App Delegate Inc

App Engine ClientLoginError

2011-08-28T00:00:00Z

I’m using a pretty specific set of tools and I’ve come across an error that I couldn’t find a solution to, so I hacked my way out of it. I’m using django-nonrel to run Django on App Engine, while preserving some important features, like the admin app. I’m also using a Google Apps account (xxxx@appdelegateinc.com) as my App Engine account and I do my development on a Mac. When I try to run python manage.py remote createsuperuser to create a new user on the App Engine hosted app, I get the following error:

google.appengine.tools.appengine_rpc.ClientLoginError: HTTP Error 403: Forbidden

Full Traceback

I followed the traceback down to an App Engine SDK file which seems to not set a variable correctly. So I opened it up and hacked and got it working correctly.

edit

/Applications/GoogleAppEngineLauncher.app/Contents/Resources/GoogleAppEngine-default.bundle/Contents/Resources/google_appengine/google/appengine/tools/appengine_rpc.py

And change the line:

account_type = "GOOGLE"

account_type = "HOSTED_OR_GOOGLE"

Here's the diff

/google/appengine/tools/appengine_rpc.py
206c206
<         account_type = "GOOGLE"
---
>         account_type = "HOSTED_OR_GOOGLE"

Where the heck is me??

2011-07-17T00:00:00Z

I was trying to figure out what the coordinates of my home were, but I couldn’t find anything online to look it up. I was hoping for something like whattimeisit.com or whatismyip.com but for my location. Since it’s nowhere to be found, I decided to write up something for myself. It’ll show a map of where you are using your browser’s location information (if supported) and display your coordinates along with your best guessed address. Try it out over at wheretheheckis.me. I hope someone finds it useful. Maybe to help test out my Google Maps Circle Overlay library.

wheretheheckis.me

Pocket Sentry Caught Some Lightning

2011-06-15T00:00:00Z

I snapped a couple nice photos using my Pocket Sentry app last Thursday during a storm. Here they are:

Pocket Sentry Motion Cam

2011-04-13T00:00:00Z

I just released Pocket Sentry, a motion activated camera for the iPhone and all other Apple devices equipped with a camera. The key features are:

Play a configurable sound when motion is detected
Snap a photo when motion is detected
Email a photo (in batches) when motion is detected
Request a webpage (HTTP Callback) when motion is detected

This is great if you have trouble with your neighbors, like I have, where they walk their dog on your lawn. It’s also good to catch the weirdo digging through your garbage. I originally wrote it as a prototype to keep the cat out of the stroller when we were getting ready for our baby to be born and it worked fantastically as you can see from this early prototype:

The quality is much better than that now though, here’s a snip from an email in the current version:

If you want to implement the callback, here’s a sample in PHP:

<?php
$device_id = $_POST["udid"];
$lat = $_POST["lat"];
$lng = $_POST["lng"];

if(array_key_exists("photo", $_FILES)) {
    $file = $_FILES["photo"];
    if($file["error"] == 0) {
        $size = $file["size"]; // Bytes
        $type = $file["type"]; // image/jpeg
        move_uploaded_file($file["tmp_name"], "/path/to/new/location");
    } else {
        // There was an error
    }
}
?>

Update that code to post the image captured to twitter and then update the callback URL in Pocket Sentry’s settings to that URL and you now have a twitter feed of all the wildlife in your back yard. Have fun!

Download Pocket Sentry Now:

RFID with Arduino and Python!

2010-10-06T00:00:00Z

This is a follow up to my initial post about using the Parallax RFID reader with the Arduino microcontroller. We’re going to spice things up by making your computer yell at intruders! With Python! I’ve opened the whole project up on GitHub at http://github.com/mattwilliamson/Arduino-RFID.

This is an RFID authentication system. It allows a person to place an RFID tag within proximity of the reader device (usually 5-6 inches with this reader) and it will look up the tag in an SQLite database to check if it should grant access. If authorization is granted, the hardware device will turn a servo motor into a second position. The primary application of this being the unlocking of a door. This implementation also speaks aloud “Access Granted” or “Access Denied” if you are running the server on OS X. You can modify the source to use something like flite on other OSes. This project is very easily modified to do any sort of task. Some ideas are:

Door locking (default implementation)
An employee punch-in/punch-out system
Real world item identification (affix a tag to every day items, such as your wallet; when scanned, open a web banking web page. attach one to your phone to initiate a Google Voice call, etc.)
Russian roulette with RFID tags instead of guns

Here's a Demo:

Hardware

To make this RFID system, you’ll need the following hardware:

1 Arduino
1 Servo (I like Hobby Partz because they are cheap) you don’t need this if you are not unlocking a door
1 RFID Reader module (buy from your local Radio Shack or Parallax) $8 if you get lucky and find one at Radio Shack
Some jumper wires to wire the whole thing together

Software

Python (2.6 tested, should work on 2.7)
pySerial
Arduino IDE
NewSoftSerial
The Arduino sketch and the rfid.py file included with this README file

Wire it Up

You can use the image schematic/reader_breadboard.png to reference the wiring diagram. Basically you just need to wire the pins as follows:

Arduino pin 8 -> RFID pin SOUT
Arduino pin 9 -> RFID pin /ENABLE
Arduino pin GND -> RFID pin GND
Arduino pin 5V -> RFID pin VCC
Arduino pin GND -> Servo wire BLACK
Arduino pin Vin (9v) -> Servo wire RED
Arduino pin 3 -> Servo wire YELLOW

You should end up with something like this.

Program the Arduino

Download the project here: http://github.com/mattwilliamson/Arduino-RFID

First you need to install the NewSoftSerial library for Arduino. Read the Arduino docs on how to install a library.

Plug the Arduino into your computer and open the sketch sketch/rfid_sketch.pde with the Arduino IDE. Select which serial port to use for the Arduino (Tools->Serial Port on OS X). Click the upload button to write the sketch to your Arduino. You may need to try a few different ports before it works. Write down the port name, we’ll need it later. It should look something like /dev/tty.usbserial-A70064Mh or COM3. It will probably be different for your machine.

Set Up the Server

Install the pySerial serial library for python. If you have setuptools installed, you can just run easy_install pyserial.

Open rfid.py with a text editor and look for this line: serial_port = '/dev/tty.usbserial-A70064Mh'. Change the /dev/tty.usbserial-A70064Mh to the serial port you wrote down in the previous step.

Usage

First, we’ll run the server to watch for the RFID tags we want to allow. Open a terminal and cd to the project directory. Then run python rfid.py serve. This tells the python server to listen to the serial device for RFID tags. It will print debug information into the console.

wraith:rfid mwilliamson$ python rfid.py serve
Starting serial server...
Connected to arduino. Awaiting RFID scans.
Checking tag: 0F03042E80
Tag access denied.
^C

Hit ctrl-c to kill the program. Now we have a tag we can enable: 0F03042E80. We’ll use the enable command to enable it.

wraith:rfid mwilliamson$ python rfid.py enable 0F03042E80
Added tag with access ENABLED

Now that we’ve added the tag, we can scan it successfully. You should see the servo spin, or a least the built in LED on pin 13 of the arduino will light up.

wraith:rfid mwilliamson$ python rfid.py serve
Starting serial server...
Connected to arduino. Awaiting RFID scans.
Checking tag: 0F03042E80
Tag access granted.

You can revoke access to a particular tag by running the disable command

wraith:rfid mwilliamson$ python rfid.py disable 0F03042E80
disabled access for tag

Errors

If you see the following, make sure your Arduino is plugged into your computer and you correctly set up the serial port in rfid.py

wraith:rfid mwilliamson$ python rfid.py serve
Starting serial server...
*ERROR*
Could not open serial port.
Edit rfid.py and make sure you define serial_port properly.

Download or view code at: http://github.com/mattwilliamson/Arduino-RFID

Motion Tracking with a Webcam

2010-08-02T00:00:00Z

I used Python and OpenCV to get a simple app together that tracks movement using a webcam. It currently averages all of the movements’ locations to determine a single point for targeting. This will not work for multiple targets, with a little modification.

Here's the code from http://github.com/mattwilliamson/Motion-Tracker/blob/master/track.py:

# Derived from http://sundararajana.blogspot.com/2007/05/motion-detection-using-opencv.html

import cv

class Target:
    def __init__(self):
        self.capture = cv.CaptureFromCAM(0)
        cv.NamedWindow("Target", 1)

    def run(self):
        # Capture first frame to get size
        frame = cv.QueryFrame(self.capture)
        frame_size = cv.GetSize(frame)
        grey_image = cv.CreateImage(cv.GetSize(frame), cv.IPL_DEPTH_8U, 1)
        moving_average = cv.CreateImage(cv.GetSize(frame), cv.IPL_DEPTH_32F, 3)
        difference = None

        while True:
            # Capture frame from webcam
            color_image = cv.QueryFrame(self.capture)

            # Smooth to get rid of false positives
            cv.Smooth(color_image, color_image, cv.CV_GAUSSIAN, 3, 0)

            if not difference:
                # Initialize
                difference = cv.CloneImage(color_image)
                temp = cv.CloneImage(color_image)
                cv.ConvertScale(color_image, moving_average, 1.0, 0.0)
            else:
                cv.RunningAvg(color_image, moving_average, 0.020, None)

            # Convert the scale of the moving average.
            cv.ConvertScale(moving_average, temp, 1.0, 0.0)

            # Minus the current frame from the moving average.
            cv.AbsDiff(color_image, temp, difference)

            # Convert the image to grayscale.
            cv.CvtColor(difference, grey_image, cv.CV_RGB2GRAY)

            # Convert the image to black and white.
            cv.Threshold(grey_image, grey_image, 70, 255, cv.CV_THRESH_BINARY)

            # Dilate and erode to get object blobs
            cv.Dilate(grey_image, grey_image, None, 18)
            cv.Erode(grey_image, grey_image, None, 10)

            # Calculate movements
            storage = cv.CreateMemStorage(0)
            contour = cv.FindContours(grey_image, storage, cv.CV_RETR_CCOMP, cv.CV_CHAIN_APPROX_SIMPLE)
            points = []

            while contour:
                # Draw rectangles
                bound_rect = cv.BoundingRect(list(contour))
                contour = contour.h_next()

                pt1 = (bound_rect[0], bound_rect[1])
                pt2 = (bound_rect[0] + bound_rect[2], bound_rect[1] + bound_rect[3])
                points.append(pt1)
                points.append(pt2)
                cv.Rectangle(color_image, pt1, pt2, cv.CV_RGB(255,0,0), 1)

            num_points = len(points)
            if num_points:
                # Draw bullseye in midpoint of all movements
                x = y = 0
                for point in points:
                    x += point[0]
                    y += point[1]
                x /= num_points
                y /= num_points
                center_point = (x, y)
                cv.Circle(color_image, center_point, 40, cv.CV_RGB(255, 255, 255), 1)
                cv.Circle(color_image, center_point, 30, cv.CV_RGB(255, 100, 0), 1)
                cv.Circle(color_image, center_point, 20, cv.CV_RGB(255, 255, 255), 1)
                cv.Circle(color_image, center_point, 10, cv.CV_RGB(255, 100, 0), 5)

            # Display frame to user
            cv.ShowImage("Target", color_image)

            # Listen for ESC or ENTER key
            c = cv.WaitKey(7) % 0x100
            if c == 27 or c == 10:
                break

if __name__=="__main__":
    t = Target()
    t.run()

Arduino Sonar

2010-07-27T00:00:00Z

I’ve been waiting 3 weeks for my ultrasonic rangefinder to come in and now it has arrived. It’s going to be used in an obstacle avoiding tank-tread robot. Here’s a quick little test.

Browser Based Map Reduce

2010-05-18T00:00:00Z

What if you could offload some of your computational work onto the visitors of your site? We thought it would be pretty cool to make a quick little app to test the theory out using Google App Engine.

We opted to use the MapReduce paradigm as made famous by Google. The workflow is like this:

Admin (you) create a new job
Admin adds data set, a line break delimited string
Admin adds a map function, written in pure javascript
Admin adds a reduce function
Client visits web page with javascript embedded
Javascript library prompts the user to contribute CPU cycles and writes to a cookie to remember
If user accepts, request a random function and data set item from all jobs that is not complete
Execute job and submit result
Request another job, if there is none, wait a while and check again, if there is, don’t delay and start step 8 again
It works like a charm! Try it out at http://bbmapreduce.appspot.com/. You can find the sources on my GitHub project page. To allow your visitors to contribute CPU cycles, simply include the script tag at http://bbmapreduce.appspot.com/client/sample in your website.

Sample:

<script src="http://bbmapreduce.appspot.com/client/wrapper/"></script>

It should be noted that was not written with security in mind and you can never trust the data that gets sent back from your visitors, but there are ways to make it more reliable, such as executing one function on multiple clients and checking the data returned against the rest which is not currently implemented. The other issue is that the jobs being submitted on this test site is completely open and thus prone to spam alert calls.

DISCLAIMER

I am not responsible for any harm this causes you, your computer, or anyone who visits the URLs provided. This is simply for educational purposes.

Google Maps Circle Overlay

2010-05-17T00:00:00Z

UPDATE: By popular request, I’ve ported this to Google Maps v3

Another feature missing from Google Maps is the ability to draw a circle overlay. Displaying radii around a location is a very common use of maps. If your website has store searching and you want to show which stores are within X miles of some zip code, you’ll want to show your users the radius on the map. This can also be used in conjunction with our point in polygon library to easily display which points are inside the circle.

Sample:

I made this library in the most logical way. I used a GPolygon and drew a fixed number of points in a circular fashion using standard geometry. In testing, I found that the optimal number of points drawn in the circle for general purposes is 40. You can easily override this using the last parameter of the constructor. Use a higher number for a smoother circle or a lower number for a higher performance circle.

Usage (default resolution):

var circle2 = new CircleOverlay(map, map.getCenter(), 50, \
    "#0000FF", 1, 1, "#0000FF", 0.25);

Usage (custom resolution):

var circle1 = new CircleOverlay(map, map.getCenter(), 100, \
    "#000000", 1, 1, "#009900", 0.25, 100);

Per usual, you can view a sample in use or view and download the code. Use freely.

Point in Polygon Checking With Google Maps

2010-05-16T00:00:00Z

The Google Maps API is missing some things. One thing it is missing is the ability to detect whether or not a point lies within a polygon. One use case for this is selecting friends in a social network using a polygon the shape of a state or county.

Sample:

To add this ability, I extended the GPolygon class with a new method called containsLatLng, just like the GLatLngBounds class. The difference is that the method in the GLatLngBounds class will only get the 4 coordinates of the bounding box, which are the farthest points in the polygon, and see if the coordinate lies within that box. My method, however, does true collision detection to see if the point lies within any shape polygon with as many sides as you want. It worked out perfectly using the Ray Cast Algorithm.

The simplest usage of this library looks like this (omitting including the GMaps API and this library):

var coordinate = new GLatLng(40, -90);
var polygon = new GPolygon([], "#000000", 1, 1, "#336699", 0.3);
var isWithinPolygon = polygon.containsLatLng(coordinate);

You can view a sample in action or download/view the source. Use freely.

Update: Tim Parkin kindly updated this library to work with Google Maps v3, check it out at http://github.com/tparkin/Google-Maps-Point-in-Polygon