, but have only gotten errors.
Thanks in advance.
Below is code that works but the last octet of the IP address can be one, two or three digits long. If it is less than three digits, I want to suppress the trailing space before I concatenate additional stuff.
The value of the variables are as follows:
SSH_CLIENT is an ip address and additional information as follows:
SSH_CLIENT=192.168.110.31 1234 56
with the code below, I am able to extract into my_ipadr as follows:
my_ipadr=192.168.110.31 :0.0 (with a trailing space after the 31 and before the:0.0)
When I concatenate the last part, the space becomes a problem. WHen it is assigned to the DISPLAY variable it looks like this:
DISPLAY=192.168.110.31 :0.0
complete with the space between 31 and :0.0
How can I "trim" the space off prior to concatenation? I can't change the substr command so it only picks up two numbers because it is possible for it to be three numbers. and if it is only one number then I will have two spaces.
I have toyed with the # (sharp) character, but have only gotten errors.
Thanks in advance.
my_ipadr=$(echo $SSH_CLIENT | awk -F. '{print $1"."$2"."$3"."substr($4,1,3)":0.0"}')
echo $my_ipadr
export DISPLAY=$my_ipadrASKER CERTIFIED SOLUTION
membership
This solution is only available to members.
To access this solution, you must be a member of Experts Exchange.
woolmilkporc
my_ipadr=$(echo $SSH_CLIENT | awk '{print $1 "0:0"}')
Sorry, typo:
my_ipadr=$(echo $SSH_CLIENT | awk '{print $1 ":0.0"}')
my_ipadr=$(echo $SSH_CLIENT | awk '{print $1 ":0.0"}')
SOLUTION
membership
This solution is only available to members.
To access this solution, you must be a member of Experts Exchange.
ASKER
farzanj,
Your script works perfectly.
I am not educated in sed, so I need to ask will this work even if the last octet of the IP address is three numbers? For example:
192.168.110.1101234 56
I believe your solution is exactly what I need... but first I need to ask a question of woolmilkporc...
woolmilkporc,
Your script works perfectly also.
So I may have a problem with using the -F switch? I am new to awk. It would appear that if I omit the -F switch, it parses the string differently. Does it depend upon finding a space, using the space as a delimiter? I notice that with my original script with the -F, it uses the "." (period) as a delimiter.
If that is the case, when the last octet is three numbers, it will look like this:
192.168.110.1101234 56 (with no space between the fourth octet and the additional number at the end of the string (1234 56). Is this going to present a problem?
Your script works perfectly.
I am not educated in sed, so I need to ask will this work even if the last octet of the IP address is three numbers? For example:
192.168.110.1101234 56
I believe your solution is exactly what I need... but first I need to ask a question of woolmilkporc...
woolmilkporc,
Your script works perfectly also.
So I may have a problem with using the -F switch? I am new to awk. It would appear that if I omit the -F switch, it parses the string differently. Does it depend upon finding a space, using the space as a delimiter? I notice that with my original script with the -F, it uses the "." (period) as a delimiter.
If that is the case, when the last octet is three numbers, it will look like this:
192.168.110.1101234 56 (with no space between the fourth octet and the additional number at the end of the string (1234 56). Is this going to present a problem?
Without -F the space is the delimiter.
$1 is the string up to, but not including, the first space, and $1 will of course contain nothing that follows this space (that would be $2 etc.)
$1 is the string up to, but not including, the first space, and $1 will of course contain nothing that follows this space (that would be $2 etc.)
ASKER
woolmilkporc,
I posted my last comment at the same time you posted yours. So, with the easier "one liner", I have to ask the same question...
I posted my last comment at the same time you posted yours. So, with the easier "one liner", I have to ask the same question...
ASKER
Well then it won't work with no space after the fourth octet, as in my example in my second post...?
The string in your second post will never appear in $SSH_CLIENT (there will always be a space following the IP address, regardless of the number of digits in the fourth octet).
So why would you construct a problem where there is actually none?
My second solution uses "inline variable editing"
%%" "* means delete (starting from the right) the first space and all that follows,
So why would you construct a problem where there is actually none?
My second solution uses "inline variable editing"
%%" "* means delete (starting from the right) the first space and all that follows,
ASKER
I did not know that there would always be a space between the fourth octet and the other numbers. I am obviously new to the SSH_CLIENT environment variable as well.
Thanks to both of you. You both have excellent solutions and I am awarding each of you half of the points.
Thanks again! I REALLY appreciate the help.
Thanks to both of you. You both have excellent solutions and I am awarding each of you half of the points.
Thanks again! I REALLY appreciate the help.
My sed command would work no matter what. It is only erasing any spaces, anywhere, nothing else.
ASKER
farzanj,
Yes, I know! Your script is the one that I favor and am using. I favor it for that VERY reason. I have tested both of them and found that with woolmilkporc's script, it has a problem if there is no space for a delimiter. He assures me that the SSH_CLIENT variable will never have a space, but I'd rather be safe than sorry.
Thanks much!
Yes, I know! Your script is the one that I favor and am using. I favor it for that VERY reason. I have tested both of them and found that with woolmilkporc's script, it has a problem if there is no space for a delimiter. He assures me that the SSH_CLIENT variable will never have a space, but I'd rather be safe than sorry.
Thanks much!
You are most welcome. Glad you liked it.