I'm a SAN admin, and I'm currently upgrading a *hit load of switches. Before I do the upgrades I need to add some information from a data collection into a spreadsheet.

Here is an example line of data:

0 (00) G_Port {hostname-#} true 16 0

I want to delete everything before G_Port (can be G, E or F_PORT) and after the 'hostname-#' (hostname and # vary). I'm using vi to edit the data collection, and the already exists, so I can't just make a .csv and create a new spreadsheet. I'm copying the *_Port and pasting it into one column, and the same with the hostname into another column. I have 143 ports in each file and have about 30 files to go through. I'm getting tired of going line by line and doing "d4w" 143 times and then going back through and deleting end of line after the hostname to get the two entries I want to keep. Is there a regular expression I can use in vi that will do this. I'm thinking something similar to "1,$s/yada/yafa/g" only instead of using the substitute flag, use something to delete the first four "words" 0 (00) and then delete everything after "hostname"?

Any suggestions are greatly appreciated.

I don't know vi but how about

grep -o '[GEF]_PORT {a-zA-Z0-9-]}'

Or:
Sorry, I only know enough vi for emergency use.

If the output looks OK, you can add the -i for in place editing and don't need to redirect to a second file.

Vim is based on ed regular expressions, as is sed:
:%s/.*\([GEF]_Port {.*}\).*$/\1/

So the only difference is the line range which precedes the command. The % means to search & replace on every line.
You could modify the current line and the next 9 lines with
:.,+9s/.*\([GEF]_Port {.*}\).*$/\1/