search of the:   for: 
code logs -> 2008 -> Thu, 17 Jan 2008< code.20080116.log - code.20080118.log >
--- Log opened Thu Jan 17 00:00:47 2008
00:03
< McMartin>
My labmate's dissertation skeletal draft includes this
00:03
< McMartin>
"CHAPTER 2: ANALYSIS OF STATIC FIELDS"
00:03
< McMartin>
"In which our brave warrior notes that meadows tend to remain in place"
00:07
< gnolam>
:D
00:08
< Reiver>
...
00:10
< McMartin>
Having forgotten which languages Reiver knows, I might as well point out that a "static field" is the OO equivalent of a global variable.
00:10
< Reiver>
And meadows?
00:11
< Reiver>
Or is he really meaning, y'know, grassy green things?
00:11
< McMartin>
That's a pun on "field".
00:11
< McMartin>
This is the "I'm making sure that LaTeX is correctly formatting my Table of Contents and chapter headings and stuff."
00:12
< McMartin>
It will, alas, not survive to the final draft, though I'm pushing for him to keep the "In which..." construct as chapter subtitles.
00:29
<@TheWatcher>
McMartin: http://fleet.starforge.co.uk/assertgtk.tar.bz2 - there you go
00:29
<@TheWatcher>
tell me if that does remotely what you're after
00:30 * McMartin grabs, takes a look
00:30
< McMartin>
I won't be folding it into the codebase any time soon, mind
00:30 * McMartin is in the process of doing inconceivable violence to the way content is organized.
00:31
<@TheWatcher>
Indeed, but I'll be using it in mine so *shrug* :)
00:32
<@TheWatcher>
(also, I haven't tried compiling it on windows, so you may or may not run into Issues there)
00:32
< McMartin>
(It wouldn't compile in Windows)
00:32
< McMartin>
This is nice-looking
00:32
< McMartin>
I bet this can be tied into our log system.
00:33
< Reiver>
...
00:34
<@TheWatcher>
Hm, I'll look into windows tomorrow probably, 0:30 is no time to fight mingw.
00:34
< Reiver>
Oh my lordie, McM. It's not a pun, it's a bad pun.
00:34 * Reiver was overthinking it~
00:34
< McMartin>
TW: You need an #include <stdio.h> in myassert.c, idly.
00:34
<@TheWatcher>
... guh, indeed.
00:34
<@TheWatcher>
(see also, 0:30 ¬¬)
00:35
< McMartin>
Heh
00:35
< McMartin>
Get some sleep
00:35
<@TheWatcher>
Not a bad idea, that.
00:35
< McMartin>
That said, this is lovely
00:38
<@TheWatcher>
I'll try to do some more digging on how to drag a stack track out of windows, too. I just have a terrible dread it involves vast amounts of code
00:41
< McMartin>
Heh
00:41 * McMartin will be cutting that bit out, because we only do this on fatals.
00:41
< McMartin>
And because we're comically multithreaded
00:41
< McMartin>
Though less comically so these days.
00:42 You're now known as TheWatcher[T-2]
00:44 You're now known as TheWatcher[zZzZ]
01:12 Derakon[AFK] is now known as Derakon
01:20 Vornicus is now known as Vornicus-Latens
01:30 Derakon is now known as Derakon[AFK]
01:40 gnolam [lenin@Nightstar-10613.8.5.253.static.se.wasadata.net] has quit [Quit: Z? (yeah, right)]
02:23 Derakon[AFK] is now known as Deraekon
02:23 Deraekon is now known as Derakon
03:42 Raif [~corvusign@67.161.90.ns-4200] has joined #Code
05:05 Ratfor [~Ratfor@Nightstar-1310.ptr.terago.ca] has joined #Code
05:05 Ratfor [~Ratfor@Nightstar-1310.ptr.terago.ca] has left #Code []
06:45 Derakon is now known as Derakon[AFK]
07:21 Pete is now known as AnnoDomini
07:30 AnnoDomini [AnnoDomini@Nightstar-29597.neoplus.adsl.tpnet.pl] has quit [Ping Timeout]
07:36 AnnoDomini [AnnoDomini@83.21.41.ns-26762] has joined #Code
07:36 mode/#code [+o AnnoDomini] by ChanServ
08:34 Raif [~corvusign@67.161.90.ns-4200] has quit [Quit: A trebuchet is not a marital aid. Many bothans died to bring us this information.]
08:39 You're now known as TheWatcher
10:34 Vornicus-Latens is now known as Vornicus
11:00 * Vornicus <3 dynamic typing and first-class functions.
11:11 * TheWatcher sets about getting a win32 gtk+2 development environment set up, haates
12:41 * Vornicus does battle with some really, really nasty set arithmetic.
13:56 gnolam [lenin@85.8.5.ns-20483] has joined #Code
13:56 mode/#code [+o gnolam] by ChanServ
14:09
< Vornicus>
oh god it's so horrible
14:09
<@TheWatcher>
Mrr?
14:09
<@TheWatcher>
That's no way to talk about gnolam ;P
14:10
<@AnnoDomini>
In Soviet Russia, gnolam talks about YOU.
14:10
< Vornicus>
This set stuff.
14:11
< Vornicus>
I think this is the first time in like three years I've used continue
14:17
< Vornicus>
goes something like this: get all the regions next to a region with a red token (a set union - for each item in the list of red tokens, index into a list of sets, and add that to the pile), then find red tokens in that list (intersect the reds with the adjacents). If there aren't any, then we're done.
14:19
< Vornicus>
But if there /are/ (and most likely there are), pick one of the reds, check to see if there's a place it can move to - which is to say, take the /rest/ of the reds, get their adjacents, and subtract both the other reds and the adjacentsfrom the whole list, and if it's not empty, then it has a place to move to...
14:20
< Vornicus>
and then swap them, and start from the beginning again.
14:21
< Vornicus>
But if it doesn't have a place to move to, try another one.
15:02
< Vornicus>
okay. Move the bit to find all adjacencies into a function. That makes the question a loop condition instead of something to break on.
15:08
< Vornicus>
...except that that's still not true - I need to create the red indices inside the loop and the condition needs those
15:57
< Vornicus>
...unless I construct the pile of red indices before entering the loop, and /manipulate/ it when I do swapping.
15:58
< Vornicus>
Which makes a vague amount of sense.
15:59 * TheWatcher sighs, notes that setting up a gtk+2 compile environment on windows /should not be this much work, damnit/
16:27
<@TheWatcher>
Finally
16:30
<@ToxicFrog>
As in, building gtk+2?
16:30 * ToxicFrog justs uses headers+binaries; building it from source is too much like work.
16:34
<@TheWatcher>
McMartin: http://fleet.starforge.co.uk/assertgtk.tar.bz2 - should now compile on linux or windows (using mingw anyway), includes dev-c++ project and linux makefile for the example
16:35
<@TheWatcher>
Now to work out how to get some kind of stack trace in win32...
16:51 * Vornicus continues to do battle with this evil awful set-graph-etc thing.
17:00
< jerith>
So, I have a coworker with the surname Brown. He wants very much to name his about-to-pop daughter Octothorpe.
17:00
< jerith>
His wife won't let him.
17:01
<@TheWatcher>
Bah, that's unfair.
17:01
< Vornicus>
heh
17:03
<@TheWatcher>
Maybe they should spend more time talking over it, I'm sure they could hash out a compromise.
17:03 * TheWatcher flrrrrd
17:04
< Vornicus>
:(
17:04 You're now known as TheWatcher[afk]
17:05
< jerith>
Bonus points to to TheWatcher. The pun there is on Hash Brown.
17:06 * Vornicus didn't even put that together untile TW got it.
17:32 Forj [~Forj@203.211.126.ns-3357] has joined #code
17:32 mode/#code [+o Forj] by ChanServ
17:34 Forj [~Forj@203.211.126.ns-3357] has quit [Quit: Gone]
18:35 You're now known as TheWatcher
18:49
<@gnolam>
TheWatcher: "should" is the word to watch out for there.
18:50
<@TheWatcher>
gnolam: well, quite
18:50
<@gnolam>
Generally speaking, when something is claimed to be buildable from source on both Linux and Windows they mean Linux/GCC and Windows/Some-specific-version-of-MSVC.
18:50
<@TheWatcher>
hence why I qualified mingw
18:51
<@TheWatcher>
whether it compiles in anything else is, frankly, SEP.
18:53
<@TheWatcher>
it's hardly a complete project, it's somehting I wrote in a couple of hours last night, mostly
18:55
<@gnolam>
Even things that claim to be buildable with MinGW tend to fall into that category. :P
18:56 * gnolam has run into about 1000 too many projects to ever trust a description that it should be compileable.
18:56
< McMartin>
UQM actually is, since that's how we get our Win32 SVN snapshots~
18:57
< McMartin>
That said, "SEP"?
18:58
<@TheWatcher>
Somebody Else's Problem
18:58
< McMartin>
Aha.
19:02
<@TheWatcher>
(not to say I wouldn't help try to get it working, but right now It Workd For Me is enough ¬¬)
19:02
<@TheWatcher>
*works
19:03 Vornicus [~vorn@Admin.Nightstar.Net] has quit [Ping Timeout]
21:40 Vornotron [~vorn@Admin.Nightstar.Net] has joined #code
21:41 Vornotron is now known as Vornicus
21:56 AbuDhabi [AnnoDomini@83.21.18.ns-4628] has joined #Code
21:56 mode/#code [+o AbuDhabi] by ChanServ
21:56 AnnoDomini [AnnoDomini@83.21.41.ns-26762] has quit [Ping Timeout]
22:13 * Vornicus returns from registering for classes. Now once again battles set-graph madness
23:00 You're now known as TheWatcher[T-2]
23:05 You're now known as TheWatcher[zZzZ]
23:08 * ToxicFrog resurrects an old project
23:11
<@ToxicFrog>
game { title, publisher, developer, platform, completion-status, tags }
23:11
<@ToxicFrog>
Sane?
23:12
< Vornicus>
Sane.
23:18 gnolam [lenin@85.8.5.ns-20483] has quit [Quit: Z?]
23:24
<@ToxicFrog>
Oh, that's a bug in string.split!
23:25
< Vornicus>
heh
23:29 * ToxicFrog rewrites it in terms of hideous string.gsub abuse
23:31
<@ToxicFrog>
Mmm, mutable state is delicious~
23:37
<@ToxicFrog>
#%&(@#!@)*%#@*%@!^#$#&%@ !#(!_%*@()$%&@^%
23:37
<@ToxicFrog>
I have a language design clawing at my mind
23:38
< Vornicus>
Heh
23:38
<@ToxicFrog>
It's such an alluring idea, too.
23:38
<@ToxicFrog>
Lua isn't powerful enough.
23:39
<@ToxicFrog>
Scheme isn't friendly enough.
23:39
<@ToxicFrog>
I must breed them
23:43 * Vornicus gives TF some lightning in the background.
23:45
<@ToxicFrog>
Of course, the problem is that the main part of the power I want from Scheme is the ability to create new keywords/operators at will, and the macro system.
23:45
<@ToxicFrog>
Both of which stem from the fact that the Scheme grammar basically doesn't distinguish between function calls, keywords, and operators.
23:45
<@ToxicFrog>
Which ruins a lot of the friendliness.
--- Log closed Fri Jan 18 00:00:54 2008
code logs -> 2008 -> Thu, 17 Jan 2008< code.20080116.log - code.20080118.log >
 search of the:   for: