asked on

convert string KB/MB/GB values

Hi,

I want to process strings within a ksh script similar to this ...

07/17/09 15:55:00 74 10G 9968M 31% 27%

and convert fields with a "G" "M" or "K" value to a KB value with no suffix.
ie convert the example string to ...

07/17/09 15:55:00 74 10485760 10207232 31% 27%

Is there a simple way to do this using sed and/or awk?

ASKER CERTIFIED SOLUTION

ozo

membership

This solution is only available to members.

To access this solution, you must be a member of Experts Exchange.

Start Free Trial

gheist

If you do not pass -h parameter to df which generates this output you get sizes in disk blocks.

gormenghast

ASKER

ozo,
It looks good but i am having trouble spotting the syntax error...

# ksh -x sn
+ awk { while( match($0,/[0-9]+M/) ){ $0=substr($0,1,RSTART) (substr($0,RSTART,RLENGTH-1)*1024) substr($0,RSTART+RLENGTH); }while( match($0,/[0-9]+G/) ){ $0=substr($0,1,RSTART) (substr($0,RSTART,RLENGTH-1)*1024*1024) substr($0,RSTART+RLENGTH); } }1
+ echo 07/17/09 15:55:00 74 10G 9968M 31% 27%
awk: syntax error near line 1
awk: illegal statement near line 1
awk: syntax error near line 1
awk: bailing out near line 1

ozo

what version of awk are you using?
can you try it it with gawk or nawk?
echo 07/17/09 15:55:00 74 10G 9968M 31% 27% | awk '{ gsub("G","MM"); while( match($0,"[0-9]+M") ){ $0=substr($0,1,RSTART) (substr($0,RSTART,RLENGTH-1)*1024) substr($0,RSTART+RLENGTH); }}1'

gormenghast

ASKER

sorted ...
# ksh -x sn
+ nawk { while( match($0,/[0-9]+M/) ){ $0=substr($0,1,RSTART) (substr($0,RSTART,RLENGTH-1)*1024) substr($0,RSTART+RLENGTH); }while( match($0,/[0-9]+G/) ){ $0=substr($0,1,RSTART) (substr($0,RSTART,RLENGTH-1)*1024*1024) substr($0,RSTART+RLENGTH); } }1
+ echo 07/17/09 15:55:00 74 10G 9968M 31% 27%
07/17/09 15:55:00 74 110485760 910207232 31% 27%

thanks ozo - top solution as always

gheist

How do you know when G is 1024x1024x1024 and when it is 1000x1000x1000 ? kind of 7% difference but who cares less...

gormenghast

ASKER

1Gb is 1024Mb
1Mb is 1024Kb