- Make sure Node.js is installed.
- On Linux make sure the followings are installed:
sudo apt install gcc make g++ pkg-config libpixman-1-dev libcairo2-dev libpango1.0-dev libjpeg8-dev libgif-dev
- Not always needed, but recommended: ensure the necessary native dependencies for canvas are installed.
- Clone this repository.
- Run
npm install
to install needed dependencies.
NOTE: For Docker instructions skip ahead
- Run 'npm install -g typescript'
- Move to 'lib' folder 'cd lib'
- Run 'npm install -g ts-node'
Note: The converter might fail on Windows if the file path containes multiple byte characters (Chinese/Japanese/Korean characters etc.). It's recommended to rename these files before passing them to the converter, or you can switch on enable Beta:Use Unicode UTF-8 for worldwide language support
options.
For Linux:
# C format
./lv_img_conv.js logo_lvgl.png -f -c CF_TRUE_COLOR_ALPHA
# Binary format (RGB565)
./lv_img_conv.js logo_lvgl.png -f -c CF_TRUE_COLOR_ALPHA -t bin --binary-format 565
# C format with dither algoritm
./lv_img_conv.js logo_lvgl.png -f -d true -c CF_TRUE_COLOR_ALPHA
For Windows:
# C format
ts-node cli.ts logo_lvgl.png -f -c CF_TRUE_COLOR_ALPHA
# Binary format (RGB565)
ts-node cli.ts logo_lvgl.png -f -c CF_TRUE_COLOR_ALPHA -t bin --binary-format 565
# C format with dither algoritm
ts-node cli.ts logo_lvgl.png -f -d true -c CF_TRUE_COLOR_ALPHA
A file called logo_lvgl.c
will be created in the same directory.
docker build -t lv_img_conv .
docker run --rm \
-u 1000:1000 \
-v /path/to/project:/usr/src/proj \
lv_img_conv \
<lv_img_conv.js arguments>
docker run -it --rm \
-u 1000:1000 \
-v /path/to/project:/usr/src/proj \
lv_img_conv
lv_img_conv.js
is in the $PATH
and callable from the interactive shell.
Parameter | Function |
---|---|
-u 1000:1000 |
Set to your host user's UID/GID, unless you don't mind root ownership on output files |
-v /path/to/project:/usr/src/proj |
Change /path/to/project to directory with image(s) to convert |
This converter was originally created by @embeddedt as an attempt to solve some of the common issues experienced with the PHP converter, such as running out of memory on large images or failing to read certain PNGs correctly.
Much of the actual conversion logic remains unchanged from the previous implementation, which can be found at https://github.com/lvgl/lv_utils/blob/b298fe71675e9c12016adabcc8889394b477b89b/img_conv_core.php.