Coronavirus (COVID-19) statistics with Slack notification

COVID-19 datasource: https://corona.lmao.ninja

Features

• Send statistics on daily basis (at 4:00AM UTC+0)
• Send statistics in chat by slack command using endpoint /corona/stats

Installation

Docker container

Project provides complete docker image for usage. To run docker container you should do the following steps:

• mkdir -p /opt/corona-stats && touch /opt/corona-slack/env.list
• Put environment variables SLACK_CHANNEL and SLACK_TOKEN to /opt/corona-slack/env.list SLACK_CHANNEL - Slack channel id
  SLACK_TOKEN - Slack bot token
• Run docker run --rm --name docker.corona-slack.service --env-file /opt/corona-slack/env.list -p 8082:8080 mdsina/corona-slack:latest
• Slack requires https protocol connection, so you need a front proxy before application that can be accessible though https. You can do that though nginx as for example:

location ~ ^/corona-bot(/?)(.*) {
    proxy_pass          http://0.0.0.0:8082/$2;
    proxy_set_header    Host             $host;
    proxy_set_header    X-Real-IP        $remote_addr;
    proxy_set_header    X-Forwarded-For  $proxy_add_x_forwarded_for;
    proxy_set_header    X-Client-Verify  SUCCESS;
    proxy_set_header    X-Client-DN      $ssl_client_s_dn;
    proxy_set_header    X-SSL-Subject    $ssl_client_s_dn;
    proxy_set_header    X-SSL-Issuer     $ssl_client_i_dn;
    proxy_read_timeout 1800;
    proxy_connect_timeout 1800;
}

• After that specify URL to https://<server>/corona-bot/corona/stats in your Slack App configuration for slash command

You can also use docker.corona-slack.service file that can pull docker image and run container without any additional configuration, except /opt/corona-slack/env.list file.

Build

Requirements

• JDK 11 and set JAVA_HOME environment variable
  • you can use GraalVM 20.0.0 if you need to build Docker image and use Native Image
• maven
  • can be used though mvnw wrapper, that downloads maven automatically and run them
• Docker, to build docker container with Native Image through GraalVM

Build and run jar file

Run in root directory ./mvnw clean install.

Jar file car be found at <project root>/target/corona-slack-0.1.jar

To run jar file you can use the following commands:

$ export SLACK_CHANNEL=<channel id>
$ export SLACK_TOKEN=<slack token>
$ java -jar ./target/corona-slack-0.1.jar

That will run http server on port 8080:

/corona-slack# java -jar target/corona-slack-0.1.jar
18:50:31.650 [main] INFO  io.micronaut.runtime.Micronaut - Startup completed in 2066ms. Server Running: http://localhost:8080

Build Docker image with AOT project compilation

First of all you need to compile jar file as described in in that README.

Update Native Image configuration

After that, if you planned to add some libraries that do not support GraalVM SVM you need to rebuild native image configuration files:

 java -agentlib:native-image-agent=config-output-dir=./src/main/resources/META-INF/native-image/,caller-filter-file=./native-image-filter.json -jar target/corona-slack-0.1.jar

After start the application, you can do some requests to endpoint /corona/stats with your bot or by hands. That will help native image aget to collect information about used classes via reflection, as described in documentation.

That will updates files in <project root>/src/main/resources/META-INF/native-image/ directory.

After that you need to remove some records in json files reflect-config.json and in resource-config.json. SVM has some issues on filtering classes that should not be loaded on building native image.

The records, that relates to the following packages should be removed to prevent runtime errors. This classes already SVM-optimized and will be collected by native image build proccess:

• io.netty
• io.micronaut
• com.github.mdsina.corona

Build Docker Image

To build Docker Image just simply run ./docker-build.sh

TODO

• Federated accessing to bot backend
• Global Slack App with registration in bot backend
• Non-docker based systemd configuration file for pure java
• Configurable list of countries to show in statistics
• ...