Weather Predictor Application

This is a demonstration of GraalVM's polyglot capabilities showing how objects and functions can be shared among different languages.

The application is a simple REST service and simple web page for predicting temperature in a given city. The predictor uses a model constructed by applying linear regression to a data set consisting of temperatures in selected cities expressed as (latitude, temperature) tuples. The temperatures are retrieved in the Ruby part of the application via a gem wrapping OpenWeatherMap API service, while the corresponding latitudes and other cities data are retrieved from a Java class that mocks a typical data access object and can be easily extended to talk to a real database.

The regression model is also built in the R module by means of the lm function. To predict the temperature in a given city the R module defines a function passing the city's latitude into the predict function, which uses the model to return the predicted temperature for the given latitude.

The predictor service is exposed in the JavaScript module of the application, which is designed as a simple Express.js based Node.js server. This module is logically divided into three stages. First, the Ruby module is loaded to initialize the OpenWeather service. Then the R module is loaded to build the prediction model. Finally, the Express.js server is launched. The server serves the web page UI at http://localhost:12836/ and one can directly query the REST API at http://localhost:12836/predict/, e.g. http://localhost:12836/predict/London. The reply to this query is a JSON object carrying the city's real and predicted temperature, such as {"city":"London","real":6.82,"predicted":14.017807794947569}.

Setup

Download the latest GraalVM image from the Oracle Technology Network. Extract the archive and set the GRAALVM_DIR environment variable to point to the graalvm directory.

Then, execute the build.sh script.

A typical problem is proxy set-up: verify that you have the http_proxy, https_proxy, and no_proxy environment variables set properly.

Run

Execute the run.sh script and open your browser at http://localhost:12836.

Note that run.sh runs the application without displaying exceptions to the terminal. It is possible to pass in arguments to enable the Chrome Inspector (--inspect and --inspect.Suspend=false) or the monitoring Agent (--agent). E.g. ./run.sh --inspect --inspect.Suspend=false --agent. The tools.sh script enables the above mentioned tools and allows printing of exceptions to terminal.