awk : Filter a set of data to parse header line and last field of multiple same match.

Shell Programming and Scripting
1

Hi Experts,

I have a data with multiple entry , I want to filter PKG= & the last column "00060110" or "00088150" in the output

file:

###############################################################################################
PKG= P8SDB     :: VGS = vgP8SOra vgP8SDB1 vgP8S001 vgP8S007 vgP8S003 vgP8S004 vgP8SOlog 
PKG= P8SDB    disks:                    
-----------------------------------------------------------------------------------------------
/dev/rdisk/disk7777   xx  00   00  xxxx  07:67  xxxx 00060110
/dev/rdisk/disk7737   xx  00   00  xxxx  02:03  xxxx 00060110
/dev/rdisk/disk7741   xx  00   00  xxxx  07:65  xxxx 00060110
/dev/rdisk/disk7744   xx  00   00  xxxx  07:68  xxxx 00060110
/dev/rdisk/disk7745   xx  00   00  xxxx  07:66  xxxx 00060110
/dev/rdisk/disk7746   xx  00   00  xxxx  07:64  xxxx 00060110
/dev/rdisk/disk7750   xx  00   00  xxxx  07:63  xxxx 00060110
/dev/rdisk/disk7751   xx  00   00  xxxx  02:00  xxxx 00060110
/dev/rdisk/disk7753   xx  00   00  xxxx  02:07  xxxx 00060110
/dev/rdisk/disk7754   xx  00   00  xxxx  07:67  xxxx 00060110
/dev/rdisk/disk7755   xx  00   00  xxxx  02:01  xxxx 00060110
----------------------------------------------------------------------------------------------
###############################################################################################
PKG= P9SDB     :: VGS = vgP9SOra vgP9SDB1 vgP9S001 vgP9S007 vgP9S003 vgP9S004 vgP9SOlog 
PKG= P9SDB    disks:                    
-----------------------------------------------------------------------------------------------
/dev/rdisk/disk7778   xx  00   00  xxxx  06:67  xxxx 00088150
/dev/rdisk/disk7739   xx  00   00  xxxx  03:03  xxxx 00088150
/dev/rdisk/disk7742   xx  00   00  xxxx  06:65  xxxx 00088150
/dev/rdisk/disk7743   xx  00   00  xxxx  06:68  xxxx 00088150
/dev/rdisk/disk7745   xx  00   00  xxxx  06:66  xxxx 00088150
/dev/rdisk/disk7747   xx  00   00  xxxx  06:64  xxxx 00088150
/dev/rdisk/disk7758   xx  00   00  xxxx  06:63  xxxx 00088150
/dev/rdisk/disk7752   xx  00   00  xxxx  03:00  xxxx 00088150
/dev/rdisk/disk7759   xx  00   00  xxxx  03:07  xxxx 00088150
/dev/rdisk/disk7751   xx  00   00  xxxx  06:67  xxxx 00088150
/dev/rdisk/disk8755   xx  00   00  xxxx  03:01  xxxx 00088150
----------------------------------------------------------------------------------------------

The output to be look like:

PKG= P8SDB     :: VGS = vgP8SOra vgP8SDB1 vgP8S001 vgP8S007 vgP8S003 vgP8S004 vgP8SOlog   :: 00060110
PKG= P9SDB     :: VGS = vgP9SOra vgP9SDB1 vgP9S001 vgP9S007 vgP9S003 vgP9S004 vgP9SOlog   :: 00088150

Thanks a lot,

2 
awk '/^PKG/&&!/disks/{ printf "%s", $0; f=1}f==1&&/^\//{ printf ":: %s\n",$NF; f=0;}' filename
1 Like
3 
awk '/::/{p=$0} p && /^\//{print p,"::",$NF; p=x}' file

sed -n '/::/{N;N;N;s/\n.* / :: /p;}' file
1 Like
4

bipin & Scrutinizer,

Thanks a lot,

Scrutinizer,
in the sed example ,

s/\n.*

means what, could not understand , also how you able to capture the last field i.e 00060110 or 00088150.

Thanks.

5

Hi, s/\n.* / :: /p means replace everything in the line buffer between the first linefeed and the last space with " :: " and print the result.

So for the first match the line buffer will contain the following after the 3 N-operations:

PKG= P8SDB     :: VGS = vgP8SOra vgP8SDB1 vgP8S001 vgP8S007 vgP8S003 vgP8S004 vgP8SOlog \nPKG= P8SDB    disks:                    \n-----------------------------------------------------------------------------------------------\n/dev/rdisk/disk7777   xx  00   00  xxxx  07:67  xxxx 00060110

so
\nPKG= P8SDB disks: \n-----------------------------------------------------------------------------------------------\n/dev/rdisk/disk7777 xx 00 00 xxxx 07:67 xxxx
will get replaced.. ( \n is representation of the linefeed in this example)

6

Scrutinizer , much better .. Thanks again,