Shell Programming and Scripting

I need something to strip out some text from a string. I basically have a variable that will always be in the same pattern but will be a different length. Example below. I need to somehow extract the text between <Strat> and </Source> from example below. I can't load it into a file for audit reasons so can only pass a variable to whatever command I use. The string length in between will also regularly change.

XXX,XXX,XXX,XXX<Strat>CALIBRI</Source>


I need the output to simply be "CALIBRI" however this text can be all sorts of lengths. Any suggestions welcome

Code:

echo 'XXX,XXX,XXX,XXX<Strat>CALIBRI</Source>'|sed 's/.*>\(.*\)<.*/\1/'

or:

Code:

echo 'XXX,XXX,XXX,XXX<Strat>CALIBRI</Source>'| awk -F"[<>]" '{print $3}'

Excellent. Thanks.

Code:

echo "XXX,XXX,XXX,XXX<Strat>CALIBRI</Source>" |sed -e s/.*\<Strat\>// | sed s/\<\\/Source\>//

Unfortunately both of Franklin's solutions do rather make an assumption that there won't be any other chevrons on the line. These aren't ideal solutions.

I prefer the answer submitted by palaisvr, although it doesn't require two separate invocations of sed and should be tidied up a little bit:

Code:

echo 'XXX,XXX,XXX,XXX<Strat>CALIBRI</Source>' | sed -e 's=^.*<Strat>==' -e 's=</Source>.*$=='

Best regards,
Mark.

using perl,

Code:

perl -nle 'print $1 if /<Strat>(.+?)<\/Source>/;'  inputfile