Finally, something that is missing from the start, the ability to write some text on display. In Doton we needed an ability to display numbers and we have done it via images. It was a partial solution but good enough. But we want more 🙂
So this time fonts, text and another step in endless struggle with displaying anything anywhere 🙂
We have a driver for ILI9486 320×480 display and we can run Doton on it. But currently, it would use only half of the screen. Our window manager has hard-coded dimension 240×320.
We need to do something with it. I think we can update manager to calculate grid from dimensions and additionally we can add tiles auto positioning.
Why? Because it seems that not setting position on the screen can be better 🙂 System can add pages when required and take care of horizontal vs vertical position.
But I think that we should be able to set coords if we want to.
Source code @ GitHub
My friend suggested me that I should write Doton in a .NET. A Huge challenge, I never thought about it but maybe we can do some research in this matter?
I installed Windows 10 IoT edition on my Raspi, started Visual Studio 2015 and.. stopped 😀 I had no idea what to do. Googling and I found something that was beautiful and crazy at the same time.
It was an information that we can use a Python… So maybe we can port a project to Windows 10 IoT? Let’s see!
We can display only 6 tiles on the screen. It is not too much. But when we redesigned GUI we prepared touch spots to scroll between pages.
A page is a group of maximum 6 tiles so one page fits on the screen. And with a few pages, we can have much more information available.
I’m little afraid of inaccuracy of touch panel. Why? See that scroll down area overlaps our two widget buttons. We may be forced to move them up to the middle row. But we will see 🙂
We have a touch driver in our GfxLCD package so we can upgrade Doton a little. We will add clickable widgets.
Our window managers will handle touch propagation. For the first widget with action, we will create a light switcher.
Source @ GitHub
We still have the weather widget to finish. But before that, I’m thinking about big changes in tile layout. Why?
Because tiles are rectangles and they cannot be used in vertical mode, square tiles would be much better.
This should give us a free space on top and bottom of the display for a nice functionality in near future (do not forget that out TFT is a touch screen) 🙂
Let’s get to work!
Source @ GitHub
Long, long time ago in a post far, far away we made a weather worker (part 1, part 2). It is time to find it and add to our project along with new widget.
And as a side effect, we will sugarcoat GUI a little bit.
Doton part 1
Doton part 3
Source code @ GitHub
We may now return to the main task, the Control Node with all data from all sources. We will present data in form of a tile.
I’m thinking about creating widgets for both Node One sensors.
Each node is the one tile. We could present light and movement with small icons and humidity and temperature with numbers. It sounds nice and simple but there is a catch. If we refresh the whole screen it would take around 5s to change its content. So we must find a way of repainting only required parts of the screen.
Part 2 – weather widget
Part 3 – window manager
Doton project @ GitHub
Docker, Docker everywhere… you open the fridge and.. docker with butter and cheese, Docker may save the day but it may also make it horrible… But it doesn’t matter if you like it or not. Docker is here, its containers are everywhere. So lets join the flow 🙂
We will dockerize ProxyLCD in CLI mode. No more running the app manually:) Laziness !
Our previous computer app was a failure. Happens from time to time:) Choosing wxPython was a huge mistake. Seems wx bindings are outdated.
After a long walk to cool of my head and think a little (but mostly catching pokemons) I recovered my strength and began second approach. This time I’m using PyQt5. It has Python3 bindings and lots of nice features, like ability to emit signals to other objects.
Spoiler: this time a full success 🙂
Get source from GitHub