Comments on: Telogis Puzzle

By: Brian Adkins

Brian Adkins — Fri, 14 Dec 2007 03:32:06 +0000

@Ralph hmm… I hadn’t thought about the problem of publishing a solution too soon. If you check back here, let me know what you think would be an appropriate waiting period before posting a solution. Although, I’m sure you can find out rather quickly if a candidate is the type to simply look up a solution instead of solving it themselves.

By: Ralph Mason

Ralph Mason — Wed, 12 Dec 2007 19:42:37 +0000

It was written in ruby originally, but was tweaked not to run in *any* language. I will be constructing another one soon – the answer to this one is everywhere now.

By: Brian Adkins

Brian Adkins — Tue, 25 Sep 2007 18:24:23 +0000

Thanks for the link Ed. I’ve copied a bunch of puzzles from that site. I’ll post them here periodically. Something tells me that the slowness of Ruby might be a hindrance for some of them; I may have to switch to Lisp.

By: Ed

Ed — Sat, 22 Sep 2007 11:25:14 +0000

http://www.itasoftware.com/careers/puzzles07.html#Sling_Blade_Runner

By: Ed

Ed — Sat, 22 Sep 2007 11:24:44 +0000

Here is another set of programming puzzles that you have to solve in order to get your foot in the door to this company (ITA):

By: Brian Adkins

Brian Adkins — Thu, 20 Sep 2007 17:40:24 +0000

Ok, as promised, here's the Ruby solution to get the name of the web page. Ruby doesn't have a post-decrement operator, so a=b-- was translated to a = b; b -= 1. Also, Ruby has a somewhat different definition of truth, so expressions of the form (x) had to be translated to (x != 0) in cases where non-zero was supposed to be true.

require 'pp'

z = 4254145 * 0x18712495
b = (z & 1) << 4
a = b
b -= 1
o = []
c = a - b
x = b - a
while z != 0
  x = (a & x) != 0 ? a & b : c + x
  if (z / a & b^x) == 0
    o[z & b^x] = (x + 6 * a + b / 5 - c).chr
    z = z / a / a
  end
  pp o
end

pp o