Reading off values from a large file

Shell Programming and Scripting
1

Hi,

I have a large output file (star.log), with many lines of the following type

*** T vavg [K] unburnt:  723.187 / burnt: 2662.000

What I would like to do is pick the values 723.187 and 2662.000 and

What I've got so far is

awk '/unburnt:.*burnt:/{Tu=$6;Tb=$NF}END{print Tu, Tb}' star.log

However, this is picking values from just one line and I'm not sure why. Any help will be appreciated. Thanks!

2

Did you try :

awk '/unburnt:.*burnt:/{print $6, $NF}' star.log
1 Like
3

Great cheer! That works!

4

Your problem with this: awk '/unburnt:.*burnt:/{Tu=$6;Tb=$NF}END{print Tu, Tb}' star.log is that you do the print in the END section. This will just print one instance.
Change it to:

awk '/unburnt:.*burnt:/{Tu=$6;Tb=$NF;print Tu, Tb}' star.log
or just
awk '/unburnt:.*burnt:/{print $6, $NF}' star.log
1 Like
5

Hi,

just one more way to get the required output.

echo "*** T vavg [K] unburnt:  723.187 / burnt: 2662.000" | awk -F" " '{print$6" "$9}'

Output will be as follows.

723.187 2662.000

Just adding if we have file then following may help.

awk -F" " '{print$6" "$9}' star.log

Thanks,
R. Singh

---------- Post updated at 04:57 AM ---------- Previous update was at 04:10 AM ----------

1 more solution for same, may help.

echo "*** T vavg [K] unburnt:  723.187 / burnt: 2662.000" | sed 's/[[:alpha:]]//g;s/[[:punct:]]//g'

Output will be as follows.

723187   2662000

Thanks,
R. Singh

6

@RavinderSingh13
I guess OP has more than one line and need the search argument that you have removed.

7

Yes and... by the way the sed statement proposed by RavinderSingh13 also removed the dot in the numbers :

The expected :

723.187   2662.000

became

723187   2662000

which - i guess - is not the expected result.
if willing to use sed you may want to give a try to something like :

sed '/unburnt:.*burnt:/!d;s/[^ .0-9]*//g' infile
1 Like