Met IndLinux guys !!!

It was a great day !!! Today I met two great guys from IndLinux. One Mr. Gaurav and another person Mr. RamaKrishna. Demonstrated LDTP project to them...

Also got few queries from them: How do you verify whether the (localized) font is rendered properly or not ? I replied like; by takin screen shot, we achieve the result. But this may not yield a right result always :) I have also told about Xvfb + LDTP. If somebody gets some alternative, please let me know :)

Uploaded Test cases to go-evolution.org

Today uploaded more than 2000 test-cases of Evolution into go-evolution.org wiki pages. Me and my team members (Poornima, Subodh, Shaheed, Shilpa) are really happy to give back our work to the community. Soon, I will try to put them in some test case repository.

LDTP hosted in freedesktop.org !!!

On September 19, 2005: LDTP has been hosted in http://freedesktop.org. Thanks to Eric Anholt who helps me to host LDTP in freedesktop.org :) I feel this as one achievement in LDTP development.

Students Crack a Hard Nut, Turn Plastic into Fuel

A team of students at Velammal Engineering College has discovered a method of producing quality petrol from one of the most problematic pollutants of modern times - non-biodegradable plastic. A final year project led mechanical engineering students Sridhar, Jaikar Sathish, V. Lakshmanan and Guru Prasad through several experiments with a `secret catalyst' that prevented plastic from melting into a mass when heated, and instead generated petroleum products - petrol, diesel and kerosene. Waste polythene and polypropylene - consisting mainly of discarded bags and biomedical waste - are subjected to `catalytic cracking' or breaking down the carbon chain. The correct ratio of the catalyst and the plastic materials are taken in the reaction flask to get greater yield. When heated at 400 degrees Celsius, the plastic yields a distillate crude. Fractional re-distillation yields a type of petrol at between 100 to 120 degrees C and kerosene at between 150 to 180 degree C. Finally, the process leaves diesel as residue. In about two hours, the students were able to generate around 2 litres of crude from 2.5 kg of plastic waste and through distillation one litre of petrol and half a litre each of kerosene and diesel. The cost works out to roughly Rs. 22 for petrol and Rs. 26 each for diesel and kerosene. Sridhar says by heating plastic in the absence of oxygen, toxic dioxin emissions are also avoided, making the fuel eco-friendly. Even the left over paraffin mass (approximately 500 gm) can be made into candles. The fuel, developed by the team, has been certified by the Indian Oil Corporation's Regional Laboratory at Korukkupet. Compared to regular petrol, this plastic-derived fuel has a marginally higher octane level; pollution emission was within stipulated limits. Sridhar says the fuel's mileage could be higher than the 40 km per litre he got on his 20-year-old scooter. The students have already obtained a patent for the process and have applied for a full patent. The Indian Institute of Technology (IIT) has agreed to quality test the products. The students were guided in their project by the College Principal C. Rathnasabapathy, and professors, D. Sivaraj and K. R. Senthil Kumar.

Orignial source: The Hindu, March 19, 2005
Online source: Link1, Link2

Copied code from at-spi examples and started recording !!!

10 days back, I have taken at-spi example code and started implementing new LDTP recording code.

Base at-spi code is event-listener-test.c. Soon this work will be published. Today I discussed this code and its implementation details with Shaheed, Prem and Subodh. They have assured that they will also contribute to an extent :D Let us see how this project grows. The reason for this new recording implementation is, with the existing recording code, we have lot of implementation issues and we were also not able to extend. With the new one, hope we won't face any such issues. But based on my current code, I saw few race conditons and I'm planing to use shared memory concept. I'm not sure how effective its going to be. Need to implement and see.

Discussed LDTP localization implementation with Poornima

Today I had a good discussion with Poornima about LDTP localization implementation. I hope this logic will work out. I can see only GNOME based applications in the locale directory. Need to check for OpenOffice.org and Mozilla based localization directory and also need to find some better solution which can be used for all common platforms, like KDE, Mozilla, OpenOffice.org.

Planing to use textdomain and bindtextdomain functions to register the locale directory and then planing to use either _() or _Q() glib function.

Shaheed's new LDTP editor

Shaheed is doing LDTP editor which looks similar to gedit. But its completely written in PyGTK. We should check with gedit author for any licensing issue.

We started this editor for debugging the scripts and a python editor with GTK interface. But things are getting in a different shape. Currently we are using it for generating two different XML files. One is data XML file (which is still under development) and the other XML file is to group the list of test cases to be executed in some order. Here grouping works in such a way that, if one test case in that group fails, then the rest of the scripts in that group will be ignored. Instead the next group execution will be started (if next group is available). Hope soon this will be published :)

Localization testing using LDTP

During weekend got new idea about localization testing using LDTP !

From LDTP we can export a function which will take the localizied .mo directory path. In pyldtp, we can just get the label stored in appmap (which should be in English) and convert the label / label_by into the localized string at run-time, using _(String) of glib function. This is still under evaluation. I hope things should work fine :D

Nice week end !!!

I don't know something happened to me, last week end. I was just writing some LDTP code, restructuring few things. Wrote some new code too ;)

Implemented 'GNU Linux Desktop Automation Project' tool. This gets a XML file as an input, which will contain the list of scripts to be executed in a sequence. Moreover group can also be specified. With grouping, if one script fails, maybe we can ignore the whole set of scripts in that group and start executing the next group.

Met with an accident :(

Unfortunately when traveling in my bike (60 KM/H) in Bangalore, I met with an accident on 27-August-2005. Even though its not a major accident, I got hurted badly.

These days coming to office by bus only :) Getting recovered slowly from injuries.