Problem 112: Bouncy numbers

(see projecteuler.net/problem=112)

Working from left-to-right if no digit is exceeded by the digit to its left it is called an increasing number; for example, 134468.
Similarly if no digit is exceeded by the digit to its right it is called a decreasing number; for example, 66420.
We shall call a positive integer that is neither increasing nor decreasing a "bouncy" number; for example, 155349.
Clearly there cannot be any bouncy numbers below one-hundred, but just over half of the numbers below one-thousand (525) are bouncy.
In fact, the least number for which the proportion of bouncy numbers first reaches 50% is 538.

Surprisingly, bouncy numbers become more and more common and by the time we reach 21780 the proportion of bouncy numbers is equal to 90%.

Find the least number for which the proportion of bouncy numbers is exactly 99%.

Algorithm

isBouncy determines whether its parameter is bouncy or not by stepping through its digits.
The rest is just a brute-force search.

Modifications by HackerRank

Any percentage can be entered.
Even though my brute-force search finds the solution to the original problem (99%) in less than 0.1 seconds, it's way too slow for
the potentially huge search space of Hackerrank. Someone indicated that one "answer requires about 90 bits".

My code

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

#include <iostream>
 
// return true if x is a bouncy number
bool isBouncy(unsigned long long x)
{
// figure out whether x is monotonic ascending or descending
// it's bouncy if neither ascending nor descending
bool ascending = true;
bool descending = true;
 
// initial digit (the right-most digit)
auto previous = x % 10;
x /= 10;
 
// still digits left ?
while (x > 0)
{
// current digit
auto current = x % 10;
 
// compare two digits
descending &= previous >= current;
ascending &= previous <= current;
 
// bouncy ?
if (!ascending && !descending)
return true;
 
// keep going ...
x /= 10;
previous = current;
}
 
// not bouncy (either ascending, descending or all digits are equal)
return false;
}
 
 
int main()
{
unsigned int tests = 1;
std::cin >> tests;
 
while (tests--)
{
// original problem: 99%
unsigned long long p = 99;
unsigned long long q = 100;
std::cin >> p >> q;
 
// brute-force ...
unsigned long long current = 100; // no bouncy numbers below 100
unsigned long long numBouncy = 0;
do
{
// check next number if bouncy
current++;
if (isBouncy(current))
numBouncy++;
} while (numBouncy * q < current * p); // same as numBouncy/current == p/q (=99%)
 
// print result
std::cout << current << std::endl;
}
 
return 0;
}

This solution contains 13 empty lines, 14 comments and 1 preprocessor command.

Interactive test

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

Number of test cases (1-5):

Input data (separated by spaces or newlines):
Note: Enter p and q such that p/q is your percentage, e.g. 90 100 => 90%

This is equivalent to
echo "1 90 100" | ./112

Output:

(please click 'Go !')

Note: the original problem's input 99 100 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)

Benchmark

The correct solution to the original Project Euler problem was found in 0.02 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

May 11, 2017 submitted solution
May 11, 2017 added comments

Hackerrank

see https://www.hackerrank.com/contests/projecteuler/challenges/euler112

My code solved 1 out of 16 test cases (score: 0%)

I failed 0 test cases due to wrong answers and 15 because of timeouts

Difficulty

Project Euler ranks this problem at 15% (out of 100%).

Hackerrank describes this problem as advanced.

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=112 - 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-112-density-bouncy-numbers/ (written by Kristian Edlund)
Java: github.com/nayuki/Project-Euler-solutions/blob/master/java/p112.java (written by Nayuki)
Scala: github.com/samskivert/euler-scala/blob/master/Euler112.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 133 solved problems had an average difficulty of 16.9% at Project Euler and I scored 11,174 points (out of 12300) 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 !