Get Google Document as HTML

Question

I had a wild idea that I could build a website blog for an unsophisticated user friend using Google Drive Documents to back it. I was able to create a contentService that c

Answer 1

I've had this problem as well. The HTML that the Document HTML Export spits out is really ugly, so this was my solution:

/**
 * Takes in a Google Doc ID, gets that doc in HTML format, cleans up the markup, and returns the resulting HTML string.
 *
 * @param {string} the id of the google doc
 * @param {boolean} [useCaching] enable or disable caching. default true.
 * @return {string} the doc's body in html format
 */
function getContent(id, useCaching) {

  if (!id) {
    throw "Please call this API with a valid Google Doc ID";
  }

  if (useCaching == null) {
    useCaching = true;
  }

  if (typeof useCaching != "boolean") {
    throw "If you're going to specify useCaching, it must be boolean.";
  }

  var cache = CacheService.getScriptCache();
  var cached = cache.get(id); // see if we have a cached version of our parsed html
  if (cached && useCaching) {
    var html = cached;
    Logger.log("Pulling doc html from cache...");
  } else {

    Logger.log("Grabbing and parsing fresh html from the doc...");

    try {
      var doc = DriveApp.getFileById(id);
    } catch (err) {
      throw "Please call this API with a valid Google Doc ID. " + err.message;
    }

    var docName = doc.getName();

    var forDriveScope = DriveApp.getStorageUsed(); // needed to get Drive Scope requested in ScriptApp.getOAuthToken();
    var url = "https://docs.google.com/feeds/download/documents/export/Export?id=" + id + "&exportFormat=html";
    var param = {
      method: "get",
      headers: {"Authorization": "Bearer " + ScriptApp.getOAuthToken()},
      muteHttpExceptions:true,
    };

    var html = UrlFetchApp.fetch(url, param).getContentText();

    // nuke the whole head section, including the stylesheet and meta tag
    html = html.replace(/.*<\/head>/, '');
    // remove almost all html attributes
    html = html.replace(/ (id|class|style|start|colspan|rowspan)="[^"]*"/g, '');
    // remove all of the spans, as well as the outer html and body
    html = html.replace(/<(span|\/span|body|\/body|html|\/html)>/g, '');
    // clearly the superior way of denoting line breaks
    html = html.replace(/
/g, '
');

    cache.put(id, html, 900) // cache doc contents for 15 minutes, in case we get a lot of requests

  }

  Logger.log(html);

  return html;

}

https://gist.github.com/xd1936/cc229d14a89e6327336177bb07ac2980