HPC

GREP

GREP (Global regular expression print) is a way of navigating text using regular expressions. It is similar to find and replace in a word document and has many very useful applications in computing.

The basic pattern for using a grep command is `grep

file'

The options are:

• i: case insensitive
• n: gives the line number of the search term
• c: gives the count of the lines the search term is found in
• v: gives the lines the search term is not found in
• -A 5 gives the five lines that come after a search term
• -B 5 gives the five lines that come before a search term
• -w gives only the lines where the whole word is found
• -l gives the filename where the match is found

Wildcards

There are some symbols that can be used to represent more than one thing. For example, . is used to designate any character so the command grep "b.g" filename could return lines with the words big, bag or bog.

To search for a . you have to escape the function using \.

The * is used to search for as many of the previous character as there is. For example `grep “b.*g” would return things like bag, bog and bug, but also blog, berg and betting.

Additional search terms include

• \w for searching for letters, numbers and _
• \d will search for numbers
• \r will search for return character
• \s Space, tab or end of line
• [A-Z] Single character in the given range
• [aeiou] Single vowel
• ^ Beginning of line character
• $ Last position before end-of-line character

Capturing and Replacing

• () captures a search result for use in a replacement term
• \1 Substitutes matched item into replacement term in numerical order

Examples

If you want to extract particular rows from a file, that include the words “key words”, you can use the following command:

grep "key words" filename

and it will list all the lines that include those words. If you want to put those lines into a csv, use the command

grep "key words" filename > newfile.csv