LILYGO UI

English

简体中文

English

简体中文

Appearance

Development quick start ​

This guide explains the development workflow described by the upstream LILYGO-UI repositories. It is intended for application developers and contributors. To write a system image to your device, start with Quick start: USB and TF card flashing.

1. Install the host tools ​

The lpm README lists the current requirements:

  • Python 3.11 or newer
  • CMake 3.21 or newer
  • Git
  • A C/C++ compiler and the host dependencies required by SDL2
  • dpkg-deb when creating Debian packages
  • SSH and SCP when deploying to a device

The exact versions and installation steps can change. Follow the upstream README for the current setup instructions before starting a project. A documented installation option is:

sh
python3 -m pip install "git+https://github.com/LILYGO-UI/lpm.git"
lpm --version

The upstream README also recommends installing lpm as a uv tool when that matches your Python workflow.

2. Create or inspect an application ​

Use the template repository as the reference project. It is a complete independent application and demonstrates:

  1. a responsive LVGL interface;
  2. C++ MVVM-style separation between UI and application state;
  3. launcher discovery through application metadata; and
  4. Debian packaging for installation on the target.

The lpm CLI is the standard project manager. Its documented workflow covers initializing from the template, configuring a host build, cross-compiling for AArch64, packaging, and deploying over SSH. For example, an empty directory can be initialized with:

sh
mkdir weather-station
cd weather-station
lpm init

The upstream README also documents lpm start for the host simulator, lpm test --report build/test-results.xml for tests, lpm pack --output dist for a Debian package, and lpm deploy --host cm0.local --user pi for SSH deployment. Use the current README for options and target names.

3. Run a host build ​

The AppKit supports a desktop host profile backed by SDL2. This lets you validate layout and interaction before deploying to a CM0 device. The normal flow is:

  1. initialize or clone an application project;
  2. configure it with CMake using the host profile;
  3. build and run the SDL2 host executable; and
  4. iterate on the LVGL UI and application state.

Use the commands in the AppKit README and the template README for the current command names and options.

4. Prepare a CM0 target build ​

The LILYGO-UI development repositories document an AArch64 Linux target workflow. The CM0BspBuilder project creates a reusable application sysroot and the matching CMake toolchain from released Raspberry Pi OS ZIP or raw IMG inputs. It also manages development-package overlays, versioned archives, checksums, and manifests.

Ensure that the target system, sysroot, and compiler toolchain use matching architectures and compatible versions. For the system image and flashing steps, see Quick start.

The builder is an application-build dependency. Its README explicitly states that it is not a bootloader, kernel, device-tree, module, or complete device-image builder.

5. Package and deploy ​

After the target build, use the packaging and deployment workflow documented by lpm. It can create Debian packages and copy them to a device over SSH/SCP. Keep the target's operating system and credentials under your normal device administration process.

Hardware setup ​

Check board specifications and interfaces in the official T-Display-CM0 Wiki (Chinese). To install the LILYGO UI system image, follow Quick start: USB and TF card flashing.

Where to go next ​

  • Review the application catalog to find a nearby example.
  • Read the development guide for AppKit targets, manifests, and runtime boundaries.
  • Check FAQ before opening an issue about hardware support or system images.