Introduction to Programming for HVAC Part-4

This article builds upon our last article, so make sure to catch up before continuing with this article.

Arguments

Before we start creating our program that will remove the background from images let’s go over arguments in shell scripts. Arguments are supplied to shell scripts are separated by a space " ", as opposed to options which start with a -<character> or --<word>.

To illustrate this, let’s change our hello-world script we wrote in the last article.

Move into our scripts directory:

cd ~/.local/bin

Make a copy of the hello-world script:

cp hello-world greet

Above we make a copy of the hello-world file and name the copy greet.

Open the greet file:

open -a TextEdit greet

Note: Because the greet file is an executable, in order to open it in the TextEdit application we must supply the -a option. Otherwise it will just run our greet program in another terminal. Use man open to read more about the open command.

Edit the greet file:

#!/bin/sh

echo "Hello, ${1}!"

Make sure to save ⌘s the file.

Take note that the quotes need to be changed to " (double quotes) from our original hello-world program.

The ${1} is indicating that we will insert / interpret the first argument passed to our program and insert it at that location. Arguments are interpreted in order they are passed in with ${0} always representing the filename of the program that is called (generally not needed / used in your scripts).

Test it out:

./greet Michael

If you’d like to supply multiple words (or arguments that contain spaces) as a single argument then you can wrap them in quotes.

./greet "Michael Housh"

Tip: Wrapping in quotes is especially useful for commands that take file paths, if any of the folders or file names contain spaces.

More Useful Program

At this point, it’s time to build a more useful program that we can use. First, we must download some dependencies for our program.

Install imagemagick:

brew install imagemagick

Note: If you’d like to check out the documentation / website for imagemagick you can run brew home imagemagick.

This will take a bit for brew to install imagemagick and it’s dependencies. When it’s completed, we can check to make sure that imagemagick is installed by running the following command.

magick --version

It should output something along the lines of this below.

Version: ImageMagick 7.1.1-17 Q16-HDRI aarch64 21569 https://imagemagick.org
Copyright: (C) 1999 ImageMagick Studio LLC
License: https://imagemagick.org/script/license.php
Features: Cipher DPC HDRI Modules OpenMP(5.0)
Delegates (built-in): bzlib fontconfig freetype gslib heic jng jp2 jpeg jxl lcms lqr ltdl lzma openexr png ps raw tiff webp xml zlib
Compiler: gcc (4.2)

Tip: Don’t forget, you can use the clear command to clear the terminal.

Create our script:

touch mktrans

We are going to name our script mktrans as a short for make transparent.

Open the file:

open mktrans

The program:

#!/bin/bash

# The input file path, passed in as the first argument.
inputfile="${1}"

# The color to make transparent, optionally passed in as the second argument.
# by default we handle / make white backgrounds transparent.
color="${2:-white}"

# Use the built-in basename command to normalize the input file name
# this will convert a file path, such as ~/Pictures/my-image.png to my-image.png.
fullfilename=$(basename -- "$inputfile")

# Use the text of the `fullfilename` up to the first '.' as the file name.
filename="${fullfilename%%.*}"

# Use the text after the last '.' as the file extension.
extension="${fullfilename##*.}"

# Create the output file name to use.
#
# For an input file of `input.png`, our output name would be
# `input-transparent.png`.
#
# This will output the file in the directory that we are
# in when we use our script (which may different than where
# the image is located)
outputfile="${filename}-transparent.${extension}"

# Make the magick happen :)
convert "${inputfile}" -fuzz 10% -transparent "${color}" "${outputfile}"

I’ve included comments in the program above, which is good practice, as there is high odds that you will forget what is going on when / if you open the file up in the future to look at it. We are using a lot of what is called parameter expansion magic in this file. You can read up more on what we are doing in the bash documentation .

This script is far from perfect, there are a lot of things to be improved upon. For example, if you download / save the banner image of this post and run this script, it will also remove some color in the wizards beard, eyes, and eye brows. However, it does work very well for my general use case, which is to remove the background from screenshots of pdf documents. It should be noted that it will not work on all types of images, some image types do not allow transparency, so it is safest to call this with input image being a .png image type, however you can use the imagemagick program that we downloaded to convert other image types to .png, but that will be left up to you to figure out.

Using Our Program

This is going to assume that you have download the banner image at the top of this article. You can do this by right-clicking and choosing Save As. This should save the image in your downloads folder, and you can keep the name of part4-banner.png.

Make the program executable:

chmod u+x mktrans

Make the image background transparent:

./mktrans ~/Downloads/part4-banner.png

Open the image:

open part4-banner-transparent.png

It should look like below.

Note: If you are viewing this site in light mode, the image does not look that bad. Hit the moon button in the top above my profile image to see some of the flaws of our program.

Tip: Remove the image from the ~/.local/bin by using rm part4-banner-transparent.png. Be aware that the rm command can not be undone, so use with caution / knowledge. It is safer, until you are more comfortable to use the Finder application and move the file to the trash. In Finder, you can show hidden directories by typing ⌘. or go directly to the folder by typing ⇧⌘G (shift + command + G) and in the pop-up typing ~/.local/bin.

That is it for this article. In the upcoming articles we will setup our shell environment so that we can use the commands we’ve built without having to navigate to the ~/.local/bin directory. Thank you for reading to the end, I hope you’re finding this series helpful.