Data exploration with awk
Help 9 / 12
Arrays

Using arrays to tally burritos

You may be surprised to learn that awk supports arrays (even multidimensional ones!). As an example, let’s track how many of each kind of burrito was ordered in this dataset.

First, let’s extract all the lines with information about burrito orders. Instead of using an if statement for each type of burrito (chicken, steak, carnitas, and vegetarian), we’ll use the tilde operator ~ to perform a regular expression match:

awk -F "\t" '{ if($3 ~ /Burrito/) print }' orders.tsv | head

Now we can use the array counts to track how many of each burrito was ordered. We’ll use the syntax counts[burritoType] to access the counts of each burrito filling, and we can output the number of chicken burritos at the end:

awk -F "\t" ' \
   { if($3 ~ /Burrito/) counts[$3] += $2 } \
   END { print counts["Chicken Burrito"] }' orders.tsv

To loop through all keys and values of the array, we can simply use a for loop:

awk -F "\t" ' \
   { if($3 ~ /Burrito/) counts[$3] += $2 } \
   END { for(k in counts) print(k, counts[k]) }' orders.tsv

Note that there are 6 unmarked burritos in our dataset, likely due to missing data!

← Previous Next →