Efficient Linux at the Command Line

...

$ grep Washing areacodes.txt
202	DC	Washington
227	MD	Silver Spring, Washington suburbs, Frederick
240	MD	Silver Spring, Washington suburbs, Frederick
⋮

Count the area codes with wc:

$ wc -l areacodes.txt
375 areacodes.txt

Find the state with the most area codes (the winner is California with 38):

$ cut -f2 areacodes.txt | sort | uniq -c | sort -nr | head -n1
     38 CA

Convert the file to CSV format to import into a spreadsheet application. Print the third field enclosed in double quotes to prevent its commas from being interpreted as CSV separator characters:

$ awk -F'\t' '{printf "%s,%s,\"%s\"\n", $1, $2, $3}' areacodes.txt \
  > areacodes.csv
$ head -n3 areacodes.csv
201,NJ,"Hackensack, Jersey City"
202,DC,"Washington"
203,CT,"New Haven, Stamford"

Collate all area codes for a given state onto a single line:

$ awk '$2~/^NJ$/{ac=ac FS $1} END {print "NJ:" ac}' areacodes.txt
NJ: 201 551 609 732 848 856 862 908 973

or collate for each state, using arrays and for loops as in “Improving the duplicate file detector”:

$ awk '{arr[$2]=arr[$2] " " $1} \
         END {for (i in arr) print i ":" arr[i]}' areacodes.txt \
  | sort
AB: 403 780
AK: 907
AL: 205 251 256 334 659
⋮
WY: 307

Turn any of the preceding commands into aliases, functions, or scripts, whatever is convenient. A simple example is the areacode script in Example 9-4.

Example 9-4. The areacode script

#!/bin/bash
if [ -n "$1" ]; then
  grep -iw "$1" areacodes.txt
fi

The areacode script searches for any whole word in the areacodes.txt file, such ...