How can I extract data from a command output with Bash?

Question

I'm new to Linux and I'd like to get the "DRIVE GB" from this output:

[randall@home gdrive]$ drive quota 
Name: Randall D
Account type:   UNLIMITED
Bytes Used: 290959662516         (270.98GB)
Bytes Free: 10995116277760       (10.00TB)
Bytes InTrash:  0                    (0.00B)
Total Bytes:    11286075940276       (10.26TB)

* Space used by Google Services *
Service                              Bytes                               
DRIVE                                270.98GB                            
PHOTOS                               0.00B                               
GMAIL                                0.00B                               
Space used by all Google Apps        270.98GB 

I want to get the "270.38" from DRIVE 270.98GB. I tried with sed reading a few posts here but couldn’t get it.

Solution

drive quota | sed -n 's/^DRIVE[[:blank:]]\{1,\}\([0-9.]\{1,\}\).*/\1/p'

Would extract the sequences of 1 or more decimal digits or . following a sequence of 1 or more blanks following DRIVE itself at the start of the line.

If your sed supports the -E option, it would be prettier with:

drive quota | sed -En 's/^DRIVE[[:blank:]]+([0-9.]+).*/\1/p'

Though then, you might as well use perl:

drive quota | perl -lne 'print $1 if /^DRIVE\h+([\d.]+)/'

With GNU grep, when built with (recent for \K) PCRE support:

drive quota | grep -Po '^DRIVE\h+\K[\d.]+'

With awk, you can also do:

drive quota | awk '$1 == "DRIVE" {print 0+$2}'

Comments

[Stéphane Chazelas]

And what will you do when it goes up to 0.921TB? It seems to me that the GB part is important here.