Striping/altering/sorting/deleting/searching strings with Bash only

So far, we have seen inkling of the power of commands available in Linux, and some of these are among the most powerful: sed and grep. However, while we can easily use these commands together, sed by itself or even using another very useful command called awk, we can leverage Bash itself to shave time and reduce external dependencies in a portable way!

So, how can we do this? Let's begin with a few examples using this Bash syntax:

#!/bin/bash
# Index zero of VARIABLE is the char 'M' & is 14 bytes long
VARIABLE="My test string"
# ${VARIABLE:startingPosition:optionalLength}
echo ${VARIABLE:3:4}

In the preceding example, we can see a special way of calling special substring functionality using ${...} , where VARIABLE is a string variable within your script (or even global), and then the following variable is the :. After the :, there is the startingPosition parameter (remember that strings are just arrays of characters and each character can be addressed by index), and there is another optional semicolon and length parameter (optionalLength). 

If we ran this script, the output would be:

$ bash script.sh
test

How is this possible, you may ask? Well, its possible with Bash's equivalent of substr (a function in C and many other programming languages), and this is achieved through the usr of the ${...}syntax. This tells bash to look inside for a variable named VARIABLE and then for two parameters: to start at byte/character 3 (technically 4 because arrays start at element 0 in Bash) and a length of 4 (to print only four characters). The result of the echo is test.

Can we do more with this, such as removing the last character? Deleting words? Searching? Of course, and all of this is covered in this recipe!