The SwiftIO board
MadMachine‘s SwiftIO board is an Arduino-like system, however it might probably run Swift code natively on the {hardware}. Swift is a good programming language for training functions, it’s a trendy language with a JavaScript-like syntax that’s straightforward to study and perceive. Swift is secure and environment friendly, this mixture makes it an ideal candidate for embedded techniques. The SwiftIO board permits us common Swift / iOS builders to construct IoT tasks utilizing our favourite programming language.
The board itself has a small footprint, it is nearly 1,57″ x 2,36″ (4cm x 6cm) huge.
It is fairly a robust microcontroller. It’s positively essentially the most reasonably priced Swift language studying system. It may be superb for STEM training functions for college students who wish to see learn how to construct digital circuits and get some fundamental Swift programming expertise on the identical time.
The Maker Package field accommodates each a 7 section and an LCD show, a servo and a DC motor with drivers, a humiture (temperature & humidity) sensor, a buzzer module a potentiometer and a light-weight sensor and plenty of extra different issues (breadboard, resistors, LEDs, buttons and cables, and so on.) that you should utilize to construct the starter tasks. You’ll have just about all the pieces within the field that you simply want as a newcomer to get began and have some enjoyable time utilizing SwiftIO. 📦
Sadly you will not discover a working GPS, WiFi or Bluetooth module for the SwiftIO board but, however hopefully these form of equipment are just some months away. There’s a new defend and a model new display screen module within the works. Oh by the best way a defend is one thing which you could plug into your board and join exterior equipment utilizing “particular” colourful cables. 🌈
Shields are boards that may be plugged on high of the Arduino PCB extending its capabilities. The completely different shields observe the identical philosophy as the unique toolkit: they’re straightforward to mount, and low cost to supply. – ArduinoShields
The board initially went on sale in July, 2020 and many individuals introduced it from all world wide. New plug-ins, extensions and modules are nonetheless being developed by the makers, however it is very important emphasize that the SwiftIO board {hardware} is in a secure state. 💪
Technical specs:
- i.MX RT1052 Crossover Processor with Arm® Cortex®-M7 core @600MHz
- Micro SD card slot, supporting customary and excessive capability SD playing cards
- Micro USB connector for energy
- On-board USB to UART for serial communication
- 46 GPIO on left and proper edges
- On-board GRB LED
- 12x 12-bit analog to digital (ADC) converters
- 4 UART, 2 CAN, 2 IIC, and a couple of SPI
- 14 PWM pins
- Many extra superior options to satisfy the wants of superior customers
Ports, communication, structure
You may join the SwiftIO board to your pc by way of a USB connector, the 2 gadgets can talk via a serial port. The USB cable will present the mandatory energy assist, however alternatively you should utilize an adapter or an exterior battery via a defend.
You may wire up extra parts utilizing the Common Goal Enter/Output pins (GPIO). You may see the precise specification on the image above, however actually I barely perceive this diagram.
Confession time: I do not know shit about electronics (but). 💩
The board can be utilized with a defend and thankfully the MakersKit arrives with a helpful instruction guide for absolute learners. I felt fortunate, as a result of I used to be capable of strategy this little gadget with my programmer mindset and I might focus extra on Swift as a substitute of constructing working circuits.
So what’s truly beneath the hood? How can I construct & run purposes for SwiftIO?
The grey space is the {hardware} itself, on high of that within the pink-ish/purple packing containers there may be this Zephyr layer which I’ve by no means heard earlier than. I am simply scratching the floor right here, however that is wonderful, since as a Swift developer we solely care in regards to the orange stuff. Lengthy story quick, there’s a customized constructed Swift toolchain for this system that enables us to construct and run Swift purposes on the board. 😅
If there’s a toolchain, then we are able to run Swift binaries, however how can we talk with the ports and stuff like that? Fortuitously the SwiftIO framework is right here for us. It supplies an easy accessibility to speak with exterior {hardware} equipment. You may learn or write digital and analog alerts utilizing the communication protocol. Are you prepared for some Swift code?
The SwiftIO setting
Though the {hardware} is secure, the software program just isn’t completed but. There are two choices obtainable, however in case you are on a Home windows machine, it is best to seize the official MadMachine IDE and use that. The reason being that the Swift Package deal Supervisor just isn’t prepared for that platform but, so you will not be capable of work with the instruments that I will present you subsequent. Sorry Home windows, no offense. 😅
So choice A, is to go together with the IDE, it is very easy to make use of and supplies a terrific DX. Choice B, study a bit extra in regards to the underlying toolkit and observe my information, it’s kind of extra difficult, however you may know extra in regards to the technical infrastructure in case you select this path. 🤔
Putting in the MadMachine SDK & CLI
As a macOS or Linux person it is best to know which you could set up the unofficial MadMachine SDK alongside a command line device referred to as mm. As you possibly can see this MadMachine SDK is written in Swift, however it is best to know that I ported a python script from the unique mm-sdk mission with the assistance of Marcus Kida. First, he made an Xcode template by invoking the unique script contained in the official MadMachine IDE, then I assumed, it could be cool to eliminate the IDE for good, and now, right here we go, now we have a standalone Swift model of the mandatory construct instruments. You may set up it like this:
git clone https://github.com/EmbeddedSwift/MadMachine.git
cd MadMachine
make set up
Now try to be prepared to make use of the mm cli app, you possibly can test the obtainable instructions contained in the README file on GitHub. I’ve not examined this on Linux but, so in case you discover any points, please do not hesitate to report or submit a PR. That is an experimental mission, so hold this in thoughts. 😅
The MadMachine toolchain
With the intention to use the MadMachine SDK you will have a working toolchain put in in your pc. You may seize the newest one by operating the next command:
mm toolchain --upgrade
It will seize the newest launch from the unofficial MadMachine toolchain repository, and place it beneath your house folder contained in the .MadMachine listing. There’s one extra factor that now it’s a must to do earlier than you possibly can begin constructing SwiftIO apps. At the moment there may be one further python script that was not ported but, as a result of it will likely be fully eradicated sooner or later. For now you continue to should obtain the official MadMachine SDK from the GitHub releases web page and place your complete contents of the mm-sdk/tools_[platform]/scripts/dist/gen_isr_tables folder into the ~/.MadMachine/legacy listing. You might need to create a legacy folder. 🙈
The SwiftIO framework
The SwiftIO framework goes to be linked with the applying binary, now we have to put in it (with the assistance of the mm-cli device) as a system library first. There’s an unofficial repo with a makefile for this:
git clone https://github.com/EmbeddedSwift/SwiftIO
cd SwiftIO
make set up
You could find the library reference for the SwiftIO framework, however we’ll see you may make it work, in just some moments. Earlier than we go additional it is best to notice that the customized MadMachine toolchain is a modified model of the Swift 5.1 toolchain. This implies which you could’t use Swift 5.3 on the board but, however hopefully the creators of SwiftIO will launch new software program parts actual quickly. 🤞
Utilizing Xcode
The SwiftIO framework might be compiled in your native machine with the native Swift toolchain (utilizing Xcode), so it’s doable to construct purposes with out concentrating on the board, and afterward you possibly can re-compile the supply recordsdata with the mm-cli command, signal the ultimate binary and deploy it to the SwiftIO board after you’ve got pressed the obtain button. That is the present workflow in a nutshell.
There’s an present Xcode template created by @kidmar that you should utilize as a place to begin.
Utilizing SPM
These days I favor to create a Swift bundle for nearly all the pieces. You should use SPM with a Makefile and your favourite editor to create SwiftIO apps. You simply should initialize a brand new executable bundle with the mandatory dependencies, for instance:
import PackageDescription
let bundle = Package deal(
identify: "myProject",
merchandise: [
.executable(name: "myProject", targets: ["myProject"]),
],
dependencies: [
.package(url: "https://github.com/EmbeddedSwift/SwiftIO", .branch("main")),
.package(url: "https://github.com/EmbeddedSwift/SHT3x", .branch("main")),
.package(url: "https://github.com/EmbeddedSwift/LCD1602", .branch("main")),
],
targets: [
.target(name: "myProject", dependencies: [
.product(name: "SwiftIO", package: "SwiftIO"),
.product(name: "SHT3x", package: "SHT3x"),
.product(name: "LCD1602", package: "LCD1602"),
]),
]
)Contained in the predominant.swift file now you possibly can write a easy Humiture sensor app that shows the present humidity & temperature on a 16×2 LCD show like this:
import SwiftIO
import SHT3x
import LCD1602
extension Float {
func format(_ f: Int) -> Float {
guard f > 0 else {return self}
var mul = 10
for _ in 1..<f {
mul *= 10
}
let knowledge = Int(self * Float(mul))
return Float(knowledge) / Float(mul)
}
}
let i2c = I2C(Id.I2C0)
let liquid crystal display = LCD1602(i2c)
let sht = SHT3x(i2c)
whereas true{
let temp = sht.readTemperature()
liquid crystal display.write(x: 0, y: 0, "Temperature:")
liquid crystal display.write(x: 0, y: 1, String(temp.format(1)))
liquid crystal display.write(x: 4, y: 1, " ")
liquid crystal display.write(x: 5, y: 1, "C")
sleep(ms: 1000)
}Now in case you open the Package deal.swift file utilizing Xcode you possibly can even construct the mission domestically utilizing the CMD+B shortcut, however do not attempt to run it, since your Mac just isn’t a SwiftIO board… 😅
If you wish to construct and run this mission it’s a must to goal the SwiftIO board. The mm-cli cannot resolve bundle dependencies (but) so it’s a must to set up the mandatory dependencies (SHT3x, LCD1602) the identical manner as we did with the SwiftIO library. You need to clone each repositories and use the makefile to put in them as native MadMachine system libraries. It will transfer the required recordsdata beneath the ~/.MadMachine/libraries folder, all of the obtainable libraries “reside” there. 🔨
After you’ve got put in the libraries, you should utilize the next Makefile for the precise binary builds:
construct:
mm construct --name myProject --binary-type executable --input . --output ./myProject
run: construct
mm board --run ./myProject/swiftio.bin
clear:
rm -r ./myProject
The mm construct command invokes the swiftc compiler from the SwiftIO toolchain with the appropriate search paths and library search paths. The mm board --run [binary] command will merely copy the mandatory recordsdata to the MadMachine board whether it is in obtain mode. You need to press the obtain button on the board earlier than you run this command. Urgent the obtain button will mount the SD card so we are able to deploy the signed binary to the system, then eject the storage, which is able to restart the board and the brand new utility might be loaded & executed straight away. 🚀
The MadMachine IDE
When you do not prefer to play with command line utilities, you possibly can all the time use the MadMachine IDE to construct and ship your tasks. It makes use of a mmp file to explain dependencies, so it might probably resolve them on the fly, however I do not like the truth that I’ve to make use of a brand new bundle supervisor and editor to work with Swift. I am in a love and hate relationship with Xcode (plus I am an old-school Chic Textual content fan). ❤️
Anyway, you could find actually nice examples with a devoted GitHub repository and plenty of video tutorials on the official YouTube channel. You may entry all of the MakerKit pattern codes, they’re each on GitHub and you’ll clone a template with a click on utilizing the MadMachine IDE.
Conclusion
For me, the SwiftIO board was a pleasing shock. I all the time needed to play with embedded techniques, know slightly extra about {hardware} and low stage stuff, however the Raspberry PI and the Arduino felt like alien planet. As soon as I’ve tried to construct an app for a Raspberry PI at a hackaton, but it surely was an entire catastrophe, I used to be upset, as a result of the instruments and the developer expertise was actually unhealthy.
The promise of MadMachine is which you could simply use Swift on such a tool and I imagine that Andy Liu created one thing that may have a shiny future on the long run. I felt in love with SwiftIO proper after I’ve assembled my first circuit and deployed my very first Swift supply. It was a seamless (ache free) expertise and I imagine that is a extremely essential issue for builders who by no means used such gadgets earlier than. This may be the start and the way forward for Swift on embedded techniques. 😍
You may construct some actually enjoyable stuff with SwiftIO, it is an ideal device for studying how electrical circuits work. I am unable to wait to see an increasing number of sensors and libraries obtainable for MadMachine. If in case you have an concept or a query be at liberty to hitch the official Discord server.