Hi,

I wish to grep for the first instance of <listen-address> value between the first <server></server> tag in an xml file.

Sample xml:

.........
    <timeout-seconds>1500</timeout-seconds>
  </jta>
  <server>
    <name>Adminserver_DEV</name>
    <listen-address>myadminhost</listen-address>
  </server>
  <server>
    <name>managed_core</name>
<listen-address>secondadminhost</listen-address>
    <ssl>
      <enabled>false</enabled>
......

Try

 
awk -F "[<>]" '/listen-address/ && $3{print $3;exit}' file

First Occurance :

sed -n -e 's/.*<listen-address>\(.*\)<\/listen-address>.*/\1/p' "Your_file.xml" | head -1

sed -n -e 's/.*<listen-address>\(.*\)<\/listen-address>.*/\1/p' "Your_file.xml" | head -n | tail -1

Try this to print the listen-address in exactly the first <server>...</server> tag, or the hostname if missing/empty:

awk     'BEGIN {"hostname" | getline HN}
         /<\/server>/ {exit}
         /<server>/   {s=1}
         /<listen-address>/&& s {gsub (/<.?listen-address>| */, "")
                                 LA = $0
                                }
         END            {print LA?LA:HN
                        }
        ' file
myadminhost

This is more close to my request as it considers the first <server> tag only. However,

1. I would like read the result in a variable (myhost) and

2. for another tag <listen-port> if missing in the first <server> tag then would like to assign "7001" to "myport" variable.

Something like this:

 awk -F"[<->]" ' BEGIN { myh="";myp="7001" }
 /<\/server>/ { exit } 
 /<server>/ { s = 1 ; }
 s && /<listen-address>/ {  myh=$3; }  
 /<listen-port>/ {  myp=$3 ;}
 END { print "host:"myh"\t port->"myp }' sample

But will this populate the hostname command if <listen-address> tag is not found ?