About Enclosures

The Mark 1 Device is housed in an Enclosure. An Enclosure is a shell that houses a Device that runs Mycroft. In the future, different Enclosures will have different capabilities.

The Mark 1 Enclosure capabilities

The Mark 1 mouth and eyes can be controlled by Skills using the self.enclosure object inherited from the MycroftSkill base class. This object acts as an interface to the Enclosure and allows the Skill creator to draw to the mouth display.

This is how the mouth and eyes are made to change during operations such as audio playback.

Drawing to the mouth display

Drawing text to the mouth display

Text can be sent to the display using the mouth_text() method of the enclosure object.

        self.enclosure.mouth_text('The meaning of life, the universe and everything is 42')

If the text is too long to fit on the display, the text will scroll.

@TODO how many characters will fit on the display before it will scroll?

Drawing images to the mouth display

Clearing an existing image from the mouth display

Before writing an image to the mouth display, you should clear any previous image.

        self.enclosure.mouth_display(img_code="HIAAAAAAAAAAAAAA", refresh=False)
        self.enclosure.mouth_display(img_code="HIAAAAAAAAAAAAAA", x=24, refresh=False)

Copied from the skill-date-time

How images are drawn on the mouth display

The mouth display is a grid, 32 pixels wide and 8 pixels high.

There are two ways to draw an image on the mouth display.

Addressing each pixel using a string encoding

You can draw an image to the mouth display by binary encoding pixel information in a string.

The binary encoding is straightforward value substitution.

and so on.

        self.enclosure.mouth_display(img_code="HIAAAAAAAAAAAAAA", refresh=False)
        self.enclosure.mouth_display(img_code="HIAAAAAAAAAAAAAA", x=24, refresh=False)

The code above clears the image by sending a string consisting of HI which stands for a Width of 7 and a height of 8 and each A stands for a segment of 4 pixels in the off state.

@TODO I kinda get this, but I think we really need a grid image here to show how it works - to make it easier to understand.

Sending a PNG image to the mouth display

Another way to draw an image on the mouth display is to create a PNG-formatted image with a width of 32 pixels and a height of 8 pixels, then use the mouth_display_png() method of the enclosure object.

The image should be black and white, with white meaning a dark pixel, and black indicating an illuminated pixel.

mouth_display_png() expects the first argument to be the image absolute path. Optional arguments are

• threshold: The value at which a pixel should be considered 'dark' or 'illuminated'

• invert: Treat white in the image as illuminated pixels, and black as dark pixels

• x: The x position (horizontal) at which the image should be displaye, in pixels

• y: The y position (vertical) at which the image should be displayed, in pixels

• refresh: clear the display before writing to it

@TODO all the above needs to be validated - the information is educated guesswork

        self.mouth_display_png('/path/to/image.png', threshold=70, invert=False, x=0, y=0, refresh=True)

Example image: ​

Tools for converting PNG image representations to string representations

If you don't want to convert PNG files at runtime (for example when creating simple animations) this short python script will convert PNG files to strings compatible with the img_code of self.enclosure.mouth_display().

Resetting the display to the default state

When the Skill is finished, you should reset the Enclosure to the default state using

        self.enclosure.reset()

This will clear the screen and blink the Mark 1's eyes once.