How to empty the App Inventor “Backpack”

You know about the App Inventor Backpack?

You can drag blocks into the “Backpack” icon at upper right of the MIT App Inventor Blocks editor. Once in the backpack, you can later drag items out of the backpack  providing a nice way to move blocks around or between apps.

Until recently, the backpack was automatically emptied when you exited App Inventor. But … that has changed recently!

A couple of months ago the Backpack behavior was changed so Backpack content is  persistent. That means the backpack retains a copy of your blocks even when you exit App Inventor.

This makes it easier to move favorite code blocks between apps or apps you download from the MIT Gallery.

To empty the backpack, you need to right-click (Ctrl-click on a Mac) somewhere in the whitespace (background) and then select Empty the Backpack from the pop up menu.

Capture

QuickSort routine for your App Inventor Apps!

Sorting data is a common need in many applications. If you have thought about writing your own sorting routine, you likely came up with a scheme that picked the first data item, and then compared it to each of the other elements to find a smaller value (if sorting into ascending order) or a larger value (if sorting into descending order). Finding a match, you switch the two values in your list of data.

Gradually, as you scan the list, the smallest (or largest) value bubbles up to the front of the list. Once the whole list is scanned, you have found the smallest (or largest) value and switched it to the beginning.

Then you would start with the next data value and compare it to the remaining data values. And so on.

Great idea if you came up with this – seriously! In computer science algorithms, this is known as the “bubble sort”. It works fine – but it has one problem – as the number of items to be sorted grows, the time it takes to do the sort grows even faster.

In fact, we say that a bubble sort runs in a time proportional to n^2 (n squared). If there are 10 items to sort, this will take 10^2 or 100 time units to sort.

If there are 30 items to sort, this will take 30^2 or 900 time units to sort. Just going from 10 items to 30 items adds 800 time units to our sorting time! Ouch!

As you can see, the bubble sort is simple but it can take a long time to run!

Computer scientists have invented other ways to sort data. One of the best known has the descriptive name “QuickSort”. In many cases, it’s sorting time is on the order of n ln n (that is n times the natural logarithm of n).

If we compare that to the Bubble Sort for n=10, we get 23 time units and for n=30, we get 103 time units. As you can see, QuickSort is much faster than the Bubble Sort.

Continue reading

appinventor.pevest.com has moved to appinventorplus.wordpress.com

The old URL http://appinventor.pevest.com automatically redirects to the new URL at https://appinventorplus.wordpress.com

The old RSS newsfeed at appinventor.pevest.com/?feed=rss2 automatically redirects to the new RSS newsfeed at https://appinventorplus.wordpress.com/feed/

If you subscribed to the blog via email, you must re-subscribe by going to the new blog URL and clicking on the Follow link near the top of the rightmost column.

If you read this blog via Facebook or Twitter, you do not need to do anything. Tumblr is not yet configured but will be soon.

Old page links, such as those on the Facebook group page like: http://appinventor.pevest.com/?p=1579

will continue to bring up the old pages, making it easy to access older material.

Most comments – except those entered in the past few days – were migrated to the new blog URL. Comments entered in the past few days (after the content was migrated but before the switch over was completed) appear here but not on the new blog.

Benefits of the new web site

  • FASTER web site access and display of web pages
  • SECURE web site access using https – this prevents Internet Service Providers, government agencies and others from spying on what you are reading online.
  • AUTOMATIC SHARING to Facebook, Twitter, Google+ and soon Tumblr. Previously I had to manually cross post items to Facebook and Google+ – and I was rarely getting caught up with Google+. Now the updates will be automatic!
  • EASIER UPDATES as I consolidate my blogs into WordPress, I will have a single interface to reach any of them, which makes publishing easier and faster.
  • MORE TIME for me as I no longer have to maintain the continuous software updates and debugging of multiple web sites. That means more time to post new App Inventor tutorials!

The only feature we lose – I think – is the summary page that automatically created a 2 or 3 line summary of each post and placed those into a single page.

Web site migration update and info on next tutorial/sample code

I posted this item last night, but took it off line as I discovered problems with my web site migration. I stayed up much of the night fixing that – and then we had no electrical power most of today due to a wind storm.
—————————
I completed a migration of my coldstreams.com Internet of Things blog to WordPress. This is a test case before I flip the switch on appinventor.pevest.com redirecting to appinventorplus.wordpress.com.

The process was more complex than I expected, especially getting the automatic redirection to the new web site working as intended. However, with that process understood now, the migration of App Inventor 2 – Learn to Code to the new server will hopefully be able to happen soon and go more smoothly. If you would like to learn how to do this and what I had to do, I’ve written this up to help others: https://coldstreams.wordpress.com/2017/04/08/how-to-move-a-self-hosted-wordpress-blog-to-wordpress-com/

Once these migrations are done, I will be able to focus on creating content and no longer have to spend so much time managing servers!

I expect my next tutorial to be on how to sort data in App Inventor.

A reader has a collection of data with both a name, and a score (it is something else but it is like a score) and would like to sort the collection sometimes by name and sometimes by score.

There are several ways to sort data – some are obvious but slow while some are really fast. My goal is to show how to implement a fast sorting algorithm that works with a list of lists.

In this concept, we will have a list of data, where each element in the list is itself a list. Sounds confusing but the idea is that we can store a data record in each list element and lists are how App Inventor stores records. This means our list will look something like:

Chang, 89
John, 86
Daksha, 92
Samantha, 91
Chandra, 90
Johan, 87
Patty, 94

When we sort by name, we want Chandra as name 1, then Chang as name 2, Daksha as name 3 and so on, with the corresponding scores still attached to their names.

We will do this with our sorting code.