dm-scripts/parseunimap.js

import {createWriteStream} from 'fs';
import {JSDOM} from "jsdom";
import fetch from "node-fetch";

(async () => {
  try {
    const url = process.argv[2];
    if (!url) {
      console.error("Usage: node script.js <registro_link>");
      process.exit(1);
    }

    const response =
        await fetch(url, {headers : {"User-Agent" : "Mozilla/5.0"}});
    const buffer = await response.arrayBuffer();
    const isoHtml = Buffer.from(buffer, "binary").toString("latin1");

    const dom = new JSDOM(isoHtml);
    const document = dom.window.document;

    let ol = document.querySelector("ol");
    if (!ol) {
      console.error("No <ol> found on the page");
      return;
    }

    ol.querySelectorAll("a, i").forEach(el => el.remove());

    const results = [...ol.children ]
                        .map(li => li.textContent.match(/\w+:(.+)/sm))
                        .filter(match => match)
                        .map(match => match[1].trim())
                        .map((x, index) => (index + 1) + ". " + x + " \n");

    const writeStream = createWriteStream("registro.md")
    const pathName = writeStream.path;
    writeStream.on('error', function(err) { /* error handling */ });
    const head = "# Registro delle lezioni \n\n";
    writeStream.write(head)
    results.forEach(value => writeStream.write(`${value}`));
    // the finish event is emitted when all data has been flushed from the
    // stream
    writeStream.on(
        'finish',
        () => { console.log(`Registro salvato nel file ${pathName}`); });

    // handle the errors on the write process
    writeStream.on(
        'error',
        (err) => {console.error(
            `There is an error writing the file ${pathName} => ${err}`)});

    // close the stream
    writeStream.end();
  } catch (error) {
    console.error("Error fetching or processing page:", error);
  }
})();