Get a Flashing Meeting Reminder with a Raspberry Pi

Figure 1 shows the parts I used for this project (minus the power supply). The Unicorn HAT on the left is facing down so you can easily see the female headers on the back of the board that match the male Pi’s GPIO headers.

Flip the Unicorn HAT over and align the GPIO Pins with the female header socket on the HAT as shown in Figure 2

With the HAT properly positioned, push down gently to seat the board securely on the Raspberry Pi as shown in Figure 3.

Right now, the HAT is floating free over the Pi, connected only through the header pins. Since the whole contraption will be sealed in a case, I didn’t feel the need to mount the HAT any more securely, but you may want to. Adafruit sells a set of spacers  you can use to secure the other side of the HAT against the Pi.

Next, install the Pi and HAT in the case as shown in Figure 4 and attach the cover.

At this point, the hardware is all installed and ready to go. When finished, the Pi will only need network and power connections, but for the remainder of the setup process, you’re going to need a display (monitor), keyboard, and mouse. Don’t connect power, you have some more work to do before you can do that.

The Raspberry PI needs an OS to boot, so grab Micro SD card then follow the instructions to download and install the OS on the SD card. Once that’s completed, insert the SD card in the Pi.
At this point, you’re ready to power up the Pi. Turn on your monitor then plug the power supply into a power outlet and the Micro USB port on the Pi (upper right corner of the Pi device shown in Figure 5).

At this point, you have a few steps to perform. The first step is to configure some system-wide settings on the Raspberry Pi. You could get away without doing this, but you may run into problems later if you don’t compete these steps. To start, open the Raspberry menu in the upper left corner of the Pi screen. In the menu that appears, select Preferences then select Raspberry Pi Configuration as shown in Figure 6.

By default, the default Raspbian image only utilizes a part of the SD card’s storage (3 GB); if the SD Card you used for your Raspbian installation is larger than 3 GB, you’ll need to expand the file system to use the whole disk. If you don’t do this, your Pi’s storage system may not have enough room to install of the additional software this project utilizes.

In the Raspberry Pi Configuration utility shown in Figure 7, click the Expand Filesystem button to enable the Pi to use the whole SD card for storage. Making this change will require rebooting your Pi, so don’t be surprised.

If you want, use this opportunity to change the host name for the Pi device. In Figure 7 you can see that I’ve named mine pi_remind so I can easily find the device on the network later. If you’re a UK-based reader, you’re done, click the OK button and let the Pi reboot. The Pi comes pre-configured for UK keyboard and other locale settings, so you’re good to go.

As a US-based hacker, I find that the default settings don’t work for me. Switch to the Localisation tab and check all of the settings exposed there to ensure its properly configured for your locale, time zone and keyboard settings. Click the OK button when you’re done to save your changes. Don’t be surprised if you’re prompted to reboot the Pi.

Next you’ll need to update the Pi’s core software and install some additional software needed to communicate with the Unicorn HAT and Google Calendar.

After the Pi reboots, open a terminal window and issue the following command:
sudo apt-get update

This command will update the Pi’s local repository of software update repositories. When that command completes, issue the following command:
sudo apt-get upgrade

This command performs a complete update of the Raspbian OS and associated programs to the latest version. This process will take quite a while, but the benefit is that at the end of the process, your Pi is running the latest and greatest supported version of all apps and code for the device.

Note: When it comes to writing code, the Raspberry Pi is too slow and the available screen resolution to small for me to work effectively. For that reason, I prefer to code on my desktop and upload the code to the Pi for testing. I coded the Python app for this project using the PyCharm IDE . The commercial version of the IDE enables me to remotely deploy the app to the pi device; you can read how I did it here. In order to be able to push code updates to the Pi, you’ll need to figure out the Pi’s IP address, the referenced article has instructions on how to do so. To simplify this on your home network, you may want to set a static IP address on the Pi so the IP address won’t change if you switch routers or network providers.

The Unicorn HAT uses its own python libraries; details can be found on the Pimoroni web site. To install the libraries on the Pi, open a terminal window and execute the following command:
sudo curl -sS get.pimoroni.com/unicornhat | bash

During installation, you’ll be asked whether you want to also install sample code and additional software; this project doesn’t make use of any of them, but you can install them just to see what else they give you.

Next, you’ll need to install the Google Calendar API client code. You can find complete instructions here. Basically just open a terminal window and execute the following command:
sudo pip install --upgrade google-api-python-client

Finally, there’s a set of scientific libraries that are used in some of the Unicorn HAT code in the project, so you will need to install them using the following command:
sudo apt-get install python-numpy

The project’s Python source code is available here. You can clone the files using Git or download the files using a browser then extract the files to a folder on the Pi’s file system. When I setup the project for the first time, I opened a new terminal window (to make sure I was pointing to the pi user’s home folder) and executed the following commands:
mkdir pi_remind
cd pi_remind
wget https://github.com/johnwargo/pi_remind/archive/master.zip
unzip -j master.zip

This creates a folder for the project’s code then pulls down the files and extracts the files to the folder. When everything finishes, you should see a file called remind.py (along with some other files) in the folder. Make note of the folder location you used for the project files, you’ll need the folder path later.

Before the project’s software can access your Google Calendar, you need to setup a developer account with Google (if you don’t already have one) and define an application in the Google Developer Console that makes use of the Google Calendar APIs. Start by opening a browser window and navigating https://console.developers.google.com. Login or create an account then look for the developer console shown in Figure 8.

Click the link for Calendar API highlighted in the figure and Enable the API then define an application for this project. Once that’s done, you’ll need to create credentials for your project as shown in Figure 9.

Once your credentials are ready, click the Download JSON button highlighted in the figure. Download the file to the folder where you extracted the python project files in the previous section (in my example, it’s the pi user’s home folder in a subfolder called pi_remind. When you save the file, Google will automatically assign a long file name including the unique identifier for the project. For this project, save the file with the file name client_secret.json.

With the client secret file in place, you’ll have to authorize the Pi to access your calendar. Open a terminal window, navigate to the folder where you extracted the project files and issue the following command:
sudo python ./remind.py
Using the folder example from earlier, you would issue the following commands:
cd pi_remind
sudo python ./remind.py
If everything is setup correctly, the python code will execute and show a screen similar to the one shown in Figure 10.

At this point, the application will open the default browser and display the page shown in Figure 11.

Login to your Google account and click the Next button. If you’re like me, you probably have multiple Google accounts, so be sure to provide the email address for your primary Google Calendar account.
After you provide your password, the browser will prompt you to authorize the project’s Python application to access your calendar as shown in Figure 12. The application name, Pi Reminder in this case, will be the same one you created in the Google Developer Console just a few steps back.

After you complete the authentication process, the browser will display a page similar to the one shown in Figure 12. At this point you can close the browser window.

The python application may quit after the authentication process completes, so you may have to restart the python application by executing the following command:
sudo python ./remind.py

The easiest way to execute the command again is to simply press the keyboard’s up arrow key and the previous command should pop onto the command line for you to execute.
At this point, the application should connect to the authorized calendar and start checking for appointments as shown in Figure 13.

To see a video of the app in action, see:

If you’re like me, you don’t want to have a keyboard, mouse and display tied up to run this little reminder app. The project is designed to run unattended, all you have to do to enable this is setup the remind.py Python script to execute when the Pi powers up. There are a lot of options described online to enable this. Here is one option and another. The one I used for my project is here.

The problem with each approach is that the app runs in the background; I couldn’t find any option (that worked anyway) that enabled me to open a terminal window at startup and execute the project’s code. See, I care about seeing the app running in a terminal window in case something goes wrong. When things fail, I want to be able to plug a display, keyboard and mouse into the Pi and see what’s going on. Now, I know I can setup an external log file and let the program write to that – then open the file remotely to see what’s happening, but that’s extra work I didn’t want to have to do.

Anyway, I spent an hour or so trying different approaches and didn’t find any that worked for me. If I figure it out, I’ll update the project’s readme and/or perhaps post an update to my blog .
For now, you can either run the program manually and unplug all the extra hardware once it’s running or you can follow the instructions in the links I provided to enable auto startup of the code. To help you, I added a pi_remind.desktop file to the project which is a sample implementation of one of the examples. I also created a shell script (start_remind.sh) and included it with the project’s files in case you want to use it for one of the other approaches.

Now, when implementing this, I just used a simple flashing pattern that lit all of the LEDs using a single color. That’s not very exciting as you’ll likely see when you get this project up and running. The Unicorn HAT sample Python code can be found here and has code you can switch into the app in order to make the lights flash more interestingly.

Right now, the code is as simple as I could make it. When the Pi checks for calendar entries, it doesn’t check to see if the appointment has a reminder set. For some placeholder appointments or less important appointments, you might not want to be reminded. So, one enhancement I plan on making is to check the appointment to see if I need alerting. Perhaps I’ll code that and add it to the project on GitHub.