uva-stalker is a Telgram bot to stalk submissions from UVa Online Judge users.
You simply tell it which users you want to follow, and it will send you notifications every time a user makes a new submission.
An instance of the bot is already running and can be interacted with by chatting to UVaStalkerBot, or you can clone this repo and set up the bot in your own server.
Follow these instructions to set up the bot in a development environment and test that it works correctly. You can then deploy it to a server.
Chat with BotFather to generate an authorization token for your new bot. Head over to the Telegram documentation for detailed instructions if you're new to creating bots.
uva-stalker uses the telegram-bot-sdk library in a Laravel environment using a (tested in Postgres) database. A suitably configured development environment has already been set up using Docker containers thanks to laradock.
After cloning the repo, you may configure the environement by typing in
$ cd laradock-5.0.3
$ cp env-example .env
And changing the environment variables contained in the .env file inside the laradock-* folder. For example, using a Postgres database, configure these variables
POSTGRES_DB=default
POSTGRES_USER=default
POSTGRES_PASSWORD=default
as needed. You should also fill in your Docker Host IP.
# Replace with your Docker Host IP (will be appended to /etc/hosts)
DOCKER_HOST_IP=10.0.75.1
Start up the environment and find out the container id of the workspace instance using docker ps.
$ docker-compose up -d nginx postgres
Install the project dependencies using docker exec -t <container_id> composer install.
Configure the database credentials in the project folder in a .env file, as well as your APP_KEY (or use php artisan key:generate to fill it in). You must also fill in the variable TELEGRAM_BOT_TOKEN with your authorization token. You may then install the migrations and execute them to build the database tables.
$ cd ..
$ cp .env.example .env
$ nano .env
... (edit values)
$ docker exec -t <container_id> php artisan migrate:install
$ docker exec -t <container_id> php artisan migrate
Finally, run daemon.php to listen for messages sent to your bot. This script uses long-polling to request updates every second from the Telegram Bot Api using the getUpdates method. It will forward the payloads to the Laravel application, which will process them under the /api/<authorization_token route, where <authorization_token> is the authorization token which BotFather provided you when you created your bot.
$ docker exec -t <container_id> php daemon.php
Starting bot with config {"id":356029581,"first_name":"UVaStalker","username":"UVaStalkerBot"}
Start polling Telegram api (will poll for updates every 1000 ms)
Poll took 0 ms (slept 1000000 µs)
Poll took 0 ms (slept 1000000 µs)
...
Any incoming updates will be printed to the console.
If the bot works normally using long polling, you may want to deploy it to a server and set a webhook so that incoming updates are pushed to you. By default, the Laravel application is listening for updates on /api/<authorization_token, so you should set up the webhook at
https://your-domain-name.com/api/<authorization_token>
Note that you need a domain name with a valid SSL certificate in order to set up the webhook.
If you have a domain name but you don't have a valid SSL certificate, a very easy way to get one is from Let's Encrypt, and install it on a Caddy web server. Laradock makes this a cinch, since Caddy is included with it. To set it up, edit the Caddyfile inside the laradock-5.0.3/caddy folder. Replace the 0.0.0.0:80 in the first line with your domain name
https://your-domain-name.com
root /var/www/public
...
and uncomment #tls self_signed at the bottom of the file, replacing self_signed with the email you registered your domain with.
# Change the first list to listen on port 443 when enabling TLS
tls youremail@gmail.com
...
All set up! You may now launch Caddy and verify that you can reach your web server via https://your-domain-name.com.
$ cd ..
$ docker-compose up caddy
Starting laradock503_applications_1
laradock503_workspace_1 is up-to-date
laradock503_php-fpm_1 is up-to-date
laradock503_caddy_1 is up-to-date
Attaching to laradock503_caddy_1
caddy_1 | Activating privacy features... done.
caddy_1 | https://your-domain-name.com
caddy_1 | http://your-domain-name.com
If all is good, launch the containers in detached mode.
$ docker-compose down
$ docker-compose up -d caddy postgres
All that's left to do is set up your webhook. You may do so using the setWebhook() method of the Bot Api. Telegram has a fantastic guide on webhooks for detailed instructions.