mjhjmtu

I've been trying to figure out how to find how many values in a column which are not present i.e. has just a "."

For example

car.txt

Car Colour mpg Year
vw_golf blue 56 2006
vw_polo red 66 2010
honda_civic white . 2007
ford_ka red . 2014

So the ones I'm interested in are the honda civic and ford ka as they don't have a number present for the mpg column and I want to find out how many values in the mpg column which doesn't have a value (in this case is 2).

The problem I'm having is the period seems to give errors when I use awk command.

Short grep approach:

grep -Ec '^S+s+S+s+.s+' file
2

• 
  -E - allow extended regular expressions


• 
  -c - print a count of matching lines


• 
  S+ - match non-whitespace character(s), it is a synonym for [^[:space:]]
  


• 
  s+ - match whitespace character(s), it is a synonym for [[:space:]]
  

Another grep method, if your file is strictly formatted to have the mpg start at column 28:

$ grep '............................' input
honda_civic white . 2007
ford_ka red . 2014

$ grep -c '............................' input
2

That's 28 periods (for "any" character) followed by an escaped period ("a period").

Or with awk, as mentioned in the comments:

$ awk '$3 == "."' input
honda_civic white . 2007
ford_ka red . 2014

$ awk '$3 == "."' input | wc -l
2