Occasional blank pages

If you see a blank page when refreshing your app in a browser after a source code update, be it on your local machine or after the upload to Google, but then with any subsequent refresh the page appearing just fine, check the Python script executed for given URL (see your app.yaml) to make sure that you are invoking the main() function correctly, i.e. something like the notorious

if __name__ == '__main__':
	main()

If you make a typo in the above snippet or should you omit it completely, you will experience the problem above.

Update: See the comments below this article for explanation of the reason of this behavior.

Static files: CSS not loading

At some stage of working on the project I introduced the CSS style sheet for the site: added a static_dir to handlers section of the app.yaml file and then created the CSS file and stored it in the directory referred to by static_dir. Then, I simply referenced the CSS file in the HTML. So my app.yaml looked like

application: corporatepoems
version: 5
runtime: python
api_version: 1

handlers:
- url: /css
  static_dir: css

- url: /.*
  script: corporatepoems.py

…and the HTML contained:

<link rel="stylesheet" href="/css/corporatepoems.css" type="text/css" media="screen" />

However, the CSS would not load. I double checked the file and directory names, all was correct. The fix to the problem was to actually increase (or chage) the application version in the app.yaml file. After that, CSS loaded as expected. The explanation to this could be that I perhaps forgotten to save the app.yaml file after adding the static_dir at first place, but I’m pretty sure this was not the case. After the fix, I never saw the problem again.

All in all, working with Google App Engine and Python feels great so far, thanks Filip for your suggestion to give it a try! To be continued.

It is simple to setup a Gmail account instead of Exchange ActiveSync account for your Google Apps account. Just use your personalized e-mail address username@yourdomain.com instead of Gmail address. This will however not help to sync your Google Calendar events.

To get my Google Calendar events to my iPod touch I set up a CalDAV Account. When adding an account on your iPod / iPhone, in the main screen listing all account types (Microsoft Exchange, mobileme etc) select “Other”, then “Add CalDAV Account” and fill in the form which follows. Server value should be www.google.com. You can now not only download the Calendar events to your device, but also add events on your iPod touch / iPhone and have them synced to your Google Apps Calendar.

This solution does not offer all the comfort for Google Apps account, but at least enables the e-mail and Calendar events synced both directions while the functionality of the Exchange account remains unchanged.

There was nothing extraordinary about the photograph of the Paradiso club, other than the feeling I had. I somehow felt that I saw this photo before. And yes indeed, I not only saw the photo before, I actually took it.

It is funny when I think about it because TEDx inspired me to actually do something about my photographs so that some of them would perhaps get published or used by someone else. And only by doing this work I found out that one of them was actually used already. And if I didn’t go to TEDx, which motivated me to work on the photos, I would have probably never found out that a photo was already used. Can this be called a synchronicity?

NB: Thanks to the magazine editors for using the photo fully according to the license under which it was published.

I had a privilege and honor to be invited to attend the TEDx event in Amsterdam last week. Although a long-time subscriber to podcast at TED website, I was curious what the character of the first-ever Dutch TEDx event would be. Open, inspiring and forward looking? Or just a chique networking event for the “lucky few” wrapped in a popular image of TED?

My fears dissolved as soon as the first speaker, Frans Timmermans began his inspiring speech on how embracing our fears can unleash creativity, can be fun and can lead towards resolution of complex problems. Within moments and then until the end of the day the conference room was charged with inspiration, hope and optimism, with sincere determination of the speakers and attendees not only to spread the ideas but also to act towards making them happen. I liked how many ideas were directly related to environmental sustainability of life on our planet. And not only ideas, but also concrete products were demoed, I personally loved for instance the product Wattcher, giving you instant information on how much energy your household consumes at any given moment.

And not to forget… the humor! Speakers like the Dutch photographer Hans Aarsman or Danish architect Bjarke Ingels provided an insightful look into their work with humor which was impossible to resist. According to the organizers, there will be a speech recording posted on YouTube every week, sure look them up there. Apropos, the organizers. I was amazed how much care, energy and attention to detail was put into organizing the event. Already the guest list selection process must have been a headache, just imagine that for every attendee there were ten of those whom the organizers had to disappoint.

Although not without a bit of preaching, TEDxAmsterdam was a very inspiring event with a potential to change the course of one’s life and thinking. It felt like attending a 20 course diner for your brain. I would like to thank the organizers for making the TEDx in Amsterdam happen, I’m looking forward to the 2010 edition. I won’t attend though, so that those who were not invited this year, have a greater chance next year. Hey, and I’m getting that Wattcher thing, not only to talk about making something happen for the planet, but also to act.

NB: There was not much room for snapshots during the day packed with talks and demos, so just one photograph: Kevin Kelly (Wired magazine) speaking.

Update: Frans Timmermans’ speech, mentioned above, is now available on Vimeo.

Download mod_wsgi

Python WSGI is a specification of interface between web servers and Python applications or application frameworks.  The mod_wsgi is an Apache module, using which it is possible to access Python applications (which adhere to WSGI speficiation) via http.

The mod_wsgi can be downloaded from http://code.google.com/p/modwsgi/.

Compile and install mod_wsgi

Versions used: Python 2.5.1, apache 2.2.11, mod_wsgi 2.3

Unpack the mod_wsgi and run the usual:

$ ./configure
$ make
$ sudo make install

On my Mac, the module was installed to /usr/libexec/apache2, I had to copy it manually to XAMPP location:

$ sudo cp /usr/libexec/apache2/mod_wsgi.so /Applications/xampp/xamppfiles/modules/

Write a sample Python WSGI application

The next step is to create a directory where the Python application will be stored. Unlike with PHP scripts, Apache server needs to be made aware of exact location of the app. We also have to tell Apache which particular script needs to be executed for any http request which points to the application directory. This is also very different from what you might be used to with PHP. To be able to do all the configuration changes in httpd.conf, we first need to create a directory and to create the main application script so we can write down all directory and file names required.

[ivk@ivk-mbpro /Applications/xampp/xamppfiles/htdocs]$ mkdir app-sample
[ivk@ivk-mbpro /Applications/xampp/xamppfiles/htdocs]$ cd app-sample
[ivk@ivk-mbpro /Applications/xampp/xamppfiles/htdocs/app-sample]$ vi main.py

main.py can be for instance a simple Hello World! app:

def application(env, start_response):
    start_response("200 OK", [])
    output = "<html>Hello World! Request: %s</html>"
    output %= env['PATH_INFO']
    return [output]

Put it all together

Modify the httpd.conf by adding the following:

LoadModule wsgi_module modules/mod_wsgi.so

WSGIScriptAlias /app-sample "/Applications/xampp/xamppfiles/htdocs/app-sample/main.py"
<Directory "/Applications/xampp/xamppfiles/htdocs/app-sample">
    Order deny,allow
    Allow from all
</Directory>

WSGIScriptAlias directive makes sure that for all requests to /app-sample, the "/Applications/xampp/xamppfiles/htdocs/app-sample/main.py"will be executed by Python.

Modify mime.types by adding “py” as follows:

text/html        html htm py

After Apache restart you should be able to access your python script main.py via http:

Tools

There are no screws on Mac Mini case which means that you will have to use a tool to open your computer. Once opened, you will need to remove a few small screws so appropriate small screwdriver will be required. The screws inside are a bit hard to reach, I found it very important to use a small magnet tool to be able to remove the screws.

Opening Mac Mini

Disconnect all cables from your Mac and turn it belly up. There are no screws or handles to help us to remove the case so I used a stainless flex tool to open up the computer. Make sure the tool is clean when used before by a craftsman. To open up the computer you need to carefully insert the tool in between the case and the body of the computer, one side after another except the one where all connectors are located (rear panel). Once you insert the tool, apply pressure carefully to pop the computer body up.

Using a screwdriver is not recommended, this could result into visible marks on the computer’s plastic body. This step calls for some patience and maybe a few repeats. I had my cat Lennon sitting on a pillow near by to help me keep relaxed and calm. A scratch or two will likely be result of this action.

Memory modules

What you will see once you open the computer is that you do not see the memory modules. Yes, there are buried inside and we need to do some more work to reach them. Your Mac now looks something like this:

Continue by removing the upper part of computer’s body which consists of the optical drive, hard drive and the CPU fan. To do so, we have to remove four screws. These screws, black colored, are located in each corner of the body. You can see three of them. Once unscrewed, use a magnet tool (or magnetic screwdriver) to remove them. I found it impossible to remove the screws otherwise (apart of flipping the whole computer upside down for a second). Fourth screw is hidden under the Wi-Fi antenna:

Remove the Wi-Fi antenna by gently pressing on two plastic clips below the antenna and pulling the antenna (gently) at the same time. Antenna will still stay connected to the computer by a cable. After removing the antenna, you will see the fourth screw. After removed, also unplug the cable which connects the upper part of the computer’s body with the rest of it:

Gently lift the upper part of the computer’s body. At one side, the upper part is still connected to the rest of the computer so you will simply have to flip the upper part on one side. This will open the access to memory modules, you can now easily exchange them:

What happened to me at this point was that my Wi-Fi antenna got disconnected from the computer. I thought that something got seriously broken. Luckily, the antenna just only got unplugged, it was no problem to put it back. Look-up the wireless adapter card on the board (on my Mac it was called AR5BXB6) and plug the antenna back in (connector MAIN). The lose antenna and the wireless adapter on the main board look something like this:

Assembling your Mac back together is simple, just follow the above steps in reverse order. I did power-up the computer once before putting the computer back just to avoid having to open it again in case something went wrong. The first experiences after the upgrade (i.e. writing this article) are very positive, there is no memory swapping to disk which I experienced when processing multiple (large) images with GIMP and having many tabs in Firefox opened at the same time, my old trusty Mac runs much smoother.

Small print

This article describes the memory upgrade for Apple Mac Mini 1.83 GHz Intel Core 2 Duo. Opening up your computer voids your Apple warranty and might lead to permanent damage to the computer. Do not try this at home unless you know what you are doing. It is advised to have this done by an authorized service company. Contact Apple for detailed options.

Barrie now tours with Primal Scream. I saw him and his band a couple of times before in Amsterdam and even ended up backstage with the band. Apart of finishing all the Heineken in the band’s fridge, we left the club with a beermat full of musical tips from Barrie.

I keep the beermat until now, to remind me of that rock’n'roll night. Primal Scream were great last night and Barrie (on the photo) rocked.

Primal Scream @ Paradiso, Amsterdam, 8 August 2009 on my Flickr

-rw-r--r--   1 ivk  staff  499 Jan 18 23:47 Info.bckup
-rw-r--r--   1 ivk  staff  499 Jan 18 23:47 Info.plist
drwxr-xr-x  18 ivk  staff  612 Jan 18 23:47 bands
-rw-r--r--   1 ivk  staff    0 Jan 18 23:47 token

Finder however still displayed the same message, even after removing the mistakenly copied files. To me it appeared as if some kind of flag was set on the home directory, instructing applications which read the flag to act as if my home dir was not a regular directory (anymore).

The actual resolution of the problem was to run SetFile utility which sets (or resets) attributes on files or directories:

$ /usr/bin/SetFile -a b /Users/ivk

Note: The SetFile man page does not list attribute b (or B) amongst those that can be applied to directory, the command however worked well. Used on HFS+ filesystem (Mac OS Extended), on OS X 10.5.6.

 
Download the XCode project (CommandLineSample.tar.gz, 48kB) – including a sample

 

The CommandLine class is very simple and offers the following functionality:

• Validates and parses the command line arguments according to user defined struture of options (NSDictionary)
• Options can be declared required or optional, followed by option value or not
• Minimum number of expected parameters (beside options) can be set
• When command line parsing fails, a message describing the error can be retrieved

A sample use of the CommandLine class:

#import <Foundation/Foundation.h>
#import "CommandLine.h"

int main (int argc, char * argv[]) {
    NSAutoreleasePool * pool = [[NSAutoreleasePool alloc] init];
    BOOL parseSuccess;

    CommandLine *commandLine =
        [[CommandLine alloc] initWithArgc: argc andArgv: argv];

    // This sample expects at least one parameter (other that options)
    int sampleMinNumberOfParams = 1;

    // Define the matrix describing expected options
    NSDictionary *sampleOptionsMatrix =
        [NSDictionary dictionaryWithObjectsAndKeys:
            CL_OPTION_REQUIRED, @"a",
            CL_OPTION_OPTIONAL, @"b",
            CL_OPTION_REQUIRED_WITH_VALUE, @"f",
            CL_OPTION_OPTIONAL_WITH_VALUE, @"x",
            nil];

    // Pass on required parameters to the parser and execute the parser
    [commandLine cL_setOptionsMatrix: sampleOptionsMatrix
    andMinNumberOfParams: sampleMinNumberOfParams];

    parseSuccess = [commandLine cL_parse];

    if (parseSuccess) {
        // Display the parsed options.
        for (NSString *key in commandLine.parsedCommandLine) {
        NSLog(
           [NSString stringWithFormat: @"Option: %@ Value: %@", key,
           [commandLine.parsedCommandLine valueForKey:key]]);
        }
        //And some more processing of a successfully parsed command line:
        if ([commandLine cL_optionIsSet: @"b"]) {
            NSLog(@"Option b was set");
        }
        NSLog(
            [NSString stringWithFormat: @"Value of the 1st parameter is %@",
            [commandLine cL_parameterGetValue: 1]]);
    } else {
        //Display the error message
        NSLog(commandLine.resultText);
    }

    [pool drain];
    [commandLine release];

    return 0;
}

A few notes:

• Alternative long option names are not supported
• The options matrix appears to be somewhat verbose, hope it is not too much overhead
• The CommandLine class source code enables some documentation to be generated using objcdoc

Feel free to use and modify the class in any way. Now off to the next step: learning the Google Data APIs Objective-C Client Library.