Better way to do tailing with awk

Shell Programming and Scripting
1

my current code:

varA=$(tail -200 /var/log/data.txt | egrep -c "Capital|capitol")
varB=$(tail -200 /var/log/data.txt | egrep -c "State|Country")
varC=$(tail -200 /var/log/data.txt | egrep -c "City|Town")

I want to do this a different way. something like:

AllVars=$(echo $(tail -200 /var/log/data.txt |  awk '/Capital|capitol/ {++c} END {print +c}' ;  awk '/State|Country/ {++c} END {print +c}' ; awk '/City|Town/ {++c} END {print +c}'))
varA=$(echo ${AllVars} | awk '{print $1}')
varB=$(echo ${AllVars} | awk '{print $2}')
varC=$(echo ${AllVars} | awk '{print $3}')

im pretty sure what i just did here is not efficient. i'm hoping someone can shorten it for me using awk.

2

You are right, this is not efficient. Pass through the file one time and use an awk script; the following shell script embeds the call to awk...

tail -200 /var/log/data.text | awk '
/[Cc]apital/ { varA=$0}
/State|Country/ { varB=$0 }
/City|Town/ { varC=$0}'

Just to point you in the right direction.

1 Like
3

You can accomplish the task efficiently in sh with a while loop to read and a case statement to match and assign.

Regards,
Alister

1 Like
4

If you want to put the output of the single awk into shell variables, you could try something like this:

read varA varB varC << EOF
$( 
  tail -200 /var/log/data.text | 
  awk '
    /Capital|capitol/ {a++}
    /State|Country/   {b++}
    /City|Town/       {c++} 
    END {print a+0,b+0,c+0}
  '
)
EOF
1 Like
5

An alternate method is

eval $(
  tail -200 /var/log/data.text |
  awk '
    /Capital|capitol/ {a++}
    /State|Country/ {b++}
    /City|Town/ {c++}
    END {print "varA="a+0, "varB="b+0, "varC="c+0}
  '
)
1 Like
6

thank you so much guys. this just saved me a lot of time!