[DFW.pm] Homework for the list, and for Oct 08 meeting

Robert Flach robert.flach at webtooldeveloper.com
Thu Sep 11 11:01:23 PDT 2014

Well, it sounds like it was a fun meeting.  I'm sorry I missed it!  I 
personally will occasionally use the FizzBuzz test when interviewing 
applicants though I usually use more complicated coding challenges.  
Maybe I could share one or more of those sometime as a "more advanced" 
challenge.  In the meantime, scroll down for my solution and for the 
things I look for when evaluating a fizzbuzz...

use strict;
use warnings;
my $x = 0;
while($x++ < 100)
   my $val = "";
   $val .= ( $x % 3 == 0 ) ? "fizz" : "";
   $val .= ( $x % 5 == 0 ) ? "buzz" : "";
   print STDOUT ( $val ? $val : "$x" ) . "\n";
Things I look for in a fizz buzz:
1. (Dealbreaker) Is the code readable. (I have a pretty loose definition 
of readable and it definitely doesn't require comments, but if it's 
unintelligible it's a dealbreaker )
2. (Dealbreaker) Will it produce correct output barring any minor syntax 
errors i.e. is the algorithm valid.
3. If doing a separate test for fizzbuzz have they optimized to a mod 15 
4. Can they (with prompting if needed) make the leap to realizing they 
can concat the two tests to produce the combined value without a 
separate test
5. If using real code: will it compile.
6. If using real code: will it compile without errors or warnings (e.g. 
for perl can I stick a use strict;use warnings; at the top of it.

Things I never care about in a fizzbuzz:
1. Do they know the correct operator for modulus (as long as they know 
modulus is possible)
2. while vs for vs. foreach, variable initialization, ternary vs. if, 
perl vs c vs (insert your favorite language argument here)
3. correct formatting of output (e.g. I don't mind if they forget to 
newline it, nor do I give extra points for printf )
4. error handling, supporting input arguments, etc. (this is an 
algorithm test, not a coding practices test and a coding practices test 
is unsuitable for the time available)

Things somewhat unrelated to fizzbuzz but discernible in some way 
related to the fizzbuzz tests that I definitely care about/give extra 
credit for.
1. Do they confirm or clarify the requirements with me in some way e.g. 
"When you say instead prints fizz buzz or fizzbuzz do you mean that you 
don't want it to print those numbers at all and only print the alternate 
text, or would you like it to always print the numbers and print that in 
addition for those meeting the requirements?"  It doesn't matter if 
their question is clearly answered in the requirements as written, 
restating the requirements as a question is a hugely valuable (and often 
undervalued) developer skill.
2. Similarly to number 1, do they ask questions about what I want in the 
solution (e.g. all those things in numbers 2-4 under things I never care 
about above) I don't care about those things (for the test; very 
different standard for "real" code), but many people do, and I DO care 
about your ability to realize that and your willingness to adjust your 
style accordingly.
3. I will always ask for some modification of the code afterwards.  I 
care a lot about how they respond to that request, whether it is a 
request to make it do something different, to change the styling, to 
comment it, etc.
4. If using pseudo code does their pseudo code look like a human 
language construction or is it just code with invalid syntax. (Either 
way, once they are done I will almost always ask for something that 
would compile/interpret correctly in some language.

Robert Flach
*Web Tools*
On 9/11/2014 12:15 PM, Lisa Cloutier wrote:
> Solution below:
> my $x = 1;
> while ($x <= 100)
> {
>     if (($x % 3 == 0) && ($x % 5 == 0))
>   {
>    print "FizzBuzz\n";
>   }
>   elsif ($x % 3 == 0)
>   {
>    print "Fizz\n";
>   }
>   elsif ($x % 5 == 0)
>   {
>    print "Buzz\n";
>   }
>   else
>   {
>    print "$x\n";
>   }
>   $x++;
> }
> I minorly "cheated" in that I didn't know the sign in Perl for "mod" 
> but had read Bob's email previously (before learning about the 
> challenge) and realized what he was doing with the % sign.   Had I 
> been in an interview I probably would have used every special 
> character in turn until I figured out the symbol for mod.
> So I guess I can "program my way out of a wet paper bag" as this 
> website <http://c2.com/cgi/wiki?FizzBuzzTest> proclaimed about the 
> FizzBuzz test.
> On Thu, Sep 11, 2014 at 11:42 AM, John Fields <wigthft at gmail.com 
> <mailto:wigthft at gmail.com>> wrote:
>     Firstly,  I want to thank John Dexter for his Docker presentation,
>     showing how to encapsulate a Mojolicious Web server and
>     application for easy deployment. He got what all presenters get, a
>     free dinner and that warm afterglow from making the world a better
>     place (with more Perl in it).  :)
>     I challenged the attendees last night to do the FizzBuzz program. 
>     We even had one programmer in attendance that had been asked to do
>     it on a job interview the previous day!  We were one day late for
>     him, but hopefully not for you..
>     Http://rosettecode.org/wiki/FizzBuzz
>     Don't cheat.. Yourself.  Do it from scratch before looking at
>     other's solutions.  We will pick 2 more, with increasing
>     difficulty with one per week.  Also new member Andy Sohn will
>     offer a short challenge and demonstrate answers live at the next
>     Mongers.
>     So send in your code, and be as creative as you can!  TIMTOWTDI is
>     a strength after all.  :)
>     Cheers,
>     John and Tommy
>     _______________________________________________
>     Dfw-pm mailing list
>     Dfw-pm at pm.org <mailto:Dfw-pm at pm.org>
>     http://mail.pm.org/mailman/listinfo/dfw-pm
> _______________________________________________
> Dfw-pm mailing list
> Dfw-pm at pm.org
> http://mail.pm.org/mailman/listinfo/dfw-pm

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.pm.org/pipermail/dfw-pm/attachments/20140911/e3672019/attachment-0001.html>

More information about the Dfw-pm mailing list