How to extract values from line?

pinga123 · 04-01-2015, 05:59 AM

I have a file in which every disk path is stored in disk parameter . However this file is dynamic and number of disk numbers can increase/decrease .
Is there a way to store this disk path values in file .

For example
This is a sample file .

disk = ['file:/OVS/repository/file1.img,xvda,w', 'file:/OVS1/repository/file1.img,xvdb,w']

I want diskpath.txt to be filled like
/OVS/repository/file1.img
/OVS1/repository/file1.img

i tried awk with my limited knowledge but couldn't able to figure the solution.
please help .

syg00 · 04-01-2015, 06:24 AM

If the data is always that well structured, I might be inclined to use colon and comma as field separators - that will give you well defined fields to print regardless of number of "file:" entries per record.

Have a go and we may be able to help if you still have trouble.

pinga123 · 04-01-2015, 06:32 AM

syg00 i tried using awk but with only one field separator using -F .

How to pick up values using two separator here its : and , . and as the values can increase decrease i m not sure how to determine the value e.g. $2 $3 in print .

pan64 · 04-01-2015, 06:33 AM

Is this suitable?
awk 'BEGIN { FS="[][,'"'"':= ]*" } { print $3" "$7 } '

syg00 · 04-01-2015, 06:52 AM

pan64 has shown one way to specify multiple separators - it can be done similarly with -F.
I would work on the premise that the number of files may be variable - 2 mostly maybe, but maybe only one, or more than 2. Use NF as an indicator.