SuttaCentral: Merge html files from `de/pli/sutta/sn` into one big that can be converted to pdf or latex using pandoc

prepareHtml.js

// This script was placed in /sc-data/html_text/de/pli/sutta/sn (needs requirements `yarn add natural-sort glob ...`)
// Outputs book.html, a file of all merged text - (hopefully) correctly ordered.
// Minimal example for creating PDF: pandoc -o book.pdf --pdf-engine=xelatex --toc book.html

const nsort = require('natural-sort');
const glob = require('glob');
const fs = require('fs');
const cheerio = require('cheerio');

glob("**/*.html", (er, files) => {
  let bookAsHtml = '';
  let currentChapter = '';

  files
    .sort(nsort()) // apply natural number sorting
    .map(file => {
      const $ = cheerio.load(fs.readFileSync(file));

      // h1 -> h2 (make title subsection)
      const title = $('.sutta h1').html();
      $('.sutta h1').replaceWith(`<h2>${title}</h2>`);

      // add new h1 section if changed 
      const chapter = $('.sutta .hgroup > p').first().text();
      if (chapter && currentChapter !== chapter) {
        bookAsHtml += `<h1>${chapter}</h1>`;
        currentChapter = chapter;
      }

      const sutta = $('.sutta article').html();

      if (sutta) {
        bookAsHtml += sutta;
      }
    });

  fs.writeFileSync('book.html', bookAsHtml);
});