mm-test.js

// Just run with node mm.js and a file "ids.txt" is generated in the same directory.
var https = require('https');
var fs = require('fs');

var submission_id = "3lhmxs";
// Last comment id on comment list, if any.
var cutoff_comment_id = "cv7cbxs";
var link = "https://www.reddit.com/r/millionairemakers/comments/" + submission_id + ".json?sort=old";

https.get(link, function (res) {
  res.setEncoding('utf8');
  var body = "";
  res.on('data', function (d) {
    body += d;
  });

  res.on('end', function() {
    var data = parseContent(body);
    writeFile(data);
  });
});

function writeFile(ids) {
  var text = ids.join("\n");
  fs.writeFile("ids.txt", text);
}

function parseContent(content) {
  // get content
  var data = JSON.parse(content);

  var comment_ids = [];
  var children = data[1].data.children;
  var outer_break = false;

  for (var i = 0; i < children.length; i++) {
    var child = children[i];
    if (child.data.parent_id === "t3_" + submission_id) {
      if (child.kind === "t1") {
        comment_ids.push(child.data.id);
      } else if (child.kind === "more") {
        var inner_children = child.data.children;
        for (var j = 0; j < inner_children.length; j++) {
          var inner_child = inner_children[j];
          comment_ids.push(inner_child);
          if (inner_child == cutoff_comment_id) {
            // if we want to break outer loop.
            //outer_break = true;
            break;
          }
        }
        // just in case the cutoff id isn't in a "more", who knows.
        // todo: fix repo?
        if (outer_break) {
          break;
        }
      }
    }
  }
  return comment_ids;
}