Display Component

The display component houses ESPHome’s powerful rendering and display engine. Fundamentally, there are these types of displays:

For the last type, ESPHome has a powerful rendering engine that can do many things like draw some basic shapes, print text with any font you want, or even show images.

To achieve all this flexibility displays tie in directly into ESPHome’s lambda system. So when you want to write some text or sensor values to the screen you will be writing in C++ code using an API that is designed to

  • be simple and to be used without programming experience

  • but also be flexible enough to work with more complex tasks like displaying an analog clock.

Display Rendering Engine

In this section we will be discussing how to use ESPHome’s display rendering engine from ESPHome and some basic commands. Please note that this only applies to displays that can control each pixel individually.

So, first a few basics: When setting up a display platform in ESPHome there will be a configuration option called lambda: which will be called every time ESPHome wants to re-render the display. In each cycle, the display is automatically cleared before the lambda is executed. You can disable this behavior by setting auto_clear_enabled: false. In the lambda, you can write code like in any lambda in ESPHome. Display lambdas are additionally passed a variable called it which represents the rendering engine object.

display:
  - platform: ...
    # ...
    lambda: |-
      // Write your display rendering code here

      // For example, draw a line from [x=0,y=0] to [x=50,y=50]
      it.line(0, 0, 50, 50);

Note

Lambdas are essentially just a lightly modified version of C++. So don’t forget to end each line with a semicolon (;). Otherwise you will be greeted by a long error message at the compilation stage.

If you compile and upload the configuration above, you should see a black (or white, depending on the display) line which starts at the top left and goes a few pixels down at a 45° angle. (If it’s in another corner, use the rotation: option to rotate the display to your liking)

This already highlights one of the things you must learn before diving into writing your own custom display code: The top left is always the origin of the pixel coordinate system. Also, all points in this coordinate system are a pair of integers like 50, 50 which represent the shift to the right and shift downwards. So, in other words, x always represents the horizontal axis (width) and y the vertical axis (height). And the convention in the rendering engine is always first specify the x coordinate and then the y coordinate.

Basic Shapes

Now that you know a bit more about ESPHome’s coordinate system, let’s draw some basic shapes like lines, rectangles and circles:

display:
  - platform: ...
    # ...
    lambda: |-
      // Draw a line from [0,0] to [100,50]
      it.line(0, 0, 100, 50);
      // Draw the outline of a rectangle with the top left at [50,60], a width of 30 and a height of 42
      it.rectangle(50, 60, 30, 42);
      // Draw the same rectangle, but this time filled.
      it.filled_rectangle(50, 60, 30, 42);

      // Circles! Let's draw one with the center at [25,25] and a radius of 10
      it.circle(25, 25, 10);
      // ... and the same thing filled again
      it.filled_circle(25, 25, 10);

All the above methods can optionally also be called with an argument at the end which specifies in which color to draw. For monochrome displays, only COLOR_ON (the default if color is not given) and COLOR_OFF are supported.

display:
  - platform: ...
    # ...
    lambda: |-
      // Turn the whole display on.
      it.fill(COLOR_ON);
      // Turn the whole display off.
      it.fill(COLOR_OFF);

      // Turn a single pixel off at [50,60]
      it.draw_pixel_at(50, 60, COLOR_OFF);

      // Turn off a whole display portion.
      it.rectangle(50, 50, 30, 42, COLOR_OFF);

For color displays (e.g. TFT displays), you can use the Color class.

display:
  - platform: ...
    # ...
    lambda: |-
      auto red = Color(255, 0, 0);
      auto green = Color(0, 255, 0);
      auto blue = Color(0, 0, 255);
      auto white = Color(255, 255, 255);
      it.rectangle(20, 50, 30, 30, white);
      it.rectangle(25, 55, 30, 30, red);
      it.rectangle(30, 60, 30, 30, green);
      it.rectangle(35, 65, 30, 30, blue);

Additionally, you have access to two helper methods which will fetch the width and height of the display:

display:
  - platform: ...
    # ...
    lambda: |-
      // Draw a circle in the middle of the display
      it.filled_circle(it.get_width() / 2, it.get_height() / 2, 20);

You can view the full API documentation for the rendering engine in the “API Reference” in the See Also section.

Fonts

The rendering engine also has a powerful font drawer which integrates seamlessly into ESPHome. Whereas in most Arduino display projects you have to use one of a few pre-defined fonts in very specific sizes, with ESPHome you have the option to use any TrueType (.ttf) font file at any size, as well as fixed-size PCF and BDF bitmap fonts! Granted the reason for it is actually not having to worry about the licensing of font files :)

To use fonts you first have to define a font object in your ESPHome configuration file. Just grab a .ttf, .pcf, or .bdf file from somewhere on the internet and place it, for example, inside a fonts folder next to your configuration file.

Next, create a font: section in your configuration:

font:
  - file: "fonts/Comic Sans MS.ttf"
    id: my_font
    size: 20

  # gfonts://family[@weight]
  - file: "gfonts://Roboto"
    id: roboto
    size: 20

  - file:
      type: gfonts
      family: Roboto
      weight: 900
    id: font2
    size: 16

  - file: "fonts/tom-thumb.bdf"
    id: tomthumb

display:
  # ...

Configuration variables:

  • file (Required): The path (relative to where the .yaml file is) of the font file. You can use the gfonts:// short form to use Google Fonts, or use the below structure:

    • type (Required, string): Can be gfonts or local.

      Google Fonts:

      Each Google Font will be downloaded once and cached for future use.

    • family (Required, string): The name of the Google Font family.

    • weight (Optional, enum): The weight of the font. Can be either the text name or the integer value:

      • thin: 100

      • extra-light: 200

      • light: 300

      • regular: 400 (default)

      • medium: 500

      • semi-bold: 600

      • bold: 700

      • extra-bold: 800

      • black: 900

    • italic (Optional, boolean): Whether the font should be italic.

      Local Fonts:

    • path (Required, string): The path (relative to where the .yaml file is) of the TrueType or bitmap font file.

  • id (Required, ID): The ID with which you will be able to reference the font later in your display code.

  • size (Optional, int): The size of the font in pt (not pixel!). If you want to use the same font in different sizes, create two font objects. Note: size is ignored by bitmap fonts. Defaults to 20.

  • glyphs (Optional, list): A list of characters you plan to use. Only the characters you specify here will be compiled into the binary. Adjust this if you need some special characters or want to reduce the size of the binary if you don’t plan to use some glyphs. The items in the list can also be more than one character long if you for example want to use font ligatures. Defaults to !"%()+=,-_.:°0123456789ABCDEFGHIJKLMNOPQRSTUVWXYZ abcdefghijklmnopqrstuvwxyz.

Note

To use fonts you will need to have the python pillow package installed, as ESPHome uses that package to translate the TrueType and bitmap font files into an internal format. If you’re running this as a Home Assistant add-on or with the official ESPHome docker image, it should already be installed. Otherwise you need to install it using pip install pillow.

Drawing Static Text

In your display code, you can render static text by referencing the font and just entering your string:

display:
  - platform: ...
    # ...
    lambda: |-
      // Print the string "Hello World!" at [0,10]
      it.print(0, 10, id(my_font), "Hello World!");

By default, ESPHome will align the text at the top left. That means if you enter the coordinates [0,10] for your text, the top left of the text will be at [0,10]. If you want to draw some text at the right side of the display, it is however sometimes useful to choose a different text alignment. When you enter [0,10] you’re really telling ESPHome that it should position the anchor point of the text at [0,10]. When using a different alignment, like TOP_RIGHT, the text will be positioned left of the anchor pointed, so that, as the name implies, the anchor point is a the top right corner of the text.

display:
  - platform: ...
    # ...
    lambda: |-
      // Aligned on left by default
      it.print(0, 0, id(my_font), "Left aligned");

      // Aligned on right edge
      it.print(it.get_width(), 0, id(my_font), TextAlign::TOP_RIGHT, "Right aligned");

As with basic shapes, you can also specify a color for the text:

display:
  - platform: ...
    # ...
    lambda: |-
      // Syntax is always: it.print(<x>, <y>, <font>, [color=COLOR_ON], [align=TextAlign::TOP_LEFT], <text>);
      it.print(0, 0, id(my_font), COLOR_ON, "Left aligned");

Formatted Text

Static text by itself is not too impressive. What we really want is to display dynamic content like sensor values on the display!. That’s where printf comes in. printf is a formatting engine from the C era and ESPHome chose to use because … well, I’m too lazy to create a fully-fledged format engine where the existing stuff is way better documented :)

printf can do way more stuff than you will probably ever need, but it’s also quite simple for the basic stuff. For example, a printf call can look like this:

sensor:
  - platform: ...
    # ...
    id: my_sensor

display:
  - platform: ...
    # ...
    lambda: |-
      it.printf(0, 0, id(my_font), "The sensor value is: %.1f", id(my_sensor).state);
      // If the sensor has the value 30.02, the result will be: "The sensor value is: 30.0"

As you can see, when you call printf most of the string is printed as-is, but when this weird percent sign with some stuff after it is encountered, it is magically replaced by the argument after the format (here id(my_sensor).state).

Every time you type a percent sign % in a printf format string, it will treat the following letters as a format tag until a so-called “specifier” is encountered (in this case f). You can read more about it here, but for ESPHome there are really just a few things you need to know.

Let’s break %.1f down:

  • % - initiate the format string

  • .1 - round the decimal number to 1 digits after the decimal point.

  • f - the specifier which tells printf the data type of the argument. Here it is a f(loat).

For example, if you would like to print a sensor value with two digits of accuracy, you would write %.2f and with zero digits of accuracy (without a decimal) %.0f.

Another interesting format string is %7.2f, which would become the right-justified string "  20.51" for a value of 20.506.

  • % - initiate the format

  • 7 - means that the number will be right-justified and be padded on the left by spaces if the result would be shorter than 7 characters long.

  • .2 - round the decimal number to 2 digits after the decimal point.

  • f - specifier: f(loat).

You can even have as many format strings as you want in a single printf call. Just make sure the put the arguments after the format string in the right order.

display:
  - platform: ...
    # ...
    lambda: |-
      // %% - literal % sign
      it.printf(0, 0, id(my_font), "Temperature: %.1f°C, Humidity: %.1f%%", id(temperature).state, id(humidity).state);

To display a text string from a text_sensor, append .c_str() to the end of your variable.

display:
  - platform: ...
    # ...
    lambda: |-
      it.printf(0, 0, id(my_font), "Text to follow: %s", id(template_text).state.c_str());

The last printf tip for use in displays I will discuss here is how to display binary sensor values. You could of course just check the state with an if statement as the first few lines in the example below, but if you want to be efficient you can use an inline if too. With the %s print specifier you can tell it to use any string you pass it, like "ON" or "OFF".

binary_sensor:
  - platform: ...
    # ...
    id: my_binary_sensor

display:
  - platform: ...
    # ...
    lambda: |-
      if (id(my_binary_sensor).state) {
        it.print(0, 0, id(my_font), "state: ON");
      } else {
        it.print(0, 0, id(my_font), "state: OFF");
      }
      // Shorthand:
      it.printf(0, 0, id(my_font), "State: %s", id(my_binary_sensor).state ? "ON" : "OFF");

Note

For displaying external data on the display, for example data from your Home Assistant instance, you can use the MQTT Subscribe Text Sensor (see the example there for more information).

Displaying Time

You can display current time using a time component. Please see the example here.

Screen Clipping

Screen clipping is a new set of methods since version 2023.2.0 of esphome. It could be useful when you just want to show a part of an image or make sure that what you draw on the screen does not go outside a specific region on the screen.

With start_clipping(left, top, right, bottom); start you the clipping process and when you are done drawing in that region you can stop the clipping process with end_clipping(); . You can nest as many start_clipping(); as you want as long you end them as many times as well.

binary_sensor:
  - platform: ...
    # ...
    id: my_binary_sensor

color:
  - name: my_red
    red: 100%

display:
  - platform: ...
    # ...
    lambda: |-
      if (id(my_binary_sensor).state) {
        it.print(0, 0, id(my_font), "state: ON");
      } else {
        it.print(0, 0, id(my_font), "state: OFF");
      }
      // Shorthand:
      it.start_clipping(40,0,140,20);
      it.printf(0, 0, id(my_font), id(my_red), "State: %s", id(my_binary_sensor).state ? "ON" : "OFF");
      it.end_clipping();

After you started clipping you can manipulate the region with extend_clipping(left, top, right, bottom); and shrink_clipping(left, top, right, bottom); within previous set clipping region.

With get_clipping(); you get a Rect object back with the latest set clipping region.

class Rect {
    int16_t x;  ///< X/Left coordinate
    int16_t y;  ///< Y/Top coordinate
    int16_t w;  ///< Width
    int16_t h;  ///< Height
    int16_t x2();  ///< Right coordinate
    int16_t y2();  ///< bottom coordinate
  };

With is_clipping(); tells you if clipping is activated.

Color

When using RGB-capable displays in ESPHome you may wish to use custom colors. A color component exists for just this purpose:

color:
  - id: my_light_red
    red: 100%
    green: 20%
    blue: 25%
    white: 0%

Alternatively, you can use <color>_int to specify the color as an int value:

color:
  - id: my_light_red
    red_int: 255
    green_int: 51
    blue_int: 64
    white_int: 0

Or, if you are more comforatble with hex values, you can use hex:

color:
  - id: my_light_red
    hex: FF3340

Configuration variables:

  • red (Optional, percentage): The percentage of the red component. Defaults to 100%.

  • red_int (Optional, integer): The brightness of the red component on a scale of 0 to 255. Defaults to 255.

  • green (Optional, percentage): The percentage of the green component. Defaults to 100%.

  • green_int (Optional, integer): The brightness of the green component on a scale of 0 to 255. Defaults to 255.

  • blue (Optional, percentage): The percentage of the blue component. Defaults to 100%.

  • blue_int (Optional, integer): The brightness of the blue component on a scale of 0 to 255. Defaults to 255.

  • white (Optional, percentage): The percentage of the white component. Defaults to 100%.

  • white_int (Optional, integer): The brightness of the white component on a scale of 0 to 255. Defaults to 255.

  • hex (Optional, string): The color in hexadecimal representation. Defaults to FFFFFF.

RGB displays use red, green, and blue, while grayscale displays may use white.

Graph Component

You can display a graph of a sensor value(s) using this component. The states used for the graph are stored in memory at the time the sensor updates and will be lost when the device reboots.

Examples:

../../_images/graph_screen.png
../../_images/graph_dualtrace.png

Graph component with options for grids, border and line-types.

graph:
  # Show bare-minimum auto-ranged graph
  - id: single_temperature_graph
    sensor: my_temperature
    duration: 1h
    width: 151
    height: 51
  # Show multi-trace graph
  - id: multi_temperature_graph
    duration: 1h
    x_grid: 10min
    y_grid: 1.0     # degC/div
    width: 151
    height: 51
    traces:
      - sensor: my_inside_temperature
        line_type: DASHED
        line_thickness: 2
        color: my_red
      - sensor: my_outside_temperature
        line_type: SOLID
        line_thickness: 3
        color: my_blue
      - sensor: my_beer_temperature
        line_type: DOTTED
        line_thickness: 2
        color: my_green

Configuration variables:

  • id (Required, ID): The ID with which you will be able to reference the graph later in your display code.

  • width (Required, int): The graph width in pixels

  • height (Required, int): The graph height in pixels

  • duration (Required, Time): The total graph history duration.

  • border (Optional, boolean): Specifies if a border will be drawn around the graph. Default is True.

  • x_grid (Optional): Specifies the time per division. If not specified, no vertical grid will be drawn.

  • y_grid (Optional, float): Specifies the number of units per division. If not specified, no horizontal grid will be drawn.

  • max_range (Optional): Specifies the maximum Y-axis range.

  • min_range (Optional): Specifies the minimum Y-axis range.

  • max_value (Optional): Specifies the maximum Y-axis value.

  • min_value (Optional): Specifies the minimum Y-axis value.

  • traces (Optional): Use this to specify more than a single trace.

Trace specific fields:

  • sensor (Optional, ID): The sensor value to plot

  • line_thickness (Optional): Defaults to 3

  • line_type (Optional): Specifies the plot line-type. Can be one of the following: SOLID, DOTTED, DASHED. Defaults to SOLID.

  • color (Optional): Sets the color of the sensor trace.

And then later in code:

display:
  - platform: ...
    # ...
    pages:
      - id: page1
        lambda: |-
          // Draw the graph at position [x=10,y=20]
          it.graph(10, 20, id(single_temperature_graph));
      - id: page2
        lambda: |-
          // Draw the graph at position [x=10,y=20]
          it.graph(10, 20, id(multi_temperature_graph), my_yellow);

color:
  - id: my_red
    red: 100%
    green: 0%
    blue: 0%
  - id: my_green
    red: 0%
    green: 100%
    blue: 0%
  - id: my_blue
    red: 0%
    green: 0%
    blue: 100%
  - id: my_yellow
    red: 100%
    green: 100%
    blue: 0%

Note

Here are some things to note: - Setting y_grid will expand any specified range to the nearest multiple of grid spacings. - Axis labels are currently not possible without manually placing them. - The grid and border color is set with it.graph(), while the traces are defined separately.

QR Code Component

Use this component to generate a QR-code containing a string on the device, which can then be drawn on compatible displays.

qr_code:
  - id: homepage_qr
    value: esphome.io

Configuration variables:

  • id (Required, ID): The ID with which you will be able to reference the QR-code later in your display code.

  • value (Required, string): The string which you want to encode in the QR-code.

  • ecc (Optional, string): The error correction code level you want to use. Defaults to LOW. You can use one of the following values:

    • LOW: The QR Code can tolerate about 7% erroneous codewords

    • MEDIUM: The QR Code can tolerate about 15% erroneous codewords

    • QUARTILE: The QR Code can tolerate about 25% erroneous codewords

    • HIGH: The QR Code can tolerate about 30% erroneous codewords

To draw the QR-code, call the it.qr_code function from your render lambda:

display:
  - platform: ...
    # ...
    pages:
      - id: page1
        lambda: |-
          // Draw the QR-code at position [x=50,y=0] with white color and a 2x scale
          it.qr_code(50, 0, id(homepage_qr), Color(255,255,255), 2);

Images

Use this component to store graphical images on the device, you can then draw the images on compatible displays.

image:
  - file: "image.png"
    id: my_image
    resize: 100x100

Configuration variables:

  • file (Required, string): The path (relative to where the .yaml file is) of the image file.

  • id (Required, ID): The ID with which you will be able to reference the image later in your display code.

  • resize (Optional, string): If set, this will resize the image to fit inside the given dimensions WIDTHxHEIGHT and preserve the aspect ratio.

  • type (Optional): Specifies how to encode image internally. Defaults to BINARY.

    • BINARY: Two colors, suitable for 1 color displays or 2 color image in color displays. Uses 1 bit per pixel, 8 pixels per byte.

    • GRAYSCALE: Full scale grey. Uses 8 bits per pixel, 1 pixel per byte.

    • RGB24: Full RGB color stored. Uses 3 bytes per pixel.

    • RGB565: Lossy RGB color stored. Uses 2 bytes per pixel.

    • TRANSPARENT_BINARY: One color, any pixel that is fully transparent will not be drawn, and any other pixel will be the on color. Uses 1 bit per pixel, 8 pixels per byte.

  • dither (Optional): Specifies which dither method used to process the image, only used in GRAYSCALE and BINARY type image. Defaults to NONE. You can read more about it here and here.

    • NONE: Every pixel convert to its nearest color.

    • FLOYDSTEINBERG: Uses Floyd-Steinberg dither to approximate the original image luminosity levels.

Note

To use images you will need to have the python pillow package installed. If you’re running this as a Home Assistant add-on or with the official ESPHome docker image, it should already be installed. Otherwise you need to install it using pip install pillow.

And then later in code:

display:
  - platform: ...
    # ...
    lambda: |-
      // Draw the image my_image at position [x=0,y=0]
      it.image(0, 0, id(my_image));

For binary images the image method accepts two additional color parameters which can be supplied to modify the color used to represent the on and off bits respectively. e.g.

display:
  - platform: ...
    # ...
    lambda: |-
      // Draw the image my_image at position [x=0,y=0]
      // with front color red and back color blue
      it.image(0, 0, id(my_image), id(red), id(blue));

You can also use this to invert images in two colors display, use COLOR_OFF then COLOR_ON as the additional parameters.

Animation

Allows to use animated images on displays. Animation inherits all options from the image component. It adds additional lambda methods: next_frame(), prev_frame() and set_frame() to change the shown picture of a gif.

animation:
  - file: "animation.gif"
    id: my_animation
    resize: 100x100

The animation can be rendered just like the image component with the image() function of the display component.

To show the next frame of the animation call id(my_animation).next_frame(), to show the previous picture use id(my_animation).prev_frame(). To show a specific picture use id(my_animation).set_frame(int frame). This can be combined with all Lambdas:

display:
  - platform: ...
    # ...
    lambda: |-
      //Ingress shown animation Frame.
      id(my_animation).next_frame();
      // Draw the animation my_animation at position [x=0,y=0]
      it.image(0, 0, id(my_animation), COLOR_ON, COLOR_OFF);

Note

To draw the next animation independent of Display draw cycle use an interval:

interval:
  - interval: 5s
      then:
        lambda: |-
          id(my_animation).next_frame();

Configuration variables:

  • file (Required, string): The path (relative to where the .yaml file is) of the gif file.

  • id (Required, ID): The ID with which you will be able to reference the animation later in your display code.

  • resize (Optional, string): If set, this will resize all the frames to fit inside the given dimensions WIDTHxHEIGHT and preserve the aspect ratio.

  • type (Optional): Specifies how to encode each frame internally. Defaults to BINARY.

    • BINARY: Two colors, suitable for 1 color displays or 2 color image in color displays. Uses 1 bit per pixel, 8 pixels per byte.

    • GRAYSCALE: Full scale grey. Uses 8 bits per pixel, 1 pixel per byte.

    • RGB24: Full RGB color stored. Uses 3 bytes per pixel.

    • RGB565: Lossy RGB color stored. Uses 2 bytes per pixel.

Display Pages

Certain display types also allow you to show “pages”. With pages you can create drawing lambdas that you can switch between. For example with pages you can set up 3 screens, each with different content, and switch between them on a timer.

display:
  - platform: ...
    # ...
    id: my_display
    pages:
      - id: page1
        lambda: |-
          it.print(0, 10, id(my_font), "This is page 1!");
      - id: page2
        lambda: |-
          it.print(0, 10, id(my_font), "This is page 2!");

You can then switch between these with three different actions:

show_next / show_previous: Shows the next or previous page, wraps around at the end.

on_...:
  - display.page.show_next: my_display
  - display.page.show_previous: my_display

# For example cycle through pages on a timer
interval:
  - interval: 5s
    then:
      - display.page.show_next: my_display
      - component.update: my_display

display.page.show: Show a specific page

on_...:
  - display.page.show: page1

  # Templated
  - display.page.show: !lambda |-
      if (id(my_binary_sensor).state) {
        return id(page1);
      } else {
        return id(page2);
      }

Note

To trigger a redraw right after the page show use a component.update action:

# For example cycle through pages on a timer
interval:
  - interval: 5s
    then:
      - display.page.show_next: my_display
      - component.update: my_display

display.is_displaying_page: This condition returns true while the specified page is being shown.

# In some trigger:
on_...:
  - if:
      condition:
        display.is_displaying_page: page1
      then:
        ...
  - if:
      condition:
        display.is_displaying_page:
          id: my_display
          page_id: page2
      then:
        ...

on_page_change: This automation will be triggered when the page that is shown changes.

display:
  - platform: ...
    # ...
    on_page_change:
      - from: page1
        to: page2
        then:
          lambda: |-
            ESP_LOGD("display", "Page changed from 1 to 2");

  • from (Optional, ID): A page id. If set the automation is only triggered if changing from this page. Defaults to all pages.

  • to (Optional, ID): A page id. If set the automation is only triggered if changing to this page. Defaults to all pages.

Additionally the old page will be given as the variable from and the new one as the variable to.

See Also