Putting together VS Code extension, Quarkus, and Deep Java Library into a simple project. Just for fun! :P
-
vscode-extension: Activate the VS Code extension by initializing the Quarkus app (with support of the backend library from Kogito Tooling). -
app-server: Quarkus application that exposes REST endpoints for editor services.
Note: The Quarkus application is automatically started up and stopped when the VS Code extension is activated and deactivated, respectively. Also, the build process takes care of embedding the Quarkus application into the VS Code extension.
- Node
16+ - Yarn
1.22.10(npm install -g [email protected]) - Lerna
4.0.0(npm install -g [email protected]) - Maven
3.6+ - Java
11
- Image classification
- Object detection
- Auto crop detected objects
- Image generation with BigGAN
- Sentiment analysis
- Open this example in VS Code 1.43+
- In the terminal, execute
yarn bootstrap && yarn build:dev F5to start debugging
Then:
- Open an image file (
*.png,*.jpg, or*.jpeg) and click on the buttons for image services. - Open a text file (
*.txt) and click on the buttons for text services.
VS Code will send a POST request to the embedded Quarkus app, which will perform the required operation on the image/text and report back the result.
Important: Running for the first time will require a longer time to complete each operation since the models need to be downloaded and cached into your machine.
Run yarn bootstrap && yarn build:prod if you want to generate the vsix file.
Once the build process is done, the vsix file will be at packages/vscode-extension/dist.
The vsix file is also generated by the GitHub workflow and uploaded as an artifact.