bash capture output between two ""

[Drenriza]

Hi all.

I have some output as follows

SYSTEM "2M Hotel"

How can i capture the text between the two "" so the only output i get is 2M Hotel ?

Thanks on advance.
Kind regards.

[WasMeHere]

I think it works if you pipe your output through

sed s/.*\ \"//|sed s/\"$//

Have fun
Olle

[Lars Noodén]

Another way is to pipe it through awk (http://manpages.ubuntu.com/manpages/oneiric/en/man1/awk.1posix.html) while setting the Output Field Separator to a double quote (")



yourscript | awk -F '"' '{print $2}'


-F sets the field separator, $2 is the second column

Obviously there are limitations to this method.

[tors]

Another sed:


#> echo 'SYSTEM "2M Hotel"' | sed "s/.*\"\(.*\)\"/\1/g"
2M Hotel
#>

[dargaud]

Another sed, which avoid parentheses:

yourscript | sed -e 's/[^"]*"//' -e 's/".*//'

[Drenriza]

Thanks for all the suggestions guys. Wish i knew what all those characters in the sed commands ment. But it looks a bit exhausting :) to be honest.

[ofnuts]

Thanks for all the suggestions guys. Wish i knew what all those characters in the sed commands ment. But it looks a bit exhausting :) to be honest.These are "regular expressions" and any programmer worth his salt should know about them.

http://xkcd.com/208/

[Lars Noodén]

Thanks for all the suggestions guys. Wish i knew what all those characters in the sed commands ment. But it looks a bit exhausting :) to be honest.

They are not just for programmers, they're essential also to system administration.

There are lots of sources of material on regular expressions. For sed you can start here:

http://www.gnu.org/software/sed/manual/html_node/Regular-Expressions.html

http://www.faqs.org/docs/abs/HTML/regexp.html