Delemangi / finki-scraper

Scripts for scraping various FINKI (FCSE) services, and providing them by a Discord webhook and a REST API

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

FINKI Scraper

Various scripts for scraping several FINKI (FCSE) services for the purpose of forwarding them to Discord using webhooks.

It's recommended, but not required to run this inside a Docker container.

Installation

For development purposes, be sure to run npm run prepare to install the Git pre-commit hooks.

Installation (Docker)

  1. git clone git@github.com:Delemangi/finki-scraper.git
  2. docker compose build

Installation (Normal)

  1. git clone git@github.com:Delemangi/finki-scraper.git
  2. npm install

Running

You may optionally pass arguments to the below commands to override which scrapers are run.

Running (Docker)

docker compose run scraper or docker compose run scraper s1 s2 ... sn

Running (Normal)

npm run start or npm run start s1 s2 ... sn

Server Mode

You can also run this app in server mode using npm run serve, which will boot up an Express based server which will open endpoints for receiving new announcements. In this mode, the app will only scrape new announcements whenever it receives a request to get them, instead of periodically. It is also possible to clear the cache to obtain the same announcements again, as well as to fetch all available scrapers. There is a separate Docker Compose configuration in docker-compose.server.yaml for running the app in server mode, and it exposes the port 3000 by default.

The exposed endpoints are:

  • GET /list to get all available scrapers,
  • GET /get/<name> to get all announcements from a certain service,
  • DELETE /delete to delete the cache of all services,
  • DELETE /delete/<name> to delete the cache of the announcements from a certain service,

where name corresponds to the name (key) in the configuration.

Config

Create a config folder and in it create config.json. Example:

{
  "successDelay": 600000,
  "errorDelay": 60000,
  "maxPosts": 20,
  "webhook": "<logging webhook url>",
  "sendPosts": true,
  "scrapers": {
    "announcements": {
      "strategy": "announcements",
      "link": "https://www.finki.ukim.mk/mk/student-announcement",
      "webhook": "<webhook url>",
      "role": "[role]",
      "enabled": true
    },
    "jobs": {
      "strategy": "jobs",
      "link": "https://www.finki.ukim.mk/mk/fcse-jobs-internships",
      "webhook": "<webhook url>",
      "role": "[role]",
      "enabled": true
    },
    "events": {
      "strategy": "events",
      "link": "https://finki.ukim.mk/mk/fcse-events",
      "webhook": "<webhook url>",
      "role": "[role]",
      "enabled": true
    },
    "projects": {
      "strategy": "projects",
      "link": "https://finki.ukim.mk/mk/fcse-projects",
      "webhook": "<webhook url>",
      "role": "[role]",
      "enabled": true
    },
    "SIC": {
      "strategy": "course",
      "link": "https://courses.finki.ukim.mk/mod/forum/search.php?id=263&words=&phrase=&notwords=&fullwords=&hfromday=1&hfrommonth=1&hfromyear=1&hfromhour=1&hfromminute=1&htoday=1&htomonth=1&htoyear=1&htohour=1&htominute=1&forumid=486&subject=&user=&perpage=25",
      "webhook": "<webhook url>",
      "role": "[role]",
      "cookie": "[cookie]",
      "enabled": true
    }
  }
}

Use the enabled property to control which scrapers will be activated on the next run.

More scrapers may be added in config.json, and each scraper should have its own scraper strategy implemented. The course scraper strategy may be reused for other Moodle courses.

About

Scripts for scraping various FINKI (FCSE) services, and providing them by a Discord webhook and a REST API

License:MIT License


Languages

Language:TypeScript 98.1%Language:Dockerfile 1.9%