I decided to create a Birthday Module to learn the Google Homepage API. I think that it's a handy widget to have on your home page, but I admit that entering the birthday data is a little difficult because you have to enter all of it in one box. Even though Google-provided modules, such as the Stock Market module, allow you to support a variable number of items through the module's preferences, that capability is not available to modules created with the Homepage API. (Hence, one box with a lot of data delimited by semicolons.)
Also, I have bundled a small Easter Egg with the Birthday Module -- try to be the first to find it and figure out how I hid it!
Saturday, January 21, 2006
Sunday, January 15, 2006
Progress on TargetAlert 0.9
Last night, I made a considerable amount of progress on TargetAlert 0.9. In my last post, I mentioned that things were going slow because I was getting bogged down with XML serialization; however, I recently abandoned that approach for storing user prefs in favor of using JSON. This allowed me to dramatically reduce the amount of boilerplate serialization code I had to write, so hopefully I'll release TargetAlert 0.9 shortly after Chickenfoot 0.6 is out the door (which is more relevant than you might expect).
Sunday, January 8, 2006
TargetAlert 0.8.9.7 Released
I've received quite a bit of email from Firefox 1.5 users asking me to restore the option to have TargetAlert icons appear after a link when a user mouses over it, so I have done so in TargetAlert 0.8.9.7. (In 0.8.9.6, instead of displaying the icon at the end of the link, it would transform the cursor into the TargetAlert icon, which a number of users found annoying.) Ultimately, I would like to provide users with the option of converting the cursor, but simply rolling back the change was the quickest fix. This option is now planned for TargetAlert 0.9.
From the TargetAlert Roadmap, it may appear that everything is planned for version 0.9, as it has been for some time. You may be wondering why there hasn't been much progress. Well, the main reason is that the bulk of the code that needs to be written is not very much fun to write. Serializing XML and error-checking user input just aren't my idea of a good time. (Please remember that I work on TargetAlert for free, as a hobby, and that will continue to be the case until some people start clicking on my PayPal Donate link!)
Besides, I've recently been working on some other things that I find more entertaining.
From the TargetAlert Roadmap, it may appear that everything is planned for version 0.9, as it has been for some time. You may be wondering why there hasn't been much progress. Well, the main reason is that the bulk of the code that needs to be written is not very much fun to write. Serializing XML and error-checking user input just aren't my idea of a good time. (Please remember that I work on TargetAlert for free, as a hobby, and that will continue to be the case until some people start clicking on my PayPal Donate link!)
Besides, I've recently been working on some other things that I find more entertaining.
Sunday, January 1, 2006
Google Web Search API in JavaScript
When I got back from India and started hacking on Firefox, I wrote a Python script to generate the JavaScript necessary to access the Google Web Search API as part of Google Dominoes. I never took the time to clean up the JavaScript code and publish it online before, but now that we have Chickenfoot, which makes it much easier to experiment with JavaScript APIs, I decided that it was finally time to make it publicly available. The Python script for converting the WSDL is called wsdl2js, and the JavaScript for the Google Search API is in Google Search.js. Happy hacking.
Subscribe to:
Posts (Atom)