Reputation: 2499
Given any random integer, create a function to find the next number that is a prime number and also a palindrome.
My attempt
def golf(number):
x = number + 1
for i in range(2, x):
if x % i == 0 or str(x) != str(x)[::-1]:
golf(number + 1)
return x
E.g golf(13) = 101
I'm actually looking for an alternative option than the recursion method i used. How can this best be accomplished without using recursion?
Thanks
Upvotes: 0
Views: 315
Reputation: 9134
Palindrome are a sparser set of numbers than primes, and you can generate palindromes directly.
Consider the sequence 98.102
These are palidrome numbers you can base on these
989, 9889, 999, 9999, 10001, 100001, 10101, 101101, 10201, 102201
ADDED
Not also that all of the palidromes with an odd number of digits will come before the palidromes with an even number of digits.
If you write this as a generator (ie using yield) get get a straightforward algorithm for generating palindromic numbers in order.
For 1..9 you generate either 9 or 18 palindromes depending upon whether you consider 1 digit numbers palindromic.
For 10..99 you generate 90 even digit and 90 odd digit palindromes.
For 100..999 you generate 900 even digit and 900 odd digit palindromes.
You have just generated all 1989 (or 1997 if including single digit numbers) of the palindromic numbers less than 1 million. There are 78,498 primes less than 1 million
Any algorithm that is based on generating primes then testing for a palindrome will be much slower that generating palindromes and then testing for primes
Upvotes: 1
Reputation: 342
improved according to Blckknght's advice, thanks.
def golf(number):
x = number
while True:
x += 1
if str(x) != str(x)[::-1]:
continue
for i in xrange(2, x):
if x % i == 0 :
break
else:
return x
Upvotes: 0
Reputation: 104762
Here's a variation on byron he's answer which adds several optimizations:
x
values (other than 2) before doing any elaborate tests, since we can trivially tell they are not prime.str(x)
once, and reuse the value later.x
value.sqrt(x)
, rather than going all the way to x
itself.break
, the else
block attached to the loop will be run.The code:
import math
def next_prime_palindrome(x):
while True:
x += 1
if x > 2 and x % 2 == 0: # even numbers greater than 2 are non-prime
continue
s = str(x) # compute str(x) just once
if x > 11 and len(s) % 2 == 0: # all even-length palindromes are multiples of 11
x = 10 ** len(s) # so jump to the next odd-length integer
continue
if s != s[::-1]: # palindrome test
continue
for i in xrange(3, round(math.sqrt(x))+1, 2): # loop over odd potential divisors
if x % i == 0: # prime test
break
else: # this else block runs only if no break happened in the loop, so x is prime
return x
Here are some tests runs, showing a few cases where the optimizations save significant time:
>>> next_prime_palindrome(1)
2
>>> next_prime_palindrome(3)
5
>>> next_prime_palindrome(9)
11
>>> next_prime_palindrome(11)
101
>>> next_prime_palindrome(99999)
1003001
>>> next_prime_palindrome(999999999)
10000500001
A further improvement might be to directly generate palindromes, rather than working with integers to start with, and doing a palindrome test to filter them. That would get quite a bit further from your original design, so I'll leave that for someone else.
Upvotes: 1
Reputation: 53535
Your solution can be slightly modified in order to create an iterative solution:
def golf(number):
x = number + 1
while True:
is_golf = True
for i in range(2, x):
if x % i == 0 or str(x) != str(x)[::-1]:
is_golf = False
break
if is_golf:
return x
x += 1
Upvotes: 0
Reputation: 114025
def golf(number):
primes = []
i = 2
while i <= number:
if isPrime(i, primes):
primes.append(i)
i += 1
answer = primes[-1] + 1
while True:
if isPrime(answer, primes):
primes.append(answer)
if str(answer) == str(answer)[::-1]:
return answer
answer += 1
def isPrime(n, primes):
for (p for p in primes if p<=n**0.5):
if n%p == 0:
return False
return True
Upvotes: 0