In the problem description it is said, fewer points are better, and this is also displayed in the ranklist for this problem. However, in the ranklist of the programming league it is reversed.
The bugs concerning the contest ranklists will be fixed in a few hours.
Yes, it is possible (the score may never exceed 1.5 points though).
There has been a stupid misunderstanding concerning Python (I was under the impression it was compiled to bytecode; all other available languages are compiled). Compilation will be introduced in the evening and solutions will be rejudged. We cannot award points for code exploiting a weakness of the judge system.
OK, but will it be possible to enter with more than one language ? Currently, only the best score is kept, which means only one language.
Well, I have to agree that my last solution may not be "in the spirit" of the problem but hey ! the rules are the same for everybody and everybody can exploit them What's the problem with that ? (I would be happy to hear other people's comments on that) At least, you will have to admit that my solution has a really nice tree shape...
We should have done it sooner.. (however, speed gain seems insignificant).
As for the fairness.. Your program certainly solved the task But, as Adrian said, it was a mistake: he thought we already had compiled Python programs to bytecode first. For example, Perl is still interpreted directly from source and it is banned in this problem because of this.
Yes, higher-ranked solutions to CTQUINE look great
But I can't resist the pleasure of publishing my solution now It scores 22 points, with 20 lines and two penalties of 1. Anyone able to beat it ? 8) (copy and paste, as the font width is not fixed here)
Sure. See the ranklists now (each plus stands for +0.5 point). So at present you have the maximum possible score (1.5 points).
(There is still a minor feature: the ranks for the problem here are generated per user, per language, and therefore your relative position is calculated differently than in the official standings here. I think such behaviour is fairly harmless and there is no point in changing it).
If you would like to benchmark your solution to CTQUINE on a piece of code similar to the special judge used for this problem, you can take a look here: http://sphere.pl/~kosowski/SPOJ/CTQUINE/. The score is calculated in the source file quine.cpp.
I wonder why I get WA with my Java program. I tested it in Linux and Windows environment, and in both cases it exactly prints its sourcecode. Maybe it is because I don't terminate the sourcecode with a newline, and spoj adds it? But my C program that also doesn't end with a newline got Accepted
I had some spare time at the weekend and I've submitted a solution to the problem (id:50900). It is obviously outside the ranking, but I would be happy to see it beaten, anyway.