Mircea Bardac /random_thoughts

I have finished today an initial version for a Command Line Argument Parser for Psi.
It adds the following possible arguments for Psi.

• --datadir=PATH - overrides PSIDATADIR
• --profile=ProfileName - specifies the profile to be automatically selected from the profiles available in the data dir (not necessarily defined through –datadir=…)
• --profiledir=PATH - this option confilcts with the first option and overrides the 2nd; it describes the directory of the only Profile going to be loaded; switching profiles is forbidden

The main reason for starting this implementation is to make Psi as USB-Stick friendly as possible. As Psi Windows Package maintainer I plan adding some customizations to the installer to allow installing to a USB stick, with full functionality.

The patch is submitted to psi-devel-maillist for reviewing.

Source: Trolltech: Trolltech to Extend Dual Licensing to Qt for Windows (February 7, 2005)

Trolltech®, a provider of leading application development software, today announced that it will extend its successful dual-licensing business model to include the Qt® cross-platform application development framework for the Microsoft Windows operating system. The availability of Qt for Windows under the open source GNU General Public License (GPL) will bring the benefits of open source software to the Windows environment, fueling the development of open source applications for the mainstream PC market. Qt for Windows is already widely used for commercial software development.

Great news for Open Source developers.
Another good reason for me to start learning & using it, even though I’m on Linux. “Hacking” Psi would be a good start :).

Qt is a complete C++ application development framework. It includes a class library and tools for cross-platform development and internationalization.

It provides classes for various GUI components, offering fast way of developing cross platform GUI applications.

Have you ever had a normal (not a project) programming homework of ~800 lines?!? I hadn’t, until now…
I spent tons of hours making it because:
- I had to learn some things which were not taught at the university before the homework
- it had many tasks to complete
- I had to make it readable (= add comments, make README, add tests) - because this counts!
I had it finished at 4:30 a.m local time. I went to bed and this evening I’ve written comments, README, packed it and sent it.

And now… you might ask yourselves what for?
The answer is simple: out of 10 (the biggest grade possible at the end of the semester), all the homework count for 2.4 (there are 12 homework). That makes 0.2 for each homework.

That means that I have written 800 lines for only 2% of the final grade!
(some of the 0.2 points could have been recovered from bonuses coming from other homework)

I wish I won’t have to do such homework ever again! (unless it counts MORE)

My quest for STL knowledge lead me to an extremely resourcefull site, actually a site for the Data Structures course of the Center for Conected Learning and Computer-Based Modelling in Northwestern University.

Resouces section in Data Structures Course:

Some of the following links date back pretty far and may be a bit
crufty, have broken links, etc.

• How to: Yell for Help, Code Good, Test Good
• About Algorithms and Data Structures
  • Chapter 8: Simple sorting algorithms
  • Chapter 9: Linked lists (commented in the original source)
  • Chapter 10: Budd’s RPN calculator
  • Chapter 10: Budd’s bank simulator
  • Chapter 11: A better backtracker
  • Chapter 12: Sets (commented in the original source)
  • Chapter 13: Trees
  • Red-black trees
  • Chapter 17: Hashtables
  • Chapter 19: Graphs

  Chapter references are to

• About the STL
  • Quick summary
  • STL resources (tutorials, implementations, references)
  • Notes on specific topics:
    • Templates and reference variables in C++
    • Iterators
    • Generic Algorithms
    • Passing functions to STL algorithms
    • Set algorithms
    • STL-think: how to program at a higher-level in C++ using the STL
    • How to define iterators

I was actually looking for methods of defining iterators but it seems I discovered a gold mine
Don’t forget to check out the STL resources link there.

The 3rd Data Structures and Alogorithms homework was a real challenge. It didn’t only require a good programming, but also some algorithms knowledge. We have received 2 versions of the same homework last week: one easy homework and a difficult one. The difficult one comes also with a bonus.

When you see the possibility to achieve more than the usual through programming, you can be sure you’ll find me there. I’ve chose the difficult version. The problem consisted of determining the convex hull for a number of points on a 2D surface. Well, if I think that I’ve done it before, but with another algorithm, I can’t really say it’s difficult. The big challenge was actually the STL implementation which should have been made both memory and time efficient. It worked well after patching it several times because of wrong behaviour in special cases.