Trying to do a count on multiple lines in a file

Shell Programming and Scripting
1

Guys I am having a problem with being able to do a count of entries in a file. What I am trying to get a count of the total number of members that are listed in the files. So I need to pull the number of the lines after members. I tried using sed but it only seems to count the first line(members line) and not the actualy members themselves.

Here is the output of Config Data that I am pulling from:

pool Happy1 {
   monitor all happy1-check
   members
      1.1.1.1:https
      1.1.1.2:https
      1.1.1.3:https
      1.1.1.4:https
      1.1.1.5:https
      1.1.1.6:https
}
pool Happy2 {
   members
      1.1.2.4:https
}
pool Happy3 {
   monitor all happy3-check
   members
      1.1.3.4:https
}
pool Happy4 {
   members
      1.1.4.4:https
}
pool Happy5 {
   monitor all happy5-check
   members
      1.1.5.4:https
}

So for the above I am looking for it to return a value of 10.

Thanks in Advance.

2 
grep -c "members" file
3

Unfortunately I can't search for anything specific in those lines cause it can be a port(80, 8080, etc...) or service name(https, dns, etc...)

Here is what I was working with and it works except for it pulls the members and the closed bracket } with it also. So I am getting incorrect data. I am not sure how to do it cleanly.

sed -n '/members/,/}/p' $CONF|wc -l

bash-3.00# sed -n '/members/,/}/p' input.txt |wc -l
     20

sed -n '/members/,/}/p' $CONF|more
4

Alternatively:

awk '/members/{c=1;next} /}/{c=0} c==1{tot++} END{print tot}' infile
5

anbu23's grep will work for that sample data, but in case "https" may occur elsewhere, or in case a member need not have "https" in its line, the following may be of use:

$ awk '$1=="members" {m=1;next} $1=="}" {m=0} m {i++} END {print i}' data
10

Assumptions: the list of members begins on the line immediately following a line whose first word is "members" and is terminated by a line beginning with a "}" (with optional leading whitespace).

Regards,
alister

6

Thx guys.

Alister, that was my assumption till I tried using your awk command. Yeah I found a wrinkle with the way the system writes the configuration file. So one version it is as listed above. Then an older version writes them as single line.

New code:

   members
      1.1.1.1:https
      1.1.1.2:https
      1.1.1.3:https

Older Code:

   members 1.1.1.1:https
   members 1.1.1.2:https
   members 1.1.1.3:https
7

hey, scottzx7rr:

If i understand you correctly, you have two different formats to deal with. perhaps:

$ cat old
   members 1.1.1.1:https
             members 1.1.1.2:https
             members 1.1.1.3:https

$ ./scottzx7rr.sh old
       3

$ ### Note that i emptied the Happy2 members section for testing purposes

$ cat new
pool Happy1 {
   monitor all happy1-check
   members
      1.1.1.1:https
      1.1.1.2:https
      1.1.1.3:https
      1.1.1.4:https
      1.1.1.5:https
      1.1.1.6:https
}
pool Happy2 {
   members
}
pool Happy3 {
   monitor all happy3-check
   members
      1.1.3.4:https
}
pool Happy4 {
   members
      1.1.4.4:https
}
pool Happy5 {
   monitor all happy5-check
   members
      1.1.5.4:https
}

$ ./scottzx7rr.sh new
       9

$ cat scottzx7rr.sh
#!/bin/sh

# Try the old format.  if egrep finds nothing, then try the new
if ! egrep 'members ([0-9]+\.){3}[0-9]+' "$1"; then
        sed -n '/members/,/}/{/members/d;/}/d;p;}' "$1"
fi | wc -l

The egrep doesn't do a strict job of matching an ip address, but you can tighten that up if the code is of any use to you and you think it's required :slight_smile:

Alister

8

Alister,

Thanks for your help. I got it working with tweaking of your script.