Cody Boone Ferguson
009e1b7522
The location of the link is in the About menu under the 'Contacting the IOCCC' and above 'Thanks for the help'. At first I had it at the very bottom, under 'IOCCC markdown guidelines', but it seemed better to have it closer to contacting the IOCCC. This is the same location in the no JavaScript menu (which, for those who do not know or forget, is not a requirement except for some screen sizes). The top level Makefile has a new 'security' rule. Updated make help. Typo fixes in make help. SECURITY.html has had a few updates as well: change the wording of entry code is unsupported (it now says 'ALL CODE IN ALL IOCCC ENTRIES IS UNSUPPORTED!!') and it now links to the mkiocccentry repo and how to report a possible security vulnerability in that repo; as that IS supported - and since there was reference to discussing security concerns is in SECURITY.html - even though the entries will not receive security updates - it seemed a good idea to have a link to report such security issues in the mkiocccentry repo. Perhaps the FAQ needs to be updated for this too.
To build:
make all
To use:
./imc [NUMBER]
where [NUMBER] is an optional number. Default is 5.
Try:
./try.sh
Judges' remarks:
This entry's algorithm is as magic as its output!
Author's remarks:
This program may be compiled with an ANSI or K&R compiler. A few
harmless warnings are displayed only if gcc -Wall is used.
This entry is really a set of library functions, but an example
main() function has been added so that the library can be tested. If
the program appears too large, consider that functions o() and s() can
each be separated from the rest of the program (except for the short
functions p() and r(), which they require) and used alone. However, the
main part of the program is the function e, which does require all the
other functions (apart from main()).
The program should be supplied with an integer parameter. If no parameter or an invalid parameter is given, then 5 is assumed. The maximum parameter is determined only by the amount of CPU time, virtual memory and display (or file) space available.
NOTICE to those who wish for a greater challenge:
If you want a greater challenge, don't read any further: just try to understand the program via the source.
If you get stuck, come back and read below for additional hints and information.
How this entry works:
OK, so you have probably seen magic square printers before. But what about one that deals with even sizes as well as odd ones, or one that prints out a different one each time (or attempts to, at any rate)?
I have formatted the important functions o(), s() and e() to show how useful
the word for is, and (in function e()) how useful the words if and
else are. In fact, hardly anything happens that isn't in one of these
instructions. I did this in order to simplify the code - the algorithms
used to make the random squares are quite complex without being shrouded
in obfuscated code! :-) Incidentally, I was surprised to find out how
useful the exclusive-or operator was while writing function s().
Here are the descriptions of the functions in the library:
o(n,a,q,d): makes a magic square of order n when n is odd and at least 3.
s(n,a,q,d): makes a magic square of order n when n equals 4.
e(n,a,q,d): makes a magic square of order n when n is even and at least 6.
In the above, a (of type int *) points to an area of memory in which
the magic square will be stored and q (also of type int *) points to
an area of memory of length at least n*sizeof(int) bytes which can be
used as a work space. Both a and q must be allocated by the caller.
The integer parameter d, usually zero, indicates the length of a gap
which will be left between adjacent rows of the magic square.
The square which is returned will (or should) be a permutation of the
numbers 1 to n*n in which all rows and columns and the two diagonals
add up to n*(n*n+1)/2.