mirror/ioccc
1
0 Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

force use of pandoc version >= 3.1.12.2

Browse Source 
The `bin/pandoc-wrapper.sh` tool will verify that the ssytem
has a recent enough version of `pandoc(1)`.

We explain why we insist on `pandoc(1)`. version >= 3.1.12.2
in comments where the $PANDOC_MIN_VERSION is set.

Added `--wrap=preserve` as a default argument to `pandoc(1)`.
While this change will impact every HTML file generated,
we believe this will improve the ability of someone to look
at the generated HTML and correlate with the original markdown.

Changed `bin/pandoc-wrapper.sh` VERSION from "1.4.2 2024-03-17"
to "1.4.3 2024-03-18".

Performed `make www` to achieve all of the above and to generate
HTML using the new  `--wrap=preserve` `pandoc(1)` mode.
This commit is contained in:
Landon Curt Noll committed 2024-03-18 01:26:39 -07:00
1 parent 9eb7a54e9f
commit b0b212acb1
402 files changed
+37061 -55132
Download Patch File Download Diff File Expand all files Collapse all files

No files matched your search

2005/anon/index.html
+255 -331
View File
@@ -129,9 +129,9 @@
<h2 id="author"><a name="author"></a>Author:</h2>
<ul>
<li>Name: <a href="../../authors.html#Anonymous_2005">Anonymous
2005</a><br> Location: <a href="../../location.html#ZZ">ZZ</a> -
<em>Unknown location</em> <br><br></li>
<li>Name: <a href="../../authors.html#Anonymous_2005">Anonymous 2005</a><br>
Location: <a href="../../location.html#ZZ">ZZ</a> - <em>Unknown location</em>
<br><br></li>
</ul>
<!-- END: next line ends content generated by: bin/output-index-author.sh -->
@@ -141,36 +141,32 @@
<h2 id="to-build">To build:</h2>
<div class="sourceCode" id="cb1"><pre class="sourceCode sh"><code class="sourceCode bash"><span id="cb1-1"><a href="#cb1-1" aria-hidden="true" tabindex="-1"></a><span class="fu">make</span></span></code></pre></div>
<p>NOTE: there is an alternate version for vi(m) like movements. See <a
href="#alternate-code">alternate version</a> below. If your terminal has
problems with the <code>stty(1)</code> commands try:</p>
<p>NOTE: there is an alternate version for vi(m) like movements. See <a href="#alternate-code">alternate
version</a> below. If your terminal has problems with the
<code>stty(1)</code> commands try:</p>
<div class="sourceCode" id="cb2"><pre class="sourceCode sh"><code class="sourceCode bash"><span id="cb2-1"><a href="#cb2-1" aria-hidden="true" tabindex="-1"></a><span class="fu">make</span> anon-no_stty</span></code></pre></div>
<p>This will compile <a href="anon.c">anon.c</a> as <code>anon</code> to
not use <code>stty(1)</code>.</p>
<p>This will compile <a href="anon.c">anon.c</a> as <code>anon</code> to not use <code>stty(1)</code>.</p>
<h3 id="bugs-and-misfeatures">Bugs and (Mis)features:</h3>
<p>The current status of this entry is:</p>
<pre><code>STATUS: INABIAF - please **DO NOT** fix</code></pre>
<p>For more detailed information see <a
href="../../bugs.html#2005_anon">2005 anon bugs</a>.</p>
<p>For more detailed information see <a href="../../bugs.html#2005_anon">2005 anon bugs</a>.</p>
<h2 id="to-use">To use:</h2>
<div class="sourceCode" id="cb4"><pre class="sourceCode sh"><code class="sourceCode bash"><span id="cb4-1"><a href="#cb4-1" aria-hidden="true" tabindex="-1"></a><span class="ex">./anon</span> x y <span class="pp">[</span><span class="ss">z</span><span class="pp">]</span></span></code></pre></div>
<h2 id="try">Try:</h2>
<div class="sourceCode" id="cb5"><pre class="sourceCode sh"><code class="sourceCode bash"><span id="cb5-1"><a href="#cb5-1" aria-hidden="true" tabindex="-1"></a><span class="ex">./anon</span> 4 4</span>
<span id="cb5-2"><a href="#cb5-2" aria-hidden="true" tabindex="-1"></a></span>
<span id="cb5-3"><a href="#cb5-3" aria-hidden="true" tabindex="-1"></a><span class="ex">./anon</span> 3 3 3</span></code></pre></div>
<p>After a very large number of moves this program will deliberately
force itself to destroy its runtime stack. This is the author’s way to
prevent you from being frustrated (which might happen as it likes to
create unsolvable puzzles just to hook you) and spending too much time
playing. :-)</p>
<p>After a very large number of moves this program will deliberately force itself
to destroy its runtime stack. This is the author’s way to prevent you from being
frustrated (which might happen as it likes to create unsolvable puzzles just to
hook you) and spending too much time playing. :-)</p>
<p>If you want to see this in action you can easily do this:</p>
<div class="sourceCode" id="cb6"><pre class="sourceCode sh"><code class="sourceCode bash"><span id="cb6-1"><a href="#cb6-1" aria-hidden="true" tabindex="-1"></a><span class="fu">yes</span> <span class="st">&#39;please crash on me :-)&#39;</span> <span class="kw">|</span> <span class="ex">./anon</span> 4 4</span></code></pre></div>
<p>What happens if you specify a larger board? For instance:</p>
<div class="sourceCode" id="cb7"><pre class="sourceCode sh"><code class="sourceCode bash"><span id="cb7-1"><a href="#cb7-1" aria-hidden="true" tabindex="-1"></a><span class="fu">yes</span> <span class="st">&#39;please crash on me :-)&#39;</span> <span class="kw">|</span> <span class="ex">./anon</span> 10 10 10</span></code></pre></div>
<p>If you specify more than three args the program might also crash or
do something strange. This might also happen if you specify excessively
large board dimensions. Try <code>100 100 100</code> for instance and
see what happens!</p>
<p>If you specify more than three args the program might also crash or do
something strange. This might also happen if you specify excessively large board
dimensions. Try <code>100 100 100</code> for instance and see what happens!</p>
<h2 id="alternate-version">Alternate version:</h2>
<p>The alternate version uses <code>vi(m)</code> like movement keys.</p>
<h3 id="alternate-build">Alternate build:</h3>
@@ -178,8 +174,7 @@ see what happens!</p>
<p>To compile the version without <code>stty(1)</code>:</p>
<div class="sourceCode" id="cb9"><pre class="sourceCode sh"><code class="sourceCode bash"><span id="cb9-1"><a href="#cb9-1" aria-hidden="true" tabindex="-1"></a><span class="fu">make</span> alt-no_stty</span></code></pre></div>
<h3 id="alternate-use">Alternate use:</h3>
<p>Use <code>anon.alt</code> as you would <code>anon</code> with the
change in movements:</p>
<p>Use <code>anon.alt</code> as you would <code>anon</code> with the change in movements:</p>
<pre><code>k | Move empty spot up
h | Move empty spot left
j | Move empty spot down
@@ -187,43 +182,36 @@ l | Move empty spot right
J | Move empty spot forwards
K | Move empty spot backwards</code></pre>
<h2 id="judges-remarks">Judges’ remarks:</h2>
<p>Taking the 4x4 board with 15 tiles to a whole new level, this entry
takes this puzzle to a whole new dimension! While you puzzle out this
puzzle, we invite you to take a good LONG look how and WHY
<code>main()</code> is recursively called and how the runtime stack is
used.</p>
<p>Taking the 4x4 board with 15 tiles to a whole new level, this entry takes this
puzzle to a whole new dimension! While you puzzle out this puzzle, we invite
you to take a good LONG look how and WHY <code>main()</code> is recursively called and how
the runtime stack is used.</p>
<h2 id="authors-remarks">Author’s remarks:</h2>
<h3 id="about-the-game-itself">ABOUT THE GAME ITSELF</h3>
<p>Remember Sam Loyd’s <a
href="https://en.wikipedia.org/wiki/15_puzzle">Fifteen Puzzle</a>? It
consists of a 4x4 board with 15 tiles numbered 1 to 15, and an empty
spot. The goal of the game is to slide the tiles around without lifting
them off the board, so that they become sorted.</p>
<p>Well, that was then. This program is the 21st century version of the
Fifteen Puzzle: not only does it allow arbitrary board sizes, it also
supports <em>3D</em> boards! The tiles are numbered from 1 to N, where N
is one less the number of spaces on the board, and the goal is to
rearrange them so that they appear sorted in the order they appear on
the screen.</p>
<p>The program takes 2 or 3 arguments, specifying the dimensions of the
board. For example, to play the original version of the Fifteen
Puzzle:</p>
<div class="sourceCode" id="cb11"><pre
class="sourceCode sh"><code class="sourceCode bash"><span id="cb11-1"><a href="#cb11-1" aria-hidden="true" tabindex="-1"></a><span class="ex">./anon</span> 4 4</span></code></pre></div>
<p>Remember Sam Loyd’s <a href="https://en.wikipedia.org/wiki/15_puzzle">Fifteen Puzzle</a>?
It consists of a 4x4 board with 15 tiles numbered 1 to 15, and an empty spot.
The goal of the game is to slide the tiles around without lifting them off the
board, so that they become sorted.</p>
<p>Well, that was then. This program is the 21st century version of the Fifteen
Puzzle: not only does it allow arbitrary board sizes, it also supports <em>3D</em>
boards! The tiles are numbered from 1 to N, where N is one less the number of
spaces on the board, and the goal is to rearrange them so that they appear
sorted in the order they appear on the screen.</p>
<p>The program takes 2 or 3 arguments, specifying the dimensions of the board.
For example, to play the original version of the Fifteen Puzzle:</p>
<div class="sourceCode" id="cb11"><pre class="sourceCode sh"><code class="sourceCode bash"><span id="cb11-1"><a href="#cb11-1" aria-hidden="true" tabindex="-1"></a><span class="ex">./anon</span> 4 4</span></code></pre></div>
<p>To play with a 3D board, do something like this:</p>
<div class="sourceCode" id="cb12"><pre
class="sourceCode sh"><code class="sourceCode bash"><span id="cb12-1"><a href="#cb12-1" aria-hidden="true" tabindex="-1"></a><span class="ex">./anon</span> 3 3 3</span></code></pre></div>
<p><strong>WARNING: do NOT run the program with more than 3 command-line
arguments. This will trigger a bug that will cause it to crash or do
something strange. :-)</strong></p>
<p>Also, resist the temptation to create very large puzzles. The program
does not attempt to resize your terminal for you, and will produce
confusing output. It may also crash on you if you insist on running it
with excessively large board dimensions. :-) (10x10x10 is considered a
safe upper limit.)</p>
<p>The empty space on the board is represented by a space filled with
<code>#</code> characters instead of digits. To slide the tiles around,
use the following keys:</p>
<div class="sourceCode" id="cb12"><pre class="sourceCode sh"><code class="sourceCode bash"><span id="cb12-1"><a href="#cb12-1" aria-hidden="true" tabindex="-1"></a><span class="ex">./anon</span> 3 3 3</span></code></pre></div>
<p><strong>WARNING: do NOT run the program with more than 3 command-line arguments.
This will trigger a bug that will cause it to crash or do something strange.
:-)</strong></p>
<p>Also, resist the temptation to create very large puzzles. The program does not
attempt to resize your terminal for you, and will produce confusing output. It
may also crash on you if you insist on running it with excessively large board
dimensions. :-) (10x10x10 is considered a safe upper limit.)</p>
<p>The empty space on the board is represented by a space filled with <code>#</code>
characters instead of digits. To slide the tiles around, use the following
keys:</p>
<pre><code>i | Slide the tile above the current empty space downwards.
j | Slide the tile to the left of the current empty space to the right.
k | Slide the tile below the current empty space upwards.
@@ -233,101 +221,87 @@ n | Slide the tile in front of the current empty space backwards.</code></pre>
<p>To quit the game:</p>
<pre><code>q | Quit the game.</code></pre>
<p>Note that, for example, sliding the tile above the current empty spot
downwards will fill up the empty spot, and create a new empty spot above
where the current empty spot was. Hence, you could think of the movement
keys in terms of how they effectively “move” the empty spot:</p>
downwards will fill up the empty spot, and create a new empty spot above where
the current empty spot was. Hence, you could think of the movement keys in
terms of how they effectively “move” the empty spot:</p>
<pre><code>i | Move empty spot up
j | Move empty spot left
k | Move empty spot down
l | Move empty spot right
n | Move empty spot forwards
o | Move empty spot backwards</code></pre>
<p><strong>NOTE:</strong> as a measure of alleviating
counter-productivity due to addiction to this game, particularly for
those hardcore gamers who insist on solving a 10x10x10 puzzle during
work hours, and also to give an upper bound to user frustration when the
program generates an unsolvable puzzle (which it does roughly half the
time, just to hook you), the program will, after a very large number of
moves (which is system-specific) terminate itself by forcefully
destroying its own runtime stack. This will result in a core dump and/or
an abnormal termination message; but this is a feature, not a bug!
:-)</p>
<p><strong>NOTE:</strong> as a measure of alleviating counter-productivity due to addiction to
this game, particularly for those hardcore gamers who insist on solving a
10x10x10 puzzle during work hours, and also to give an upper bound to user
frustration when the program generates an unsolvable puzzle (which it does
roughly half the time, just to hook you), the program will, after a very large
number of moves (which is system-specific) terminate itself by forcefully
destroying its own runtime stack. This will result in a core dump and/or an
abnormal termination message; but this is a feature, not a bug! :-)</p>
<h3 id="compile-notes">COMPILE NOTES</h3>
<p>This program complies with ANSI/C99. It uses ANSI terminal control
sequences, and so requires a terminal that understands ANSI sequences,
such as VT100 or xterm.</p>
<p>This program complies with ANSI/C99. It uses ANSI terminal control sequences,
and so requires a terminal that understands ANSI sequences, such as VT100 or
xterm.</p>
<p>Some compilers may emit lots of warnings about how things ought to be
parenthesized (e.g. <code>gcc -Wall</code>). This is deliberate, since
if everything were parenthesized “properly” the program would fail to
work. It (ab)uses C operator precedence to accomplish some cool hacks.
:-) However, there are no compiler errors. I have tested this with
<code>gcc -pedantic -std=c99</code>.</p>
<p>The program also uses <code>stty cbreak</code>. If this breaks on
your system, compile it with <code>-DNO_STTY</code>. The game interface
will be a bit annoying (you will need to hit enter after every key), but
it works.</p>
<p>The game requires lots of space on the runtime stack space due to
deep recursion, among other things. :-)</p>
parenthesized (e.g. <code>gcc -Wall</code>). This is deliberate, since if everything were
parenthesized “properly” the program would fail to work. It (ab)uses C
operator precedence to accomplish some cool hacks. :-) However, there are no
compiler errors. I have tested this with <code>gcc -pedantic -std=c99</code>.</p>
<p>The program also uses <code>stty cbreak</code>. If this breaks on your system, compile it
with <code>-DNO_STTY</code>. The game interface will be a bit annoying (you will need to
hit enter after every key), but it works.</p>
<p>The game requires lots of space on the runtime stack space due to deep
recursion, among other things. :-)</p>
<p>See spoilers below for more notes about portability.</p>
<h3 id="about-the-code">ABOUT THE CODE</h3>
<p>Clearly, the author is very, very confused. Although he has broken up
the file into sections labelled with helpful headings, and although he
has left ample whitespace between all those <code>#define</code>s to
make them easier to read, :-) and although he has helpfully formatted
the code with beautiful indentation, something is clearly very, very
wrong.</p>
<p>In fact, the closer you look at this code, the more you realize
something is horribly wrong with it. For example, why is it that the
first statement in <code>main()</code> is a <code>return</code>? What do
all those nested <code>do</code> loops do (pun intended)? Why is the
first statement inside the first <code>do</code> loop a mere integer
literal? Why are there such strange statements, like that empty string
on line 65, or that isolated <code>||</code> on line 68?</p>
<p>But wait… why aren’t any statements terminated by semicolons? Why is
there only ONE semicolon in the entire program?! Why is the author
opening nested blocks by such strange things as <code>||</code> and
<code>(</code>, and closing them by <code>||</code> and <code>))</code>?
The <code>for</code> loop isn’t a <code>for</code> loop! The
<code>do</code> loop isn’t a <code>do</code> loop! Or… are they? Well,
the <code>for</code> loop is defined in terms of a <code>while</code>
loop. But wait, the <code>while</code> loop is defined in terms of the
<code>do</code> loop. And the <code>do</code> loop is defined in terms
of … Huh???</p>
<p>And that <code>switch</code> statement toward the end of the program…
wait a minute, that’s not a <code>switch</code> statement! Why is a
<code>case</code> defined in terms of a <code>switch</code> statement?!
And that initial return isn’t a return! Well, it is… but not what you
might think!</p>
<p>Clearly, the author is very, very confused. Although he has broken up the file
into sections labelled with helpful headings, and although he has left ample
whitespace between all those <code>#define</code>s to make them easier to read, :-) and
although he has helpfully formatted the code with beautiful indentation,
something is clearly very, very wrong.</p>
<p>In fact, the closer you look at this code, the more you realize something is
horribly wrong with it. For example, why is it that the first statement in
<code>main()</code> is a <code>return</code>? What do all those nested <code>do</code> loops do (pun intended)? Why
is the first statement inside the first <code>do</code> loop a mere integer literal? Why
are there such strange statements, like that empty string on line 65, or that
isolated <code>||</code> on line 68?</p>
<p>But wait… why aren’t any statements terminated by semicolons? Why is there
only ONE semicolon in the entire program?! Why is the author opening nested
blocks by such strange things as <code>||</code> and <code>(</code>, and closing them by <code>||</code> and <code>))</code>? The
<code>for</code> loop isn’t a <code>for</code> loop! The <code>do</code> loop isn’t a <code>do</code> loop! Or… are they? Well,
the <code>for</code> loop is defined in terms of a <code>while</code> loop. But wait, the <code>while</code> loop is
defined in terms of the <code>do</code> loop. And the <code>do</code> loop is defined in terms of …
Huh???</p>
<p>And that <code>switch</code> statement toward the end of the program… wait a minute,
that’s not a <code>switch</code> statement! Why is a <code>case</code> defined in terms of a <code>switch</code>
statement?! And that initial return isn’t a return! Well, it is… but not what
you might think!</p>
<p>Let’s see if it all makes sense after we run it through CPP…</p>
<p>Uh oh. The result looks like the diseased imagination of a mind more
confused than the one who wrote the original code. Unfortunately, unless
you’re a Lisp expert, it is probably easier to understand the program in
its original form. (And even then.) But that’s just so, so wrong…</p>
<p>The entire program consists of a single expression, returned from
<code>main()</code>. There are no <code>if</code> statements, no
<code>for</code> loops or any looping or control constructs besides the
<code>return</code>, in spite of the misleading appearance of the
original source. :) Yes I know, this has been attempted before, using
<code>?:</code> and the bad old comma operator. BUT this program does
NOT use the <code>?:</code> operator, and there are only TWO comma
operators in the entire program–and they are only there because
<code>srand()</code> and <code>exit()</code> return void.</p>
<p>The entire program consists of a single boolean expression returned
from <code>main()</code>. Looping is achieved by recursion, of course,
and branching is achieved by exploiting short-circuit evaluation of
boolean expressions. :-)</p>
<p>Uh oh. The result looks like the diseased imagination of a mind more confused
than the one who wrote the original code. Unfortunately, unless you’re a Lisp
expert, it is probably easier to understand the program in its original form.
(And even then.) But that’s just so, so wrong…</p>
<p>The entire program consists of a single expression, returned from <code>main()</code>.
There are no <code>if</code> statements, no <code>for</code> loops or any looping or control constructs
besides the <code>return</code>, in spite of the misleading appearance of the original
source. :) Yes I know, this has been attempted before, using <code>?:</code> and the bad
old comma operator. BUT this program does NOT use the <code>?:</code> operator, and there
are only TWO comma operators in the entire program–and they are only there
because <code>srand()</code> and <code>exit()</code> return void.</p>
<p>The entire program consists of a single boolean expression returned from
<code>main()</code>. Looping is achieved by recursion, of course, and branching is achieved
by exploiting short-circuit evaluation of boolean expressions. :-)</p>
<p>But this is not all.</p>
<p>It is comprehensible that the program can achieve branching and
looping using only boolean expressions and recursion. But HOW does the
program STORE DATA??? There are no variables of any kind in the program
except for the two arguments to <code>main()</code>. There are certainly
no arrays declared anywhere. The program does not call any memory
allocation routines, nor any function that indirectly returns allocated
memory. Yet it happily looks up array references without crashing.
How??</p>
<p>One might be tempted to think that this program would be highly
non-portable because of such insane hacks. However, it has been tested
and proven to work on the following architectures:</p>
<p>It is comprehensible that the program can achieve branching and looping using
only boolean expressions and recursion. But HOW does the program STORE DATA???
There are no variables of any kind in the program except for the two arguments
to <code>main()</code>. There are certainly no arrays declared anywhere. The program does
not call any memory allocation routines, nor any function that indirectly
returns allocated memory. Yet it happily looks up array references without
crashing. How??</p>
<p>One might be tempted to think that this program would be highly non-portable
because of such insane hacks. However, it has been tested and proven to work
on the following architectures:</p>
<ul>
<li>i386/linux</li>
<li>sparc/solaris</li>
@@ -339,132 +313,107 @@ and proven to work on the following architectures:</p>
<li>ia64/linux</li>
<li>sparc64/linux</li>
</ul>
<p>It probably will also compile and work on many (all?) other *nix
platforms as well. How does it achieve this incredible feat?</p>
<p>It probably will also compile and work on many (all?) other *nix platforms as
well. How does it achieve this incredible feat?</p>
<h3 id="spoiler">SPOILER</h3>
<p>Read the following only if you give up figuring it out :-)</p>
<h4 id="the-evil-hack">The Evil Hack</h4>
<p>The program (ab)uses an implicit assumption underlying practically
all architectures, and that is that parameters to C functions are
implicit local variables allocated on the runtime stack. This runtime
stack is a contiguous section of memory, and therefore, after
<code>N</code> recursive calls of <code>main()</code> to itself, there
is enough room on the stack to store whatever the program desires, so
long as it does not exceed the maximum stack size.</p>
<p>(Theoretically, the runtime stack doesn’t <strong>have</strong> to be
contiguous… but all current architectures that are worth consideration
make this assumption. I submit that this implies this program is 100%
portable, even though theoretically it shouldn’t even work in the first
place.) :-)</p>
<p>There are two types of architectures the program must handle: those
that allocate the stack top-down, and those that allocate the stack
bottom-up. The program correctly detects this at runtime, and adjusts
itself accordingly. :-) (It is a simple matter of testing if the address
of <code>argc</code> in a recursive call is greater or less than in the
parent call.)</p>
<p>To avoid having to handle issues of <code>int</code> sizes, memory
alignment, what exactly is on the stack between successive locations of
<code>argc</code>, etc., the program (ab)uses the fact that
<code>argc</code> must necessarily be allocated on an aligned address,
and therefore, the range of memory between two <code>argc</code>s is
aligned, and therefore safe to use as an array of <code>long</code>s.[*]
The program also compares the pointers as <code>long*</code> pointers,
so that it will not have to deal with mapping bytes to larger types. The
initial memory allocation is done by recursively calling
<code>main()</code> until there is enough space on the stack (the exact
number of recursive calls is, of course, architecture-dependent).</p>
<p>[*] <code>long</code> was chosen instead of <code>int</code>, in
order to skirt around pointer-size issues on architectures like ia64. It
really can be made anything large enough to hold both <code>int</code>s
and pointers. <code>:-)</code></p>
<p>After the stack is deemed to have grown deep enough to be a
sufficiently <code>long</code> array (pun intended), the program begins
its real code. It uses <code>argc</code> to keep track of what it’s
supposed to be doing (since <code>main()</code> is the only function),
and <code>argv</code> to point to the base of this array (which is
computed differently based on how the architecture allocates the stack).
Since this fills up the only two variables in the program, everything
else is stored in the <code>MASS</code> [**] array.</p>
<p>The program (ab)uses an implicit assumption underlying practically all
architectures, and that is that parameters to C functions are implicit local
variables allocated on the runtime stack. This runtime stack is a contiguous
section of memory, and therefore, after <code>N</code> recursive calls of <code>main()</code> to
itself, there is enough room on the stack to store whatever the program desires,
so long as it does not exceed the maximum stack size.</p>
<p>(Theoretically, the runtime stack doesn’t <strong>have</strong> to be contiguous… but all
current architectures that are worth consideration make this assumption. I
submit that this implies this program is 100% portable, even though
theoretically it shouldn’t even work in the first place.) :-)</p>
<p>There are two types of architectures the program must handle: those that
allocate the stack top-down, and those that allocate the stack bottom-up. The
program correctly detects this at runtime, and adjusts itself accordingly.
:-) (It is a simple matter of testing if the address of <code>argc</code> in a recursive
call is greater or less than in the parent call.)</p>
<p>To avoid having to handle issues of <code>int</code> sizes, memory alignment, what exactly
is on the stack between successive locations of <code>argc</code>, etc., the program
(ab)uses the fact that <code>argc</code> must necessarily be allocated on an aligned
address, and therefore, the range of memory between two <code>argc</code>s is aligned, and
therefore safe to use as an array of <code>long</code>s.[*] The program also compares the
pointers as <code>long*</code> pointers, so that it will not have to deal with mapping
bytes to larger types. The initial memory allocation is done by recursively
calling <code>main()</code> until there is enough space on the stack (the exact number of
recursive calls is, of course, architecture-dependent).</p>
<p>[*] <code>long</code> was chosen instead of <code>int</code>, in order to skirt around pointer-size
issues on architectures like ia64. It really can be made anything large enough
to hold both <code>int</code>s and pointers. <code>:-)</code></p>
<p>After the stack is deemed to have grown deep enough to be a sufficiently <code>long</code>
array (pun intended), the program begins its real code. It uses <code>argc</code> to keep
track of what it’s supposed to be doing (since <code>main()</code> is the only function),
and <code>argv</code> to point to the base of this array (which is computed differently
based on how the architecture allocates the stack). Since this fills up the
only two variables in the program, everything else is stored in the <code>MASS</code> [**]
array.</p>
<p>[**] for “Memory Allocated by Stack Smashing” :-)</p>
<p>Note that from this point on, all bets are off as to what the
contents of the runtime stack are below the current recursive call to
<code>main()</code>. We are overwriting frame pointers, return
addresses, etc., with data. Hence, the program must never return from
this call, otherwise it will crash horribly. That’s why I had to call
<code>exit()</code> even though it meant I can no longer claim the
program doesn’t need the comma operator. :-) (I was going to skirt
around the issue by indirectly calling <code>exit()</code>, such as via
<code>bsearch()</code> or <code>raise()</code>, but I was getting too
close to the size limit.)</p>
<p>Note that from this point on, all bets are off as to what the contents of the
runtime stack are below the current recursive call to <code>main()</code>. We are
overwriting frame pointers, return addresses, etc., with data. Hence, the
program must never return from this call, otherwise it will crash horribly.
That’s why I had to call <code>exit()</code> even though it meant I can no longer claim the
program doesn’t need the comma operator. :-) (I was going to skirt around the
issue by indirectly calling <code>exit()</code>, such as via <code>bsearch()</code> or <code>raise()</code>, but I
was getting too close to the size limit.)</p>
<h4 id="some-notes-about-the-macros">Some notes about the macros</h4>
<p>In spite of the frightening appearance, it is probably easiest to
understand the code in its original form. (Don’t even try running
<code>indent</code> on it: on my system it chokes and complains about
nesting errors.)</p>
<p>Most of the <code>#define</code>s are just to keep the code size
down, due to the abusive amounts of type casting employed. The others
define a bizarre “programming language” (in the crudest, most twisted
sense of the word) that the author used to write this mess.</p>
<p><code>R(i)</code> and <code>W(i,j)</code> are macros to read/write
the <code>i</code>th position of the <code>MASS</code>.</p>
<p><code>A(i)</code> retrieves the <code>i</code>th command-line
argument. This is slightly tricky, because by the time the
<code>MASS</code> is setup, we no longer have access to the original
values of <code>argc</code> and <code>argv</code>. So what is done is
that the initial memory allocation code will save <code>argc</code> and
<code>argv</code> into positions 8 and 9 in the <code>MASS</code> after
the allocation. It does this by using <code>argv</code>, which at that
time contains a pointer to the top-level <code>argc</code>, to compute
the location of the top-level <code>argv</code>. This is done before the
rest of the <code>MASS</code> is used, since otherwise they may be
overwritten.</p>
<p><code>do(x,y)</code> has nothing to do with <code>do</code> loops.
:-) It defines a “state” in <code>main()</code>, essentially, what to do
when <code>main()</code> is called with <code>argc</code> containing a
particular value. (This is why the user must never run the program with
more than 3 arguments. :-) The state number is <code>x</code>, and
<code>y</code> is the code. The program’s boolean expression consists of
a number of states <code>&amp;&amp;</code>d together. Within each state,
it is assumed that statements are separated by <code>||</code>. All
statements are assumed to evaluate to false. This makes it easy to write
conditional statements: <code>if (a) { b } ...</code> can simply be
written <code>a &amp;&amp; b || ...</code>. All states created by
<code>do(x,y)</code> are assumed to evaluate to false. There is only 1
state which can return true or false: state 5, a “manually-implemented”
state which determines whether the system uses top-down or bottom-up
stack allocation. It’s the only state that needs to return a value from
the boolean expression, since once the <code>MASS</code> is initialized
any return value can simply be stored in a well-known location in the
<code>MASS</code>.</p>
<p><code>if(p,q,r,s)</code> is, in fact, an <code>if</code> statement.
:-) It is equivalent to:</p>
<p>In spite of the frightening appearance, it is probably easiest to understand
the code in its original form. (Don’t even try running <code>indent</code> on it: on my
system it chokes and complains about nesting errors.)</p>
<p>Most of the <code>#define</code>s are just to keep the code size down, due to the abusive
amounts of type casting employed. The others define a bizarre “programming
language” (in the crudest, most twisted sense of the word) that the author
used to write this mess.</p>
<p><code>R(i)</code> and <code>W(i,j)</code> are macros to read/write the <code>i</code>th position of the <code>MASS</code>.</p>
<p><code>A(i)</code> retrieves the <code>i</code>th command-line argument. This is slightly tricky,
because by the time the <code>MASS</code> is setup, we no longer have access to the
original values of <code>argc</code> and <code>argv</code>. So what is done is that the initial memory
allocation code will save <code>argc</code> and <code>argv</code> into positions 8 and 9 in the <code>MASS</code>
after the allocation. It does this by using <code>argv</code>, which at that time contains
a pointer to the top-level <code>argc</code>, to compute the location of the top-level
<code>argv</code>. This is done before the rest of the <code>MASS</code> is used, since otherwise they
may be overwritten.</p>
<p><code>do(x,y)</code> has nothing to do with <code>do</code> loops. :-) It defines a “state” in <code>main()</code>,
essentially, what to do when <code>main()</code> is called with <code>argc</code> containing a
particular value. (This is why the user must never run the program with more
than 3 arguments. :-) The state number is <code>x</code>, and <code>y</code> is the code. The program’s
boolean expression consists of a number of states <code>&amp;&amp;</code>d together. Within each
state, it is assumed that statements are separated by <code>||</code>. All statements are
assumed to evaluate to false. This makes it easy to write conditional
statements: <code>if (a) { b } ...</code> can simply be written <code>a &amp;&amp; b || ...</code>. All
states created by <code>do(x,y)</code> are assumed to evaluate to false. There is only 1
state which can return true or false: state 5, a “manually-implemented” state
which determines whether the system uses top-down or bottom-up stack
allocation. It’s the only state that needs to return a value from the boolean
expression, since once the <code>MASS</code> is initialized any return value can simply be
stored in a well-known location in the <code>MASS</code>.</p>
<p><code>if(p,q,r,s)</code> is, in fact, an <code>if</code> statement. :-) It is equivalent to:</p>
<div class="sourceCode" id="cb16"><pre class="sourceCode c"><code class="sourceCode c"><span id="cb16-1"><a href="#cb16-1" aria-hidden="true" tabindex="-1"></a><span class="cf">if</span> <span class="op">(</span>p<span class="op">)</span> <span class="op">{</span> q<span class="op">;</span> r <span class="op">}</span> <span class="cf">else</span> <span class="op">{</span>s<span class="op">}</span></span></code></pre></div>
<p>Why are <code>q</code> and <code>r</code> split up? Because it’s
cleaner (in whatever twisted sense of the word that can be applied to an
IOCCC entry) to implement using boolean expressions. Keep in mind that
all statements are assumed to evaluate to false. The proof that this
macro in fact implements an <code>if-then-else</code> construct is left
as an exercise for the reader.</p>
<p><code>while(x,y,z)</code> is actually a <code>while</code> loop after
all. :-) It implements the loop using, of course, recursive calls to
<code>main()</code>. However, this macro doesn’t actually run a
<code>while</code> loop; it implements the loop in the state numbered
<code>x</code>. It is used for defining a <code>while</code> loop, which
can then be called from another state using <code>f(x)</code>. (The
reason we can’t directly implement a <code>while</code> loop inside
another state ought to be clear: we can’t recursively call
<code>main()</code> to enter into the middle of a state.)</p>
<p>However, it <em>is</em> possible to implement a <code>for</code> loop
directly inside a state! The <code>for(s,i,x,y,z,Z)</code> macro does
exactly this. Well, actually, technically speaking this macro can only
be used as the last thing in a state, because it secretly closes the
current state and opens a new one behind the programmer’s back. :-) Its
intended usage is this: <code>s</code> is the state number for the loop
body, and <code>i</code> is the index variable (well, the index into the
<code>MASS</code> of the index variable); <code>x</code> and
<code>y-1</code> define the starting and ending values of
<code>i</code>, and <code>z</code> is the loop body. <code>Z</code> is
the post-loop code in the current state. For example:</p>
<p>Why are <code>q</code> and <code>r</code> split up? Because it’s cleaner (in whatever twisted sense of
the word that can be applied to an IOCCC entry) to implement using boolean
expressions. Keep in mind that all statements are assumed to evaluate to
false. The proof that this macro in fact implements an <code>if-then-else</code> construct
is left as an exercise for the reader.</p>
<p><code>while(x,y,z)</code> is actually a <code>while</code> loop after all. :-) It implements the loop
using, of course, recursive calls to <code>main()</code>. However, this macro doesn’t
actually run a <code>while</code> loop; it implements the loop in the state numbered <code>x</code>. It
is used for defining a <code>while</code> loop, which can then be called from another state
using <code>f(x)</code>. (The reason we can’t directly implement a <code>while</code> loop inside
another state ought to be clear: we can’t recursively call <code>main()</code> to enter
into the middle of a state.)</p>
<p>However, it <em>is</em> possible to implement a <code>for</code> loop directly inside a state! The
<code>for(s,i,x,y,z,Z)</code> macro does exactly this. Well, actually, technically speaking
this macro can only be used as the last thing in a state, because it secretly
closes the current state and opens a new one behind the programmer’s back.
:-) Its intended usage is this: <code>s</code> is the state number for the loop body, and
<code>i</code> is the index variable (well, the index into the <code>MASS</code> of the index
variable); <code>x</code> and <code>y-1</code> define the starting and ending values of <code>i</code>, and <code>z</code>
is the loop body. <code>Z</code> is the post-loop code in the current state. For example:</p>
<div class="sourceCode" id="cb17"><pre class="sourceCode c"><code class="sourceCode c"><span id="cb17-1"><a href="#cb17-1" aria-hidden="true" tabindex="-1"></a><span class="cf">for</span><span class="op">(</span>s<span class="op">,</span>i<span class="op">,</span> <span class="dv">0</span><span class="op">,</span> <span class="dv">10</span><span class="op">,</span></span>
<span id="cb17-2"><a href="#cb17-2" aria-hidden="true" tabindex="-1"></a>some_code <span class="op">,</span></span>
<span id="cb17-3"><a href="#cb17-3" aria-hidden="true" tabindex="-1"></a>other_code<span class="op">)</span></span></code></pre></div>
@@ -473,29 +422,25 @@ the post-loop code in the current state. For example:</p>
<span id="cb18-2"><a href="#cb18-2" aria-hidden="true" tabindex="-1"></a> some_code<span class="op">;</span></span>
<span id="cb18-3"><a href="#cb18-3" aria-hidden="true" tabindex="-1"></a><span class="op">}</span></span>
<span id="cb18-4"><a href="#cb18-4" aria-hidden="true" tabindex="-1"></a>other_code<span class="op">;</span></span></code></pre></div>
<p>Note the “wrong” indentation in the former: the author prefers to
write it that way, since the analogy to the usual <code>for</code> loop
is clearer. This is why it’s easier to understand the pre-CPP code,
since after CPP, the macro reshuffles <code>some_code</code> after
<code>other_code</code> so that it can close the current state and
define the one containing the loop body. (Have I told you that the
author of this program is horribly, horribly, confused? :-) This
confusion also shows up in the superficial appearance of the program,
which heavily suffers from wrong indentation.)</p>
<p>In spite of all appearances, <code>switch()</code> really is just a
shorthand macro for reducing code size. The evil choice of macro name is
just to instill more confusion into the would-be decipherer of this
entry. :-) After all, <code>case()</code> really <em>is</em> the
equivalent of a <code>case</code> statement, implemented using
<code>&amp;&amp;</code>, and conveniently encapsulating code common to
the cases in the only <code>switch</code> statement in the program,
which is the code that decides what to do with a key press. What could
be more fun than to define such a case in terms of a
<p>Note the “wrong” indentation in the former: the author prefers to write it
that way, since the analogy to the usual <code>for</code> loop is clearer. This is why it’s
easier to understand the pre-CPP code, since after CPP, the macro reshuffles
<code>some_code</code> after <code>other_code</code> so that it can close the current state and define
the one containing the loop body. (Have I told you that the author of this
program is horribly, horribly, confused? :-) This confusion also shows up in
the superficial appearance of the program, which heavily suffers from wrong
indentation.)</p>
<p>In spite of all appearances, <code>switch()</code> really is just a shorthand macro for
reducing code size. The evil choice of macro name is just to instill more
confusion into the would-be decipherer of this entry. :-) After all, <code>case()</code>
really <em>is</em> the equivalent of a <code>case</code> statement, implemented using <code>&amp;&amp;</code>, and
conveniently encapsulating code common to the cases in the only <code>switch</code>
statement in the program, which is the code that decides what to do with a
key press. What could be more fun than to define such a case in terms of a
<code>switch</code>? :-)</p>
<p>Finally, for those who give up deciphering this horribly confused
code, here are the mappings of various locations of the
<code>MASS</code> to their usage in the program, and what the various
state numbers are meant to do:</p>
<p>Finally, for those who give up deciphering this horribly confused code, here
are the mappings of various locations of the <code>MASS</code> to their usage in the
program, and what the various state numbers are meant to do:</p>
<h4 id="mass-usage-map">MASS usage map</h4>
<pre><code>0 | Function return value
1 | Function argument 1
@@ -516,16 +461,14 @@ state numbers are meant to do:</p>
18 | Scratch register used by state 18 :-)
31 | field width for printing tiles
32 | Start of board data</code></pre>
<p>(<strong>CAVEAT:</strong> not all of these mappings ended up in the
final version of the program, mainly due to size limitations which
entailed the trimming off of a lot of neat features in the game. It is
left as an exercise for the reader to figure out which of these are
actually used.)</p>
<p>Note that <code>argv</code> and <code>argc</code> necessarily need to
stay away from lower indices of the <code>MASS</code>, because otherwise
they may overwrite their original values when the memory allocation code
saves them. Index 8 is deemed “safe enough”. (It seems to be on all
tested architectures.)</p>
<p>(<strong>CAVEAT:</strong> not all of these mappings ended up in the final version of the
program, mainly due to size limitations which entailed the trimming off of a
lot of neat features in the game. It is left as an exercise for the reader to
figure out which of these are actually used.)</p>
<p>Note that <code>argv</code> and <code>argc</code> necessarily need to stay away from lower indices of
the <code>MASS</code>, because otherwise they may overwrite their original values when the
memory allocation code saves them. Index 8 is deemed “safe enough”. (It seems
to be on all tested architectures.)</p>
<h4 id="state-number-assignments">State number assignments</h4>
<pre><code>1 | showhelp
2 | showhelp
@@ -543,62 +486,43 @@ tested architectures.)</p>
17 | inner loop of 16.
18 | main loop
19 | checkwin() loop</code></pre>
<p>Note that some of the manually-implemented states (those not
implemented by <code>do()</code>) respond to multiple state numbers.
Also note that states 1-4 conveniently correspond to what the program
should do when 1-4 arguments are given at the command-line. :-)</p>
<p>The original intention was to include a parity-checking routine in
the game so that it will always generate a solvable game. However, size
limits necessitated the removal of this code. And besides, the game is
more addictive when it’s sometimes unsolvable. Especially when it taunts
the user with a smiley. :-)</p>
<h2 id="copyright-and-cc-by-sa-4.0-license">Copyright and CC BY-SA 4.0
License:</h2>
<p>This file is Copyright (c) 2023 by Landon Curt Noll. All Rights
Reserved. You are free to share and adapt this file under the terms of
this license:</p>
<p>Note that some of the manually-implemented states (those not implemented by
<code>do()</code>) respond to multiple state numbers. Also note that states 1-4
conveniently correspond to what the program should do when 1-4 arguments are
given at the command-line. :-)</p>
<p>The original intention was to include a parity-checking routine in the game so
that it will always generate a solvable game. However, size limits
necessitated the removal of this code. And besides, the game is more addictive
when it’s sometimes unsolvable. Especially when it taunts the user with a
smiley. :-)</p>
<h2 id="copyright-and-cc-by-sa-4.0-license">Copyright and CC BY-SA 4.0 License:</h2>
<p>This file is Copyright (c) 2023 by Landon Curt Noll. All Rights Reserved.
You are free to share and adapt this file under the terms of this license:</p>
<pre><code>Creative Commons Attribution-ShareAlike 4.0 International (CC BY-SA 4.0)</code></pre>
<p>For more information, see:
https://creativecommons.org/licenses/by-sa/4.0/</p>
<p>For more information, see: https://creativecommons.org/licenses/by-sa/4.0/</p>
<!-- END: this line ends content for HTML phase 21 by: bin/pandoc-wrapper.sh via bin/md2html.sh -->
<!-- START: this line starts content for HTML phase 22 by: bin/output-index-inventory.sh via bin/md2html.sh -->
<!-- START: this line starts content generated by: bin/output-index-inventory.sh -->
<h1 id="inventory-for-2005anon"><a name="inventory"></a>Inventory for
2005/anon</h1>
<h1 id="inventory-for-2005anon"><a name="inventory"></a>Inventory for 2005/anon</h1>
<p><a href="#top">Jump to top</a></p>
<h2 id="primary-files">Primary files</h2>
<ul>
<li><a
href="https://github.com/ioccc-src/temp-test-ioccc/blob/master/2005/anon/anon.c">anon.c</a>
- entry source</li>
<li><a
href="https://github.com/ioccc-src/temp-test-ioccc/blob/master/2005/anon/Makefile">Makefile</a>
- entry Makefile</li>
<li><a
href="https://github.com/ioccc-src/temp-test-ioccc/blob/master/2005/anon/anon.alt.c">anon.alt.c</a>
- alternative source</li>
<li><a
href="https://github.com/ioccc-src/temp-test-ioccc/blob/master/2005/anon/anon.orig.c">anon.orig.c</a>
- original source</li>
<li><a href="2005_anon.tar.bz2">2005_anon.tar.bz2</a> - download entry
source</li>
<li><a href="https://github.com/ioccc-src/temp-test-ioccc/blob/master/2005/anon/anon.c">anon.c</a> - entry source</li>
<li><a href="https://github.com/ioccc-src/temp-test-ioccc/blob/master/2005/anon/Makefile">Makefile</a> - entry Makefile</li>
<li><a href="https://github.com/ioccc-src/temp-test-ioccc/blob/master/2005/anon/anon.alt.c">anon.alt.c</a> - alternative source</li>
<li><a href="https://github.com/ioccc-src/temp-test-ioccc/blob/master/2005/anon/anon.orig.c">anon.orig.c</a> - original source</li>
<li><a href="2005_anon.tar.bz2">2005_anon.tar.bz2</a> - download entry source</li>
</ul>
<h2 id="secondary-files">Secondary files</h2>
<ul>
<li><a href="index.html">index.html</a> - entry home page</li>
<li><a
href="https://github.com/ioccc-src/temp-test-ioccc/blob/master/2005/anon/.entry.json">.entry.json</a>
- description of entry in JSON</li>
<li><a
href="https://github.com/ioccc-src/temp-test-ioccc/blob/master/2005/anon/.gitignore">.gitignore</a>
- list of files that should not be committed under git</li>
<li><a href="https://github.com/ioccc-src/temp-test-ioccc/blob/master/2005/anon/.entry.json">.entry.json</a> - description of entry in JSON</li>
<li><a href="https://github.com/ioccc-src/temp-test-ioccc/blob/master/2005/anon/.gitignore">.gitignore</a> - list of files that should not be committed under git</li>
<li><a href=".path">.path</a> - directory path from top directory</li>
<li><a
href="https://github.com/ioccc-src/temp-test-ioccc/blob/master/2005/anon/README.md">README.md</a>
- markdown source for index.html</li>
<li><a href="https://github.com/ioccc-src/temp-test-ioccc/blob/master/2005/anon/README.md">README.md</a> - markdown source for index.html</li>
</ul>
<!-- END: next line ends content generated by: bin/output-index-inventory.sh -->