I have an .xml file in which I need to I need to parse and extract data from via a ksh script.

The file (simplified) looks like this:

<Site>
blah blah blah
VARIABLE 1
</Site>
<Site>
blah blah blah
VARIABLE 1
</Site>

What I need to do is print all information between and including <Site> and </Site> of every occurrence to individual files and preferably only do so if VARIABLE 1 is matched.

I'm sure some sort of if/sed combo could do it, but I just don't know how.

All help is very much appreciated!

There is probably an easier way, but this will work:

#!/usr/bin/ksh
results=$(sed -e "s/$/:zzaqmkoz:/g" < tfile3 | tr -d "\n"|sed -e "s/<\/Site>/<\/Site>\n/g"|sed -e "s/:zzaqmkoz:<Site>/<Site>/g"|grep $1)
case $results in
"")
print BLANK
;;
*)
print $results|sed -e "s/:zzaqmkoz:/\n/g"|tr -s "\n"
;;
esac

This looks great... but a little complicated for my newbie (lack of) skills!

Could you please explain it a little more? Is tfile3 the .xml file I want to extract data from?

Many thanks!

Yeah sorry, I used tfile3 as my input file. You would just change taht to whatever your input file is. The script takes a variable as whatever you want to search for.

Ths script is not as complicated as it actually look, let me try to break it down for you. In defining $results the first command:
sed -e "s/$/:zzaqmkoz:/g" < tfile3
takes your file and throws :zzaqmkoz: at the end of every line (I use that as a custome delimiter so I can easily go back and replace it later, and im 99% sure it wont be found "naturally"

The second command:
tr -d "\n" removes the new line characters in the file, so now you have one HUGE line of text.

The third command:
sed -e "s/<\/Site>/<\/Site>\n/g"
takes any <\Site> it finds and replaces it with <\Site> followed by a new line. So now every line starts with :zzaqmkoz:<Site> then has your text, then ends with a </Site>, so it looks like:
:zzaqmkoz:<site> text text </site>
this makes it easy to search for your variable.

The next command:
sed -e "s/:zzaqmkoz:<Site>/<Site>/g"
finds and line that starts with my custom delimiter ":zzaqmkoz:" and the <Site>, and simply changes that back to <Site>.

The last command is a grep and simply searches for your variable.

The case command takes the results line above as its input. If it is blank, it simply prints blank. If it finds a result, it takes the custom delimiter :zzaqmkoz: and changes it to new line, then "squeezes" multiple newlines to one (the tr -s "\n" command).