iCalendar feeds

July 21, 2007

Last weekend I sat down and implemented iCalendar feeds for your projects. There are several to select from, both personal and for all people in your ClockingIT company.

After finding the icalendar gem for Ruby and reading a bit of documentation, implementation was a breeze. What took most time was finding an export format that contained correct information so that Google Calendar could parse it. It seems that if you use an end time of less then an hour, Google Calendar sets it to one hour, but if you specify a duration it gets it right. The gem also output the wrong tag for a VTODO percentage, but that was easily fixable with a gsub.

The end result is that Google Calendar now polls my feeds once a day, and I get a nice view of the work I log and the tasks I close. And as I split different information into different feeds, I can filter the view easily to see just the information I’m looking for.

Google Calendar Feeds

Update time…

July 9, 2007

It’s been a while since I’ve written any thing here, so here’s a short run down of what’s happened to ClockingIT lately.

  • Tasks can have dependencies
  • Tasks can repeat daily, weekly, monthly, on a specific day, on a specific day of a specific week, or after a number of days
  • Files are no longer stored in the database, making for easier backups and allowing for larger uploads
  • Uploaded logos are resized if they’re too large
  • Some of the filtering select boxes are merged into a single one, with highlights and indenting to easily find what you’re looking for

It’s all committed to my darcs repository, so feel free to test before I deploy in a couple of days.

After promising to open up the source for ClockingIT for quite a while, I finally found the time and energy to do it this weekend. There is now a public Darcs repository available for all interested parties. Have a look at the wiki if you’re interested.

It’s not hard to get it up and running, but I’m looking to automate the installation process a bit before creating an archive and offering up a general download.

Feel free to attempt to install it on your own servers, and let me know if you run into any problems. Patches and contributions are most welcome, too – I really want ClockingIT to keep improving all the time.

Reports rewrite

April 3, 2007

The reports rewrite reached a usable state this weekend, and I deployed the first version of the new reports system yesterday. It’s turned out to be quite flexible, and also sports CSV downloads for those Excel moments.


February 3, 2007

.. has gotten a makover as well. After meaning to clean it up for quite a while, I finally had both the motivation and energy to get it done. This will also go live on the next deployment, which will be soon unless some new problems surface during further testing.

Next up will be some kind of schedule/calendar, showing task and milestone due dates, and probably some simple event registration. We’ll see what happens once I get started.

Development update..

February 3, 2007

I’ve added support for assigning multiple users to a task, which means that the task will show for everyone assigned when viewing their tasks. This is already being used on my install at work, and I expect the change to go live after a couple of more days of testing.

There will also be a new report, Users by Date, when I deploy the next time. This report lets you easily see how much time people have logged for a date range, and you can easily see who needs more to do, or who is overworked.

Juggernaut & Comet fixes

January 25, 2007

After struggling a bit with getting the Flash bridge working on my Linux machines, I did some research and discovered that the latest Flash refuses to fetch the crossdomain.xml file from anywhere else than the xmlsocket server when operating on port 443. So, I hacked the push_server a bit, and added something like this to get it to work:

if @line_buffer.include? "<policy-file-request/>"
Logger.debug "Sending policy file.."
self.send_data "<?xml version=\"1.0\"?>\n<cross-domain-policy>\n<allow-access-from domain=\"*.clockingit.com\" to-ports=\"*\"/>\n<allow-access-from domain=\"192.168.*\" to-ports=\"*\"/>\n</cross-domain-policy>\n"

I also fetched the latest Juggernaut version from SVN and cleaned up the push_server quite a bit. There is for instance a lacking rescue on the JSON parse statement, meaning if someone sends bogus data on port 443 (ie. attempts a ssl request) it crashes.

To round of the evening, I created a first version of a print stylesheet, which makes it possible to print your task list and have it look nice and usable.

New version live..

January 10, 2007

The lastest and greatest version of ClockingIT is now live, after quite a bit of reworking and rewriting. I’ve had a lot of focus on making it easier to find the tasks you’re supposed to work on across all projects, as well as making a huge number of tasks workable.

Personally, I think it’s worked out great, and we’ve been able to handle over 800 tasks and issues across 40+ people at work with ease. Everyone knows what to do next, and there’s barely any need for status meetings any longer as everyone is up to speed on what’s been done and what’s up next.

Feel free to give it a try – ClockingIT is free to use, no ads, and no restrictions.

The new look is complete…

December 22, 2006

The wife had a new look for ClockingIT waiting for me when I came home from work yesterday, and I was able to complete it today. Personally, I think it looks fabulous. Expect the next version to go live before new years.

New Overview

Currently, I’m trying to write a tutorial and a wee bit of documentation, as some things have changed quite a bit (for the better, trust me) and I want the transition to the new way of working to be as smooth as possible for people.  I’ve even installed a wiki and going down that route instead of trying to create some static pages. Hopefully some people will join in and help me keep the documentation current. One can hope, at least.

Ferret and acts_as_ferret

December 20, 2006

I started implementing search in ClockingIT yesterday, and was prepared for a long session and massive codeage to get it done. Enter Ferret and acts_as_ferret, and the whole thing including the results page with highlighting of the search terms was done in a bit over two hours. Consider me amazed.

And when I got home from work today, the Wife had created a new look for the site as well. Even cleaner and slicker than it was. Things are looking great for the launch of the next version sometime before new-years. Just half a permissions system left to implement, and a bit of spit-shine and we’re good to go.