4 Things to know about awk

1. Language structure: The basic format of scripts are ```awk pattern1 { ACTION; ACTION; … }

pattern2 { ACTION; …}

The input is always one line at a time of a text file. Each pattern is checked; if it matches/passes, the actions within the braces are run. Also if you leave off the pattern, it runs the actions on every line, 

2. **Fields**: awk splits the files into "fields". You use `$1` to get the first column, `$2` for second, etc. `$0` prints the entire line. So if you want to rearrange the columns to be (4, 2, 3, 1), you'd write
```awk
{print $4, $2, $3, $1}

1. Variables You can initialize variables inside the brace of a BEGIN { ... } block:

   # initialize a variable x to 0
   BEGIN { x = 0 } 
   # add up all of the second column
   { x += $2 }
   # print it after going through all the lines
   END { print x }
   

   There are “special” variables, which include the input/output “field separator” FS/OFS (by default any amount of white space) or input/output “record separator” RS/ORS (by default \n)

2. Comparisons: You can do comparisons which are “fuzzy matching” like in javascript: numbers are coerced to strings, strings to numbers:

   # print lines where column 1 is the string "a", column 2 is a number > 10
   ($1 == "a") && ($2 > 10) {print $0}
   

Based on (1), I just learned that why my most commonly used awk one liner, ps aux | grep <blah> | awk {print $2} | xargs kill will run on every line and has to be surrounded by braces. - Any apparently is redundant with the grep? Now I know I can do ps aux | awk '/blah/' {print $2}

References

1. https://ferd.ca/awk-in-20-minutes.html
2. https://learnxinyminutes.com/docs/awk/