Problem 55: Lychrel numbers

(see projecteuler.net/problem=55)

If we take 47, reverse and add, 47 + 74 = 121, which is palindromic.

Not all numbers produce palindromes so quickly. For example,
349 + 943 = 1292,
1292 + 2921 = 4213
4213 + 3124 = 7337
That is, 349 took three iterations to arrive at a palindrome.

Although no one has proved it yet, it is thought that some numbers, like 196, never produce a palindrome.
A number that never forms a palindrome through the reverse and add process is called a Lychrel number.
Due to the theoretical nature of these numbers, and for the purpose of this problem, we shall assume that a number is Lychrel until proven otherwise.
In addition you are given that for every number below ten-thousand, it will either
(i) become a palindrome in less than fifty iterations, or,
(ii) no one, with all the computing power that exists, has managed so far to map it to a palindrome.
In fact, 10677 is the first number to be shown to require over fifty iterations before producing a palindrome: 4668731596684224866951378664 (53 iterations, 28-digits).

Surprisingly, there are palindromic numbers that are themselves Lychrel numbers; the first example is 4994.

How many Lychrel numbers are there below ten-thousand?

Algorithm

Intermediate numbers may become quite huge (as hinted in the problem), therefore I decided to store all numbers in a BigNumber,
which is a std::vector where each digit is stored as a single element (lowest digits first), e.g. 3401 = { 1, 0, 4, 3 }.

Then, reversing the digits becomes super-easy:
auto reverse = number;
std::reverse(reverse.begin(), reverse.end());

And checking whether we have a palindrome
if (number == reverse)

Adding the digits is based on the basic algorithm taught in school.

My code

… was written in C++11 and can be compiled with G++, Clang++, Visual C++. You can download it, too.

The code contains #ifdefs to switch between the original problem and the Hackerrank version.
Enable #ifdef ORIGINAL to produce the result for the original problem (default setting for most problems).

#include <vector>
#include <map>
#include <algorithm>
#include <iostream>
 
//#define ORIGINAL
 
// a sequence of digits, lowest digits first, e.g. 3401 = { 1, 0, 4, 3 }
typedef std::vector<unsigned int> BigNumber;
 
// count all numbers converging to BigNumber
std::map<BigNumber, unsigned int> finalNumber;
 
// return true if x is a Lychrel number, stop after maxIterations (and return true if "indecisive")
bool isLychrel(unsigned int x, unsigned int maxIterations)
{
// split integer into its digit, store each digit in a separate cell
BigNumber number;
while (x > 0)
{
number.push_back(x % 10);
x /= 10;
}
 
// try to find a palindrome in the first 60 iterations
for (unsigned int iteration = 0; iteration < maxIterations; iteration++)
{
auto reverse = number;
std::reverse(reverse.begin(), reverse.end());
 
// check if palindrome
#ifdef ORIGINAL
if (iteration > 0) // originally, the initial number is allowed to be a palindrome
#endif
// no, can't be a Lychrel number
if (number == reverse)
{
finalNumber[number]++;
return false;
}
 
// add the reverse
auto sum = number;
unsigned int carry = 0;
for (size_t digit = 0; digit < number.size(); digit++)
{
// get digit "from the other end"
sum[digit] += reverse[digit] + carry;
 
// overflow ?
if (sum[digit] >= 10)
{
sum[digit] -= 10;
carry = 1;
}
else
{
// add one more digit
carry = 0;
}
}
if (carry > 0)
sum.push_back(carry);
 
number = std::move(sum);
}
 
// yes, we have a Lychrel number
return true;
}
 
int main()
{
// consider a number to be a Lychrel number if no palindrome after that many iterations
#ifdef ORIGINAL
unsigned int iterations = 50;
#else
unsigned int iterations = 60;
#endif
 
unsigned int maxNumber = 10000;
std::cin >> maxNumber;
 
// count all Lychrel number
unsigned int count = 0;
for (unsigned int i = 0; i <= maxNumber; i++)
if (isLychrel(i, iterations))
count++;
 
#ifdef ORIGINAL
std::cout << count << std::endl;
 
#else
 
unsigned int bestCount = 0;
BigNumber bestNumber;
// find number most converged to
for (auto f : finalNumber)
if (bestCount < f.second)
{
bestCount = f.second;
bestNumber = f.first;
}
 
// print single digits, highest digits first (they were stored last in BigNumber)
std::reverse(bestNumber.begin(), bestNumber.end());
for (auto digit : bestNumber)
std::cout << digit;
std::cout << " " << bestCount << std::endl;
#endif
 
return 0;
}

This solution contains 18 empty lines, 17 comments and 12 preprocessor commands.

Interactive test

You can submit your own input to my program and it will be instantly processed at my server:

This live test is based on the Hackerrank problem.

Input data (separated by spaces or newlines):

This is equivalent to
echo 130 | ./55

Output:

(please click 'Go !')

(this interactive test is still under development, computations will be aborted after one second)

Benchmark

The correct solution to the original Project Euler problem was found in 0.01 seconds on a Intel® Core™ i7-2600K CPU @ 3.40GHz.
(compiled for x86_64 / Linux, GCC flags: -O3 -march=native -fno-exceptions -fno-rtti -std=c++11 -DORIGINAL)

See here for a comparison of all solutions.

Note: interactive tests run on a weaker (=slower) computer. Some interactive tests are compiled without -DORIGINAL.

Changelog

February 27, 2017 submitted solution
April 21, 2017 added comments

Hackerrank

see https://www.hackerrank.com/contests/projecteuler/challenges/euler055

My code solves 6 out of 6 test cases (score: 100%)

Difficulty

Project Euler ranks this problem at 5% (out of 100%).

Hackerrank describes this problem as easy.

Note:
Hackerrank has strict execution time limits (typically 2 seconds for C++ code) and often a much wider input range than the original problem.
In my opinion, Hackerrank's modified problems are usually a lot harder to solve. As a rule thumb: brute-force is never an option.

Links

projecteuler.net/thread=55 - the best forum on the subject (note: you have to submit the correct solution first)

Code in various languages:

Python: www.mathblog.dk/project-euler-55-how-many-lychrel-numbers-are-there-below-ten-thousand/ (written by Kristian Edlund)
Haskell: github.com/nayuki/Project-Euler-solutions/blob/master/haskell/p055.hs (written by Nayuki)
Java: github.com/nayuki/Project-Euler-solutions/blob/master/java/p055.java (written by Nayuki)
Go: github.com/frrad/project-euler/blob/master/golang/Problem055.go (written by Frederick Robinson)
Javascript: github.com/dsernst/ProjectEuler/blob/master/55 Lychrel numbers.js (written by David Ernst)
Scala: github.com/samskivert/euler-scala/blob/master/Euler055.scala (written by Michael Bayne)

Heatmap

green problems solve the original Project Euler problem and have a perfect score of 100% at Hackerrank, too.
yellow problems score less than 100% at Hackerrank (but still solve the original problem).
gray problems are already solved but I haven't published my solution yet.
blue problems are solved and there wasn't a Hackerrank version of it at the time I solved it or I didn't care about it because it differed too much.

Please click on a problem's number to open my solution to that problem:

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
176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200
201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225
226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250
The 163 solved problems had an average difficulty of 22.2% at Project Euler and I scored 11,907 points (out of 13200) at Hackerrank's Project Euler+.
My username at Project Euler is stephanbrumme while it's stbrumme at Hackerrank.
more about me can be found on my homepage, especially in my coding blog.
some names mentioned on this site may be trademarks of their respective owners.
thanks to the KaTeX team for their great typesetting library !