Background

Google Docs (or Drive) offers a very handy web form and you can make and share it in seconds. If you are wondering how to make a registration form for your low budget meetings or short-run projects, Google Form would be a perfect choice.

However, when used with the default setting, the Google form works not that perfect: You may have wondered how to force it send an auto-response message for the submitter. I suppose this is intentional not to make it a playground for spammers. But don’t they allow us to have such a functionality even for temporary websites, or in limited-access areas?

The answer is YES. You can make it work with Google Apps Script. See, for example, Tutorial: Automating a Help Desk Workflow. The idea is simple: Watch the submission event and let it trigger an email sending script.

Why I didn’t fully appreciate it

In the above mentioned tutorial site, you saw a very small piece of script that works perfectly. But I was not content with that solution. In common situations, I suppose, you, programmer, set up a web form and your non-programmer colleague actually uses it. So, I wondered: How do they edit the template for the auto-response message? Should you explain that \n means a new line, put a semicolon at the end of the sentence, don’t use a quotation mark between quotation marks and stuff like that? Who becomes happy in such a situation?

What I did

Here I show one way to make the script read a template that is easily editable for non-programmers. What I wanted is something like the relationship between HTML and CSS:

CSS is designed primarily to enable the separation of document content (written in HTML or a similar markup language) from document presentation, including elements such as the layout, colors, and fonts. (Quoted from Wikipedia)

Separation is always good. So, let’s do it with Google form too.

Step 1. Create a form. Because this is a demonstration only, I put only two text boxes: Name and E-mail. …Sorry for the Japanese in the pic

Step 2. Create a document like THIS (This doc is publicly visible). Step 2 and 3 are interchangeable. I present this way to specify the requirement of the script. After creating the template, look at the URI and find the document ID (in my case, it’s 14O7df3d_kNiKB6tDhgzrLz2vhaDKCc4fbiS3O_KmC88).

Template is split into three parts by ===’s (more than 3).

• In the 1st part, the header information is set. You could dynamically change the subject too, but here we just don’t do that.
• The body of the message goes in the 2nd part. When sending the message, the submission contents for Name and E-mail, will replace the doubly-bracketed {{Name}} and {{E-mail}}, respectively. You can also use the default field Timestamp.
• 3rd part is ignored.

Your non-programmer colleague will be happy with this template? If you think not enough, please make the template neater and the script better.

Step 3. Here is what I did. (I’m not a pro-programmer, so please teach me if you know a better solution. I really appreciate it.)

Scripting

Let’s specify the Doc ID and the name of the column where recipients’ email addresses are stored.

loadTemplate() function (ll. 42-) reads the template file and returns the header and body while cutting off the 3rd (memo) part.

The core is at ll. 22-23. The doubly-bracketed keywords are replaced with the corresponding form entries.

// Doc ID of the template file
var id = "koko_ni_id_wo_iretekudasai";
var mailCol = "E-mail";

function onFormSubmit(e){
    try {
      var ss = SpreadsheetApp.getActiveSpreadsheet();
      var r = ss.getLastRow();
      var c = ss.getLastColumn();
      var rg = ss.getDataRange();

      // Processing the body part
      var t = loadTemplate(id);
      var body = t.body;
      var v = {};
      for ( var j = 1; j <= c; j++ ){
        // Replacing the template keyword into the input value
        var cname = rg.getCell(1, j).getValue();
        var entry = rg.getCell(r, j).getValue();
        v[cname] = entry;

        var re = new RegExp("{{"+cname+"}}", "g");
        body = body.replace(re, entry);
      }

      // Processing the header part
      var mail = v[mailCol];
      var subject = t.head.subject;
      var opt = { cc: t.head.cc,
                  bcc: t.head.bcc,
                  replyTo: t.head.replyTo,
                  name: t.head.name };

      // Sending the message
      MailApp.sendEmail( mail, subject, body, opt );
    } catch (e) {
        // When exception raised
        MailApp.sendEmail(Session.getActiveUser().getEmail(), "Error report", e.message);
    }
}

function loadTemplate(docId){
  // Load your template doc file
  // docId = Google Docs Document ID
  // Find it from doc's URI
  var t = DocumentApp.openById(docId).getText();

  // Comment out strings after double slashes
  t = t.replace(new RegExp("//.*\n", 'g'), "");

  // Body part is placed between a pair of lines with more than 3 consecutive ='s
  var tsplit = t.split(/===+.*\W+/);

  // Retrieving the header part
  // that should be placed before the first ===
  var head = {};
  var lines = tsplit[0].split(/\n+/);
  for ( var i in lines ){
    var line = lines[i];
    if ( null != line.match(/(\w+)\s*=\s*(.+)$/)) {
        head[RegExp.$1] = (RegExp.$2).trim();
    }
  }

  // Retrieving the body part
  // that should be place between the firs and last ==='s
  var body = tsplit[1];

  return { head: head, body: body};
}

Result

The script works like the following pic. So far so good.

#!/usr/bin/env python
# -*- coding: utf-8 -*-

import matplotlib.pyplot as plt

def scatterplot(data, data_name):
    """Makes a scatterplot matrix:
    Inputs:
      data - a list of data [dataX, dataY,dataZ,...];
             all elements must have same length

      data_name - a list of descriptions of the data;
                  len(data) should be equal to len(data_name)

    Output:
      fig - matplotlib.figure.Figure Object

    """

    N = len(data)
    fig = plt.figure()

    for i in range(N):
        for j in range(N):
            ax = fig.add_subplot(N,N,i*N+j+1)

            if j == 0: ax.set_ylabel(data_name[i],size='12')
            if i == 0: ax.set_title(data_name[j],size='12')
            if i == j:
                ax.hist(data[i], 10)
            else:
                ax.scatter(data[j], data[i])

    return fig

# Example
if __name__ == "__main__":
    import numpy as np
    import numpy.random as npr

    X = npr.randn(100)
    Y = 1.2 * X + npr.normal(0.0, 0.1, 100)
    Z = - Y ** 2 + X + 0.05 * npr.random(100)
    W = X + Y - Z + npr.normal(0.0, 2.0, 100)

    data = [X, Y, Z, W]
    data_name = ['Data X', 'Data Y', 'Data Z', 'Data W']

    fig = scatterplot(data, data_name)

    fig.savefig('scatterplot.png', dpi=120)
    plt.show()

the output of the example looks like the figure below…

Six Lessons I’ve Learned Since I Started Working for Myself: http://lifehacker.com/5663011/six-lessons-ive-learned-since-i-started-working-for-myself

The figures in the report indicate that the state of the Japanese business and innovation is brilliant but the economic sophistication is overshadowed by the government inefficiency. The too heavy burden hampers the Japanese market to grow faster. We all agree that social security is important, but the engine of the economy, which is the source of the social security, should be market trading. Why doesn’t it look at its efficiency?

The statistics of Japanese economy is usually gloomy these days. We should try harder to regain the central position of Asian economy. I hope the government stay stable against the political turbulence. The government and the citizens have no time to keep looking inward. The global economy is incessantly moving.

After hosting three major meetings, we are having the feeling that our conferences were somewhat inward-looking; I mean the targets of the Johogaku.net conferences has been Japanese IT engineers, investors and students. We don’t have any English talks, English-speaking speakers, or English-speaking participants (actually we had one French participant). Although the Japanese engineers have the technology and people that can globally compete with and even improve the global standard, the market doesn’t recognize them as big deals.

This fact can be found with regrettable figures in the Global Information Technology Report‘s by WEF; the global (relative) competitiveness of the Japanese IT industry is getting worse. Globalization has been one of the top-rated issues for Japanese people and the government. But there’s been little fruit of the agenda.

Johogaku.net is gonna be holding next big conference in May, 2011. To fight against the closed nature of Japanese IT industry, we’ve set the slogan of it as the “global expansion.” Re-iterating the issue should convey important messages to the Japanese or international students who are going to create the better future. We’re gonna discuss how Japanese IT business can regain its appeal to the international market among other topics. We’re trying to contribute to make the situation better by offering a junction of energetic ideas of young creators and engineers and wisdom of all generations.

If you are interested, please feel free to contact me or the conference organizers. Join us as a participant, a sponsor, or an organizer. In any case, you’ll be most welcome.

I missed the important news that Dropbox started to offer 8GB disk space for free! I was sooo stupid… If you have no idea what Dropbox is, please just google it. You can find a lot of information on it. Roughly speaking, it’s real awesome combination of a nice software and a great online storage service.

If you don’t have an account, join through my referral link: https://www.dropbox.com/referrals/NTEwMjg2NDI5.
We’ll both get 250MB additional space. (The minimum storage is 2GB. You can get additional space if you invite other people to Dropbox, up to 8GB. The limit used to be 3GB.)

The new version, 3.0, of WordPress, on which this website is running, has arrived lately. It has quite nice features. Among them, I’m pleased because you are finally able to manage multiple blogs on a single WordPress program. You know, my website has two identical WordPress core programs to manage Japanese and English versions. Editing small things can be a nuisance because I had to log out and in to move to the other site. So, this version-up can be an excitement to me.

But it is not so easy to upgrade all of my website immediately because I slightly modified several PHP files to somewhat optimize the appearance for each language and I’m not quite sure how things change after employing the multi-blog feature.

Now is the time to ponder.

Slight Modification of the Built-in Commands

To fully enjoy quix in the address bar, I strongly recommend to set a “default” search engine. Since this script requires, by default, a command, such as ‘g’, it causes a lot of troubles for someone who loves the simple behavior of the address bar. Let’s get rid of this by rebuilding “quix.txt” file. Download it and open it with your favorite text editor. Add a line at the bottom of it with the following statement:

* http://www.google.com/search?q=%s Untriggered search

By this line, the script interprets an input without any command as a to-be searched phrase. It works quite well. If you wish, you are allowed to add commands for your favorite search engines by editing quix.txt. Follow the instructions given in http://quixapp.com/help/syntax/

Because I’m Japanese, I thought it was necessary to add Japanese counterparts of Google and Amazon. You can’t forget Google Scholar. My custom commands are as follows:

@Custom
@Commands that I added
* http://www.google.com/search?q=%s Untriggered search
aj      http://www.amazon.co.jp/gp/search/?index=blended&keywords=%s&_encoding=utf-8 	Amazon.co.jp Search
gj      http://www.google.co.jp/search?q=%s          Google Japan Search
sc     http://scholar.google.com/scholar?q=%s        Google Scholar

Reflecting the Revision to the Script

Firstly, upload the quix.txt file to a web server. If you don’t have any account, you can’t use the Public folder of Dropbox, which I think is the most wonderful idea. It’s the easiest way to edit and upload. Quix offers a simple script to reflect the revision. Go to the following site. http://quixapp.com/extend/

Integration of Google Chrome and quix

Once you made your custom script, it’s time to integrate it with Chrome. See the following site.

1. Open the script you’ve got in the previous step with a text editor. Modify it as follows.

2. Install the script to Google Chrome. You can see a detailed instruction in the above mentioned website.

3. Enjoy quix!

Note. If you wish to revise commands after all those processes, you can do it just by modifying quix.txt on your server. After revising and uploading, execute the “debug” command on the “quixed” address bar. You can see that the new commands will be working.

Some transcription from the interview with Ed Catmull, the president of Pixar, held by The Economist.

The notion that you’re trying to control the process and prevent error screws things up. We all know the saying it’s better to ask for forgiveness than permission. And everyone knows that, but I Think there is a corollary: if everyone is trying to prevent error, it screws things up. It’s better to fix problems than to prevent them. And the natural tendency for managers is to try and prevent error and over plan things.