Shell Programming and Scripting

Hi,
I have looked all over for this. I am attempting to get a the substring of a string using sed since it seemed the best solution for this.

For example my string is: "zzz foo to you and bar123 or foo"

I would like to extract the text between "and" and "or" (it could be anything, but that was my test string). "and" and "or" are unique in the string and the example is a single line, but it would be good to have it work on an arbitrary chunk of text.

To do so I thought I could do a match from the beginning of the string and just replace from the start to and with nothing. Then do the same from the reverse:

echo "zzz foo to you and bar123 or foo" | sed -e 's/[\*]and//' | more

(I was piping the text into sed vs a file just for easy of copying and pasting, but it will be the output of another process.)

Now, I tried various things that I thought would do a wildcard match "from the beginning until 'and'".

Suggestions as to where I went wrong? (Or am I missing an obvious way to do this with sed/awk/grep/egrep etc?)

:-)

Thanks

There's no need for sed (and even less need for more):

Code:

string="zzz foo to you and bar123 or foo"
temp=${string##*and}
printf "%s\n" "${temp%%or*}"

If the ouput of the process is a single line, save the output of the other process in a variable then use the above lines:

Code:

string=$( another_process ... )

If there will be multi-line output, use sed or awk, e.g.:

Code:

another_process | sed 's/.*and\(.*\)or.*/\1/'

Thanks for the tips. ;-)