Why do I publish my solutions ?
Almost always the solutions for problems at Project Euler consist of two parts:
- finding a mathematical way to break down the problem's structure into its elements
- writing an efficient program for step 1
There are already several blogs or code repositories (with the vast majority being hosted at GitHub) where you can find solutions to the Project Euler problems.
I strive to:
- explain my choice of algorithms and data structures
- write fully commented C++ source code that compiles without warnings and without any external libraries
- offer interactive tests for most problems
- link to relevant Wikipedia / MathWorld / OEIS pages
- link to other solutions, especially those that are written in other programming languages
Let's not forget that this website helps me, too: only if I can explain a solution to someone else then I can be sure that I truly understood it in the first place.
And practicing some of the lesser used features of C++ (such as algorithm hidden inside STL, like std::next_permutation
) improves my overall coding skills as well.
Sounds like a win-win situation ... ☺
"But You Shouldn't Publish Your Solutions !"
Every time you enter the correct result on the Project Euler website, you see the following text:
Please do not deprive others of going through the same process by publishing your solution outside Project Euler.
If you want to share your insights then please go to thread [xyz] in the discussion forum.
I have a different point of view:
- 99% of my knowledge is based on things I was taught, I saw somewhere or I stumbled across
- and maybe 1% is "original"
- I'm pretty sure it's not just me - it's the way how all of us gain knowledge
- having a good teacher
- access to well-equipped library
- and probably most important today: your skills in working with a search engine
That's the only way how knowledge can be spread - all famous scientists wrote books.
Leonhard Euler was one of the most productive mathematicians and he published 866 papers/books/etc. He shared his knowledge.
And a substantial number of my solutions is based on some of his formulas, I only solved them because I could look up his works.
Admittedly, there is no use in publishing lists of the results to Project Euler problems. In my opinion, these numbers don't matter at all:
noone really cares whether the result of problem 1 is 233167, 233168 or 233169.
That's why you find algorithms, explanations, links, code, ..., basically everything on my website - but not the results.
Because these numbers don't teach you anything.
"The Solutions On This Website Are Too Complicated ..."
"... someone else wrote much shorter/simpler programs !".
Yes, several solutions to the original Project Euler problems are indeed much more straightforward.
But they usually:
- don't scale well (see notes about Hackerrank)
- have about the same number of lines of code, but no comments
- use language-specific features, such as Python's BigNum support
HackerRank
The commercial website HackerRank offers an online programming environment. They have a huge variety of coding exercises and competitions for beginners and "masters", too.
Some Project Euler problems are modified by HackerRank and in most cases they become (much) harder.
There is a strict time limit - often 2 seconds for C++ code - whereas Project Euler just recommends that your program should finish in less than 1 minute.
The majority of my programs have a perfect 100% score at HackerRank, unless indicated otherwise.
The input format of my interactive tests is usually identical to HackerRank's.
If HackerRank's modifications substantially change my algorithm then I need to insert #ifdef ORIGINAL
,
whereas ORIGINAL
refers to the original Project Euler problem and #ifndef ORIGINAL
(or just #else
)
contains HackerRank specific code.
Note: the test cases at Hackerrank "discovered" a few bugs in my code. Even though I had the correct result for Project Euler,
some of my programs failed to pass all Hackerrank tests.
@Hackerrank: Thanks for including those nasty edge cases most of us forget about !
When do you solve problem XYZ ?
I solve a problem whenever I have the time and feel the "thirst" ...
There is no certain order - even though I solved most of the first 100 problems almost in ascending order.
Sometimes I can apply knowledge gained from one problem to solve a harder one. And sometimes I randomly select a problem.
The difficulty is rising faster than the percentage indicates: "70-percentager" among the first 300 problems are hard but doable whereas
a substantial number of "50-percentager" of the 300+ problems is completely unsolveable to me because I have no idea even what theroems/conjectures/etc. to look for.
Interactive Tests
I created this feature because I often had trouble in finding some hidden bugs in my code:
I had the right algorithm, hopefully the best data structure and somehow my results were incorrect.
If you use my interactive tests then you can hopefully find your bugs, too.
Please note that I set a execution timeout of 1 second on my server. Memory is typically limited to 64 MByte.
If too many users are running tests simultaneously, then some processes might get killed in order to keep the server responsive.
In addition, some input/output values are forbidden - especially the default input from the original Project Euler problem.
My logfiles show that about 50% of all users still attempt to enter those inputs in order to see the solution.
If you are that lazy and stupid then look somewhere else.
Note: this feature is still under development and requires Javascript to be enabled.
Heatmap
Please click on a problem's number to open my solution to that problem:
green | solutions solve the original Project Euler problem and have a perfect score of 100% at Hackerrank, too | |
yellow | solutions score less than 100% at Hackerrank (but still solve the original problem easily) | |
gray | problems are already solved but I haven't published my solution yet | |
blue | solutions are relevant for Project Euler only: there wasn't a Hackerrank version of it (at the time I solved it) or it differed too much | |
orange | problems are solved but exceed the time limit of one minute or the memory limit of 256 MByte | |
red | problems are not solved yet but I wrote a simulation to approximate the result or verified at least the given example - usually I sketched a few ideas, too | |
black | problems are solved but access to the solution is blocked for a few days until the next problem is published | |
[new] | the flashing problem is the one I solved most recently |
I stopped working on Project Euler problems around the time they released 617.
1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | 14 | 15 | 16 | 17 | 18 | 19 | 20 | 21 | 22 | 23 | 24 | 25 |
26 | 27 | 28 | 29 | 30 | 31 | 32 | 33 | 34 | 35 | 36 | 37 | 38 | 39 | 40 | 41 | 42 | 43 | 44 | 45 | 46 | 47 | 48 | 49 | 50 |
51 | 52 | 53 | 54 | 55 | 56 | 57 | 58 | 59 | 60 | 61 | 62 | 63 | 64 | 65 | 66 | 67 | 68 | 69 | 70 | 71 | 72 | 73 | 74 | 75 |
76 | 77 | 78 | 79 | 80 | 81 | 82 | 83 | 84 | 85 | 86 | 87 | 88 | 89 | 90 | 91 | 92 | 93 | 94 | 95 | 96 | 97 | 98 | 99 | 100 |
101 | 102 | 103 | 104 | 105 | 106 | 107 | 108 | 109 | 110 | 111 | 112 | 113 | 114 | 115 | 116 | 117 | 118 | 119 | 120 | 121 | 122 | 123 | 124 | 125 |
126 | 127 | 128 | 129 | 130 | 131 | 132 | 133 | 134 | 135 | 136 | 137 | 138 | 139 | 140 | 141 | 142 | 143 | 144 | 145 | 146 | 147 | 148 | 149 | 150 |
151 | 152 | 153 | 154 | 155 | 156 | 157 | 158 | 159 | 160 | 161 | 162 | 163 | 164 | 165 | 166 | 167 | 168 | 169 | 170 | 171 | 172 | 173 | 174 | 175 |
176 | 177 | 178 | 179 | 180 | 181 | 182 | 183 | 184 | 185 | 186 | 187 | 188 | 189 | 190 | 191 | 192 | 193 | 194 | 195 | 196 |