Problem 119: Digit power sum

(see projecteuler.net/problem=119)

The number 512 is interesting because it is equal to the sum of its digits raised to some power: 5 + 1 + 2 = 8, and 8^3 = 512.
Another example of a number with this property is 614656 = 28^4.

We shall define a_n to be the nth term of this sequence and insist that a number must contain at least two digits to have a sum.

You are given that a_2 = 512 and a_10 = 614656.

Find a_30.

Algorithm

A simple but extremely slow brute-force program revealed that a_30 has more than 10 digits:
I was iterating through numbers from 10 to 10,000,000,000 and didn't find enough terms a_i:
81, 512, 2401, 4913, 5832, 17576, 19683, 234256, 390625, 614656, 1679616, 17210368, 34012224, 52521875, 60466176, 205962976, 612220032, 8303765625, ...

Then I turned the whole problem upside-down: there are 90 two-digit numbers but only 18 different digit sums.
Even better: there are 9000 four-digit numbers but only 36 different digit sums !
Actually each digit is always between 0 and 9 therefore the maximum digit sum is \sum_i{digit_i} = 9i.

My program looks at all potential digit sums from 1 to 9*20 (I thought that maybe a_30 < 10^20, and that's actually true).
For each potential digit sum, I call them base in my code, all exponents are computed until base^{exponent} overflows.
If digitSum(base^{exponent}) = base then a solution has been found.

Solutions will be found in a "random" order, so I add them to a sorted std::set and read its 30th element.

Modifications by HackerRank

The search space is much bigger: all numbers up to 10^100 (that's a Googol) have to be printed.
Even worse, numbers can be encoded in any base from 2 to 1000.

My BigNum class was already able to store its digits in arbirary bases - but I always had powers of 10 in my mind.
Addition and multiplication is reduced to in-place operations to avoid slow memory allocations.
Pretty simple but effective is BigNum::convert which converts a BigNum to a different base.

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 <iostream>
#include <set>
 
//#define ORIGINAL
#ifdef ORIGINAL
 
// add all digits (base 10)
unsigned int digitSum(unsigned long long x)
{
unsigned int result = 0;
while (x > 0)
{
result += x % 10;
x /= 10;
}
return result;
}
 
int main()
{
std::set<unsigned long long> solutions;
 
// based on my experiments with BigNum I knew that the result fits in 64 bit
// which is about 10^20, each digit <= 9, therefore their sum <= 20*9
unsigned int MaxBase = 20 * 9;
// base is the sum of all digits
for (unsigned int base = 2; base <= MaxBase; base++)
{
unsigned int exponent = 1;
// step-by-step: base^exponent, then base^(exponent+1), ...
unsigned long long current = base;
 
while (current < ULLONG_MAX / base) // 1^64 - 1
{
auto sum = digitSum(current);
// digit sum equals base ? (note: single-digit numbers excluded)
if (sum == base && current >= 10)
solutions.insert(current);
 
// next iteration
current *= base;
exponent++;
}
}
 
// sorted results
auto i = solutions.begin();
// skip 30 - 1 = 29 elements
std::advance(i, 30 - 1);
// print the 30th
std::cout << *i << std::endl;
return 0;
}
 
#else
 
#include <vector>
 
// store single digits with lowest digits first
// e.g. 1024 is stored as { 4,2,0,1 } (when base = 10)
struct BigNum : public std::vector<unsigned int>
{
unsigned int maxDigit;
 
// store a non-negative number
BigNum(unsigned long long x = 0, unsigned int base = 10)
: maxDigit(base)
{
do
{
push_back(x % maxDigit);
x /= maxDigit;
} while (x > 0);
}
 
// add two big numbers
void operator+=(const BigNum& other)
{
// add in-place, make sure it's big enough
if (size() < other.size())
resize(other.size(), 0);
 
unsigned int carry = 0;
for (unsigned int i = 0; i < size(); i++)
{
carry += operator[](i);
if (i < other.size())
carry += other[i];
else
if (carry == 0)
return;
 
if (carry < maxDigit)
{
// no overflow
operator[](i) = carry;
carry = 0;
}
else
{
// yes, we have an overflow
operator[](i) = carry - maxDigit;
carry = 1;
}
}
 
if (carry > 0)
push_back(carry);
}
 
// multiply a big number by an integer
void operator*=(unsigned int factor)
{
unsigned long long carry = 0;
for (size_t i = 0; i < size(); i++)
{
carry += operator[](i) * (unsigned long long)factor;
operator[](i) = carry % maxDigit;
carry /= maxDigit;
}
// store remaining carry in new digits
while (carry > 0)
{
push_back(carry % maxDigit);
carry /= maxDigit;
}
}
 
// compare two big numbers
bool operator<(const BigNum& other) const
{
if (size() < other.size())
return true;
if (size() > other.size())
return false;
for (int i = (int)size() - 1; i >= 0; i--)
{
if (operator[](i) < other[i])
return true;
if (operator[](i) > other[i])
return false;
}
return false;
}
 
// convert to a different radix
BigNum convert(unsigned int newRadix) const
{
BigNum result(0, newRadix);
for (auto i = rbegin(); i != rend(); i++)
{
result *= maxDigit;
result += *i;
}
return result;
}
};
 
 
// add all digits
unsigned int digitSum(const BigNum& x)
{
unsigned int result = 0;
for (auto digit : x)
result += digit;
return result;
}
 
 
int main()
{
std::set<BigNum> solutions;
unsigned int radix = 10;
std::cin >> radix;
 
// do not exceed 10^100
BigNum googol(1, 10);
for (unsigned int digits = 1; digits <= 100; digits++)
googol *= 10;
// convert to current radix
BigNum max = googol.convert(radix);
 
// analyze all base^exponent below 10^100
// base is the sum of digits
for (unsigned int base = 2; base < (radix-1)*max.size(); base++)
{
unsigned int exponent = 1;
// step-by-step: base^exponent, then base^(exponent+1), ...
BigNum current(base, radix);
// still below a googol
while (current < max)
{
auto sum = digitSum(current);
// digit sum equals base ? single-digit numbers excluded
if (sum == base && current.size() >= 2)
{
// store result in decimal notation
BigNum decimal = current.convert(10);
solutions.insert(decimal);
}
 
// next iteration
current *= base;
exponent++;
}
}
 
// sorted results
for (auto i : solutions)
{
// print digits (note: they are stored in reverse)
for (auto j = i.rbegin(); j != i.rend(); j++)
std::cout << *j;
std::cout << " ";
}
 
return 0;
}
#endif

This solution contains 27 empty lines, 34 comments and 6 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):
Note: Enter a base, e.g. 2 stands for the binary number system

This is equivalent to
echo 2 | ./119

Output:

(please click 'Go !')

Note: the original problem's input 10 cannot be entered
because just copying results is a soft skill reserved for idiots.

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

Changelog

May 13, 2017 submitted solution
May 13, 2017 added comments

Hackerrank

see https://www.hackerrank.com/contests/projecteuler/challenges/euler119

My code solved 20 out of 20 test cases (score: 100%)

Difficulty

Project Euler ranks this problem at 30% (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=119 - 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-119-sum-of-digits-raised-power/ (written by Kristian Edlund)
Java: github.com/nayuki/Project-Euler-solutions/blob/master/java/p119.java (written by Nayuki)
Scala: github.com/samskivert/euler-scala/blob/master/Euler119.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 already 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
The 126 solved problems had an average difficulty of 16.0% at Project Euler and I scored 11,074 points (out of 12500) at Hackerrank's Project Euler+.
more about me can be found on my homepage.
some names mentioned on this site may be trademarks of their respective owners.
thanks to the KaTeX team for their great typesetting library !