Created as The Advanced Project for MSBI 31600 Winter 2022 Muhammad Aji Muharrom
The main function of this simple application is to enable Google Chromecast-enabled devices to play certain verses of the Qur'an independently. To put simply, the Qur'an can be thought of as an audiobook that contains several chapters, where each chapter contain variable number of verses. With this application, a user can choose certain verses and play them on a Cast device connected on the same local network.
I create this little app for my own use, as I own a Google Home Mini and want this capability. However, I could not find a readily available solution. Existing solutions include mobile applications which can play certain verses on the device. To play it on the cast, I have to set the Cast as a bluetooth speaker and connect it to my device. With this app, I do not need to keep my device connected to the Cast, because it just sends instructions and the Cast is the one who downloads and plays the audio files.
The class themes I utilized in this project includes GUI (which in itself pertains a whole bunch of OOP programming), OOP in using external library, and a little Networking.
What I would do differently, or rather, what I would improve, is certainly the user interface. I would love to have a more pleasant app doing this capability. I was considering using Django and create the interface using HTML and all the responsive web good stuff, but ended up trying Kivy instead because I don't want the hassle of running a web server endlessly and want to simply have an app I can run on my phone. If I am to redo this project, I would probably try to use the "Kv Language", some sort of layouting language that kivy provides to simplify the code for UI layouts.
First, make sure all dependencies are available. This project is made and tested using Python 3.9.7, so no guarantee it will run on other python versions. To be safe, create a conda environment using this version and activate it.
conda create -n py397-qcast python=3.9.7
conda activate py397-qcast
I believe kivy and pychromecast are the only external libraries I used here,
other libraries such as requests were part of Python 3 standard distribution.
So first, install kivy by running conda install -c conda-forge kivy=2.0.0
Then, we need to install pychromecast. Unfortunately, it's not available in conda, so we need to use pip
conda install pip
pip install pychromecast==10.3.0
Then, on the project directory, execute python app.py to run the application.
Once the app runs, click "Discover" to start the device discovery process. Cast-enabled device(s) found will appear on the selector. Select the device and click "connect" to use the device.
After chapters and verses are selected, click "Play" to play the selected verses on the device!
Note: I included requirements.txt as part of this project, which includes all the packages needed to run this app. However, I could not find an easy way to create a conda environment out of the requirements due to the package(s) availability -- I could not find a single conda repository that has every requirement.
Among the challenges that I have to overcome during the project is understanding the external libraries.
Though pychromecast is powerful, it lacks documentation. Well, to be fair, even Google's Cast SDK docs are not that complete. However, pychromecast does include a bunch of examples which were really useful to learn from.
Kivy has better documentation, though not as good as perhaps tkinter. However, it is enough to at least get the app to work!
The Quran.com team provides a well-documented API.
Acknowledgements thus go to the authors and contributors to these cool libraries and APIs.
References:
- https://github.com/home-assistant-libs/pychromecast
- https://github.com/home-assistant-libs/pychromecast/tree/master/examples
- https://kivy.org/doc/stable/gettingstarted/intro.html
- https://quran.api-docs.io/v4/getting-started/introduction
Another challenge and learning I got from this project was doing asynchronous programming in Python. Having some experiences with Javascript, I would say that async programming requires more work in Python compared to JS.
Extra credits I attempted are the use of interesting external library - kivy, and also concepts slightly covered in class: threading. I tried to implement more asynchronous programming to avoid the application hangs while certain processes are being executed.
I also attempted try..except error checking, one of them is inside the DeviceComponents.on_button_play_release() method. If the "Repeat count" selector value has not been set, thus raising ValueError when casted, it will catch the error and assign the value to "1" before proceeding.