Overview
Atlassian SourceTree is a free Git and Mercurial client for Windows.
Atlassian SourceTree is a free Git and Mercurial client for Mac.
Project Euler Solutions
The Project Euler is a website hosting a series of computational problems. This repository also mirrors all the problem statements (with some delay).
Statement fetcher
Use fetch.sh
to fetch all the problem statements locally.
See fetch.sh h
for more details.
Objectives
The main objective is to be smart and to grasp a deep understanding of every problem.

Write the fastest possible code under a limit condition. The 'limit' variable is to make clear that nothing is precomputed.

The only output should be the result.

Some problem solutions are not an algorithm but merely a mathematical derivation (also known as a "pen and paper" solution).

Code should be clear and short.

Focus on the algorithm, avoid language implementationspecific optmization. Hardwarespecific optimizations are usually discarded, although some of them, which are the essence of computer science, can be used (e.g. binary operations).

Every implementation should run in less than a minute on any decent computer as suggested on the home page. Strive to keep runningtime below a second, or even 10 ms on a modern computer. Compiler optimization is disabled.

Explain the main algorithm in head comment if needed. Point out the tricks and pitfalls, explain why some approaches would be slow.

If possible or reasonable, write portable code, use the standard of the language only, without external library nor external file. Every problem should be straighforward to compile and execute.

There might exist various solutions where complexity is unclear or where performance depends on the input. There are usually tradeoffs in memory/speed or speed/code length. These implementations are kept together for comparison.

Code has to be proven and should not be working just because of some lucky shot. Indeed, a wrong algorithms can output the right result.

Some problems evolve around big numbers, thus assisting the resolution with a big number library makes the resolution trivial and pointless. The goal is to either avoid going through big numbers manipulation, or to find an efficient way to store them.
Note
A lot of solutions available on the forum and out there on the Internet are plain brute force: I think the authors missed the point. Many problems suggested on Project Euler are rather simple to implement using a naive approach. The real purpose of the project is to make you think of a fast and elegant resolution.
Contribution
Feedback on solved problems is very welcome. Please do not send me anything on problems yet to be solved.