Count on crashes

like there’s ever going to be wrap around with a 32-bit
unsigned int! how long would that take? Let’s see... even if
we somehow generated 100 packets per second, we’d need to
run for over 1 year, 4 months straight. I suspect this
client, or the machine it’s running on, will crash way way
way before then. Ah, windows. Being able to COUNT on crashes
is SUCH a lovely thing. F*ck, whatever.

Fight Club

 * The first rule of fight club is:
 *   You do not include ssl/ssl.h.
 * The second rule of fight club is:
 *   You DO NOT include ssl/ssl.h.
 * The third rule of fight club is:
 *   If you include util/error.h before ssl/ssl.h,
 *   the compilation is over!
 * The fourth rule of fight club is:
 *   Only one inclusion of ssl/ssl.h at once.
 * The fifth rule of fight club is:
 *   If you include ssl/ssl.h, it comes right after elinks.h.
 * The sixth rule of fight club is:
 *   No ssl/ssl.h in other .h file, no complaints about
 *   the rules.
 * The seventh rule of fight club is:
 *   The compilation will go as long as there is still
 *   something to compile.
 * The eight, and final rule of fight club is:
 *   If this your first peek at this .h file, you
 *   (Apologies to Chuck Palahniuk.) */

Steal my girlfriend

* to Bill Robbins for trying to steal my girlfriend. Not only
* did you show me no respect but you’re a manipulating
* prick who tried to take away the most important thing in
* the world to me with no guilt whatsoever, and for that I
* wish you nothing but pain. Die.

Lazy MSIE 6 developers

For some stupid reason that i don’t want to know, 
those alerts are ALWAYS returning the same shit under 
that shitty browser (Microsoft Internet Explorer 6), 
when i’m using element.getAttribute(‘foo’) i want to 
_GET_ THE F*CKING ‘foo’ _ATTRIBUTE_ of element NOT A 
STUPID CHILD named ‘foo’.
Thanks again lazy MSIE 6 developers!

Work guarantee

/* Yeah, yeah. Cheap hack, but it’s guarenteed to work! :)
 * The following just hacks the returned track name into
 * the variable

You’ve been warned

######## No User serviceable part below ! ########
# You should really not look any further:
# The following is private unexported undocumented 
# unblessed... code time to hit "q" ;-) !

# Hmmm... ok, you really want to know ?

# You’ve been warned… Here it is...

# Array storing the parsed descriptions
variable OptDesc;
array set OptDesc {};
# Next potentially free key id (numeric)
variable OptDescN 0;

# Inside algorithm/mechanism description:
# (not for the faint hearted ;-)

I have NO idea how this happened

/* what the f*ck... this happened once when the source was
 * somehow added twice... i have NO idea how this happened,
 * but giFT shouldn’t be crashing because of it */

A security standpoint

 * Map other pairs to security names
 * here. For now just let non-IPv4 transport always succeed.
 * WHAAAATTTT. No, we don’t let non-IPv4 transports
 * succeed! You must fix this to make it usable, sorry.
 * From a security standpoint this is insane. — Wes

Sorry to the developers

# don’t blame me if this is wrong! It took me quite a while
# to find in st**id tcsh docs *grrr*
# (insert “sorry-to-the-developers” here).

A rewrite from scratch

Originally this was a hacked version of shout.
Then, I wrote it in perl, the music kept breaking up on pauses
Then, I wrote it in C, the music kept breaking up on pauses
Then, I wrote another one in C using shared memory between two processes.
It worked.
Then, I discovered what happens when you run out of shared memory keys
Then, I ran fsck and pressed ‘y’ lots.
Started again in C, same idea.
Shared memory became unwieldy.
This is a rewrite from scratch using pipes between processes.
I am pissed off.
Page 3 of 8« First...2345...Last »