00:07:07 back for a sec 00:07:15 http://rafb.net/p/wYc8k789.html <-- from rc/funge's main.c 00:07:24 proof that Riley is a bad programmer as well 00:07:33 (learn to use "continue" in C -_-) 00:08:17 wait... is that removing the program name from argv by comparing it against the name from a Version Management System? 00:08:20 and he drops the max index he found when parsing arguments 00:08:36 ais523, I don't know 00:08:38 also, that strcpy without allocating memory is suspicious 00:08:41 there are NO comments 00:08:49 and yes 00:08:51 it is 00:08:58 anyway, time to go home I think 00:08:59 bye 00:09:03 ais523, cfunge uses whatever getopt says is the last option 00:09:06 then strdups that 00:09:07 Additionally, why the fuck does he have argv and ArgV? 00:09:29 and you don't want to see the compare stuff above 00:09:34 he manages to crash on --help 00:09:37 no clue how 00:09:41 *what?!?* 00:09:41 bye everyone 00:09:42 also -YS != -Y -S 00:09:44 um 00:09:48 what is wrong with that code... 00:09:51 :| 00:09:54 pikhq, I think -YS cause memory corruption in rc/funge 00:09:59 it seems so 00:10:01 well, it's hard to figure out what without any context, of course 00:10:08 I think calling him a bad programmer is a bit rude. 00:10:14 *facepalm* 00:10:26 ehird, he trows away an index from command line parsing 3 lines above 00:10:33 and then rechecks that index 00:10:43 fine, fine, but i still think it was rude to say that 00:11:16 oh --help works fine if followed by a program 00:11:22 except... it doubles output 00:11:25 GGOODD 00:11:25 rc/funge ran fungot, didn't it? 00:11:25 ehird: but mine is simpler, one moment. i still have doubts concerning the exact implications :) fnord/ fnord/ images/ fnord/ fnord 00:11:31 he's obviously not a terrible programmer 00:11:35 ehird, yes it did, but older verion 00:11:36 if he can get a befunge-98 system to work. 00:11:37 version* 00:11:48 true. but he can't get command line parsing to work 00:11:59 And command line parsing is easy... 00:12:02 maybe he has some stupid, obscure, and crap parts of code, but that doesn't mean he can't program 00:12:05 Especially with getopt (GNU or otherwise) 00:12:10 $ ./funge -S ~/src/cfunge/trunk/mycology/sanity.bf 00:12:10 0 1 2 3 4 5 6 7 8 9 Exiting with return code = 0 00:12:13 $ ./funge -S --help ~/src/cfunge/trunk/mycology/sanity.bf 00:12:13 0 0 1 1 2 2 3 3 4 4 5 5 6 6 7 7 8 8 9 9 Exiting with return code = 0 00:12:20 pikhq, he doesn't *use* that 00:12:25 he got a huge loop with strcmp 00:12:31 Who *doesn't*? 00:12:34 $ ./funge --help 00:12:34 Segmentation fault 00:12:40 I'd like to point out that command-line parsing was probably not the top thing on his mind. 00:12:45 pikhq, Mike Riley, the author of this program 00:12:47 It was likely an afterthought. 00:12:57 ehird: Command line parsing takes all of half an hour to do right. 00:13:00 ehird, well even so, crashing on it is messy 00:13:10 pikhq: I imagine he hacked something up in 5 minutes before releasing it. 00:13:20 pikhq, no it was much faster, at least using getopt(), but even without it, it takes 1 minute 00:13:23 err 00:13:26 10 minutes 00:13:27 or so 00:13:27 Indicative of a poor programmer. 00:13:35 GOODG:O ODD :p uDs hpeuss h500 e 00:13:35 s 500 00:13:35 GOODG:O OMD :p uMs hpeuss h1000 e 00:13:35 s 1000 00:13:47 pikhq: Where is your Befunge-98 system that can run an entire IRC bot using a bunch of fingerprints and not crash? 00:13:48 I disagree. That isn't GOOD 00:15:00 one thing to note. is that while the v1 segfaults on it, it don't cause memory corruption on --YS 00:15:25 it just ignores that argument 00:15:27 -!- KingOfKarlsruhe has quit (Remote closed the connection). 00:15:32 however --help still crashes it 00:15:34 badly 00:16:06 /home/arvid/funges/interpreters/rcfunge/mterm.c:23: warning: the `gets' function is dangerous and should not be used. 00:16:14 I think that proves "bad programmer" 00:16:16 without doubt 00:16:30 -!- optbot has set topic: the entire backlog of #esoteric: http://tunes.org/~nef/logs/esoteric | it's kinda trivial.. 00:16:32 a bad programmer would not be able to write rc/funge, stop being jerks 00:16:48 ah 00:16:51 #1 0x000000000040a2d4 in main (argc=2, argv=0x7fff674ca9d8, envp=0x7fff674ca9f0) at main.c:1334 00:16:51 1334 if (strcmp(argv[i],ProgName) == 0) f = 1; 00:16:55 (gdb) print ProgName 00:16:55 $1 = 0x0 00:16:57 (gdb) print argv[i] 00:16:57 $2 = 0x7fff674cb91a "--help" 00:17:06 that explains that bit 00:17:16 ehird, right, ehehehehe,,,,,,,,,,,,,,,, 00:17:33 Let me know when you're not being insensitive assholes to Mike Riley. 00:17:34 -!- ehird has left (?). 00:18:03 oh poor fanboy 00:18:06 heh 00:18:26 hrrm. I wonder why he loves Mike Riley so much :/ 00:18:30 seems strange 00:18:37 the football coach? 00:18:37 considering he made that MKRY himself 00:18:43 mbishop, who? 00:18:52 http://www.rcfunge98.com/ 00:18:54 that one 00:19:00 Author is called Mike Riley 00:19:11 possibly someone else got same name 00:20:00 http://en.wikipedia.org/wiki/Mike_Riley 00:20:30 mbishop, don't think it is the same person 00:20:52 mbishop, nor any of those on http://en.wikipedia.org/wiki/Mike_Riley_(disambiguation) 00:22:42 What does the Rc stand for? 00:22:50 reminds me of those rosetta code implementations 00:26:08 -!- ehird has joined. 00:26:18 AnMaster: I am totally a fanboy because I point out you're being a fuckwit by making personal attacks. Totally. 00:26:19 -!- ehird has left (?). 00:26:30 I didn't make any personal attack 00:26:52 not more than MKRY :/ 00:27:10 -!- ehird has joined. 00:27:17 and I'm going to bed 00:27:19 night all 00:27:22 You repeatedly called him a bad programmer. MKRY was just a light-hearted prod at his english style. 00:27:23 -!- ehird has left (?). 00:28:40 -!- jix has quit ("CommandQ"). 00:51:40 -!- psygnisfive has joined. 01:41:02 -!- poiuy_qwert has joined. 03:54:15 -!- Corun_ has quit (Remote closed the connection). 03:54:29 -!- Corun has joined. 04:10:36 -!- Corun has quit ("Pants."). 04:43:19 -!- ab5tract has joined. 05:06:20 My home page's color scheme is now automatically generated 8-D 05:10:40 boomtown 05:11:01 I see. 05:13:16 can you link me to that neural net again plz 05:13:42 http://codu.org/colormatch/ 05:13:52 Also see http://codu.org/ 's fancy colors ^^ 05:14:45 aweseome 05:14:51 Interesting. 05:15:47 GregorR next step a restful api for grok'ng the "present scheme" 05:16:22 It's actually getting not all that bad results... 05:16:48 I haven't seen any bad results. Some are a bit ... bright, but other than that :P 05:17:18 oi 05:17:31 is anyone here good with music? like, recognizing pieces, styles, etc? 05:19:37 I have friends who are :P 05:20:30 psygnisfive: i am 05:20:59 lament, can you do me a favor and listen to a brief bit of piano music and see if you recognize it? 05:21:08 ok 05:21:08 I'm vaguely so. 05:22:52 ok lemme record it off into an mp3 for you 05:26:45 erk 05:26:48 its gonna have to be wav 05:26:52 since i dont have flac 05:26:56 er 05:26:58 not flac 05:26:59 lame 05:29:47 this is a horribly large file for nary a minutes worth of audio, sorry: http://www.wellnowwhat.net/transfers/dennett_piano.wav 05:33:47 psygnisfive: i don't think it's anything famous 05:33:58 i dont either 05:33:59 but i WANT it 05:34:04 if you could give me like 05:34:12 note directions 05:34:18 uhhhh. 05:34:19 e.g up down same same up etc 05:34:28 Generated color scheme: #2E0A78, #85D70A, #F2CEBA is not particularly nice :) 05:34:41 i have music recognition resources that rely on note change directions 05:34:41 i don't think you'd find this piece anywhere. 05:34:43 but it's mostly pretty nice 05:34:50 oh no i think this one place would have it 05:35:01 its pretty good at recognizing random music 05:35:34 i'm not sure it was even ever published 05:35:44 it had to have been 05:35:57 all the music in this show was retrieved from free music archives 05:36:01 it's probably a study 05:36:14 so somehow they found a recording 05:36:38 do you atleast have an idea of what genre, or composer? 05:36:43 it could have been a czerny study 05:37:04 genre? 05:37:28 genre: romantic, or imitation romantic (if it's recent) 05:37:30 Asztal: I've seen worse :P 05:37:41 really? 05:37:42 hm 05:37:48 shostakovich was romantic, wasnt he? 05:37:57 Erm, romantic ERA ... 05:38:05 Romantic era != romantic necessarily :P 05:38:06 i know :P 05:38:15 psygnisfive: shostakovich was very very late romantic 05:38:20 ok 05:38:30 i heard a shostakovich piece, ages ago, that reminded me of this 05:38:30 this is early, "normal" romantic 05:38:50 but then i looked up shostakovich and couldnt find anything 05:39:22 shostakovich is fucking amazing 05:39:33 this music is complete junk in comparison :) 05:39:40 but i think it's a study, not a real piece 05:40:33 it's pretty easy to play, you could learn it 05:40:57 i cant follow the hands 05:41:03 i dont know where the low notes go :( 05:41:09 i love this piece tho 05:41:16 its simple and clean and beautiufl 05:41:32 and embodies a single, pure concept for me 05:41:40 i think it might be czerny 05:41:48 ok 05:41:50 thank you 05:41:50 <# 05:41:51 <3 05:42:04 if it is, ill give you head whenever you want. :D 05:42:05 :P 05:42:06 but that's an unlikely guess, just the best match i can think of 05:42:09 awesome 05:42:20 this is like a free lottery ticket :) 05:47:42 ok, maybe not czerny 05:48:05 ?? 05:48:14 im listening to some of czerny and it sounds like it MIGHT be 05:48:29 why not czerny? 05:48:56 at least it's not in any czerny i can find online 05:51:12 -!- poiuy_qwert has quit. 05:55:25 but... listen to shostakovich instead :P 05:58:34 czerny has some pieces that could be it 05:58:39 lots of them are like 10 minutes long 05:58:44 and itunes only has brief clips so 05:59:04 lots of people suggested it might be glass's etude 1 or so, but its not 05:59:08 its similar tho 06:00:17 its very glassy in its extreme use of high arpeggios and stranded low notes 06:10:01 -!- moozilla has quit (Nick collision from services.). 06:10:02 -!- metazilla has joined. 06:13:09 -!- metazilla has quit (kornbluth.freenode.net irc.freenode.net). 06:14:42 -!- metazilla has joined. 06:16:30 -!- optbot has set topic: the entire backlog of #esoteric: http://tunes.org/~nef/logs/esoteric | and stuff. 06:39:14 ize finding nothing :( 06:41:49 -!- metazilla has quit (kornbluth.freenode.net irc.freenode.net). 06:42:18 -!- metazilla has joined. 06:49:24 -!- metazilla has quit (kornbluth.freenode.net irc.freenode.net). 06:49:44 -!- metazilla has joined. 07:06:29 http://rafb.net/p/wYc8k789.html <<< if you find this hard to read, you're bad programmers 07:06:48 but it's clearly obfuscated 07:14:26 -!- metazilla has quit (Read error: 54 (Connection reset by peer)). 07:26:22 (on second read i retract the obfuscation statement) 07:31:53 also doesn't ehird mock people all the time? 07:32:20 what? 07:32:57 he /parted dramatically during my night when AnMaster was putting down Miker Riley's code 07:33:19 http://rafb.net/p/wYc8k789.html 07:33:21 this code 07:33:24 well 07:33:33 not mike's code, but mike's ability to code 07:35:06 in my opinion the non crappy way to do that is to implement some sort of gc on top of c, write a list class and do a cut-in-half on the index of the program's name 07:35:24 but in actual c, i wouldn't say that's a bad way to do it 07:44:44 Oh, more IRC-drama. 07:47:07 drrrrrrrrrrammmmmmmmmmmm 07:50:18 Well, the use of a fixed-width ArgV[50][80] and strcpy instead of an array of pointers and strdup or something (or just those original pointers) *is* reasonably ugly, but other than that I'm not sure what's so very wrong in the loop, or how it would be significantly improved with "continue", except that it'd remove one level of nesting. 07:50:39 Indentation, I mean. 07:51:22 ^cho drama 07:51:23 dramaramaamamaa 07:51:36 That's a reasonably ^cho-able word. 07:59:59 -!- clog has quit (ended). 08:00:00 -!- clog has joined. 08:00:39 -!- ab5tract has quit. 08:11:18 -!- immibis has joined. 08:16:53 -!- Judofyr has quit. 08:41:04 ^cho llama 08:41:04 llamalamaamamaa 08:44:16 -!- oerjan has joined. 09:09:55 -!- olsner has quit ("Leaving"). 09:14:53 -!- immibis has quit (Read error: 60 (Operation timed out)). 09:21:15 fizzie, it was wrong because he knew this index 3 lines above when he parsed argumnets 09:21:21 arguments* 09:21:27 and then he threw it away 09:21:50 also it is the reason it crashes on "any option except -h and no program name" 09:22:15 http://www.club.cc.cmu.edu/~ajo/funge/#Fungus <-- this looks fun 09:27:55 * AnMaster yawns 09:30:54 AnMaster: you've got something stuck between your teeth 09:31:16 * AnMaster walks over to a mirror 09:31:19 ah yes 09:31:30 fixed 09:34:40 ^cho esoteric 09:34:40 esotericsotericoterictericericricicc 09:34:55 ^cho ````ccc 09:34:55 ````ccc```ccc``ccc`ccccccccc 09:35:08 subtle cough is so pretty 09:35:57 ^ul (:(oklopol)^):^ 09:35:57 ...bad insn! 09:36:02 oh right 09:36:04 ^ul (:(oklopol)):^ 09:36:10 i wanted to make REALLY INTENSE COUGH at some point, but couldn't really find a pure and beautiful way to add another continuation-related operation that made it tc 09:36:17 ^ul (:^(oklopol)):^ 09:36:17 ...too much prog! 09:36:20 wtf 09:36:22 that is a new one? 09:36:27 No. 09:36:28 err 09:36:34 you're not doing tail recurzion 09:36:37 "too much stack" I have seen 09:36:41 oklopol, true 09:36:45 I wanted it to grow 09:36:47 you call the program recursively 09:36:47 I think ais523 or someone did "too much prog" already, when I asked. 09:36:53 so the program stack grows 09:36:55 yes 09:37:05 ^ul (::^):^ 09:37:06 It's printed when the program length extends too much to the left. 09:37:06 ...too much stack! 09:37:22 ^ul (:^):^ 09:37:23 ...out of time! 09:37:29 AnMaster: different stack, that's the data stack, i mean the "call stack", kinda 09:37:34 ah right 09:37:37 makes sense 09:37:57 although in underload it's not really a stack, but you know, it does what "the stack" usually does, holds things you'll need when returning from procedures 09:38:16 The ^ instruction is implemented by simply inserting the top-of-stack so that it's final character replaces the ^, and then moving the IP to the start of that string. 09:38:32 yes, but that's such an ugly practical way to explain it. 09:40:55 you need to use high-level concepts that only somewhat apply, and use technical terms like "kinda" 09:41:04 hah 09:41:36 try { socket = new Socket( ia, port ); } catch( Exception e ) { System.out.println( e ); } 09:41:38 try { writer = new BufferedWriter( new OutputStreamWriter( socket.getOutputStream() ) ); } catch ( Exception e ) { System.out.println( e ); } 09:41:41 try { reader = new BufferedReader( new InputStreamReader( socket.getInputStream() ) ); } catch ( Exception e ) { System.out.println( e ); } 09:41:44 try { writer.write( "JOIN " + name + "\n" ); } catch ( Exception e ) { System.out.println( e ); } 09:41:47 try { writer.flush(); } catch ( Exception e ) { System.out.println( e ); } 09:41:48 i make java pretty :) 09:42:42 i find exception handling bad programming style, should really wrap up the socket class into one that suppresses exceptions, or just crashes the whole program 09:43:53 Is there a particular reason that's not all in a single try { } block, except for the fact that now you can get some NullPointerExceptions from writer.write() and such? 09:44:45 after i stop doing try-catching separately for each line, i will have to start thinking what the optimal way to do that is 09:45:02 haha 09:45:04 because i do not want to actually catch these exceptions, these blocks do not actually make any sense, so i don't want to have to think. 09:45:20 So it's either butt-ugly or most optimal, nothing in the middle? 09:45:27 they are there because java wants me to be clean. 09:45:34 That's not clean. 09:45:40 fizzie: it needs to be perfect :) 09:45:41 oklopol, a solution could be using some other language 09:45:52 fizzie: clean as in, handle exceptions 09:45:57 I wouldn't call that perfect, either. 09:46:08 A single catch block for all lines would be as "clean". 09:46:24 fizzie: syntactically cleaner, conceptually just the same 09:46:41 um no, for the reason that you continue if you fail 09:46:48 when that makes no sense 09:47:59 java doesn't let me make socket suppress easily, and it doesn't let me not catch these, so i find it conceptually cleanest to surround everything with try-catches that can throw. i could, in this case, surround these with just one block 09:48:13 but if i added something in the middle that actually threw an exception i want to handle 09:48:29 then i would have to write the exception blocks around each other 09:48:41 heh 09:48:50 and the exception handling for the socket would clutter my thinking 09:49:01 oh well 09:49:04 Personally I find it pretty ugly to do all the later lines when you already know they're just going to throw NullPointerExceptions when they refer to uninitialized cruft. 09:49:44 fizzie: that's thinking. you clearly don't see my point here, think i'm actually being rational? :P 09:49:55 thinking is not good. you shouldn't think about things that matter not. 09:49:57 I don't expect rationality from you, no. 09:50:15 Maybe you should go the other way and add "throws Throwable" to all your functions. 09:50:23 It looks silly, also. 09:50:28 :D 09:51:05 nah, that's too much work 09:52:02 * oerjan suddenly gets "throws Smurfable" into his head 09:52:11 You can run all your Java code through CPP and use #define BEGIN throws Throwable {, then code your functions as void (foo) BEGIN heh(); heh(); } 09:52:26 :D 09:52:48 then i'll need #define END } 09:52:58 but yeah, otherwise not such a bad idea 09:53:19 Maybe not; the mismatch between BEGIN and } can be an artistic statement on the class imbalance between the rich and the poor, or something. 09:53:35 -!- M0ny has joined. 09:53:48 anyway it's not like i hate the fact java wants me to catch all exceptions, it's just when i'm writing the code in the first place, thinking about exceptions is so goddamn annoying i want to choke my computer. 09:54:06 fizzie: i don't like art :D 09:54:42 plop 09:55:41 well, i guess what i want is #ifdef CLEANMODE \n #define BEGIN { \n #else #define BEGIN throws Throwable { \n #endif 09:56:59 but if i had to choose just one of those, it would definitely not be cleanmode, because i simply don't want to care about exceptions when writing the code, that's just clutter. 09:57:48 but, i guess i do like hating java, so it's not that bad. 09:58:11 some things i don't even like hating, they're the ones that really annoy me 09:58:16 M0ny: plop. 09:58:23 ^^ 09:58:30 i know a guy named plop 09:59:27 plopmania, actually, he was here once, asked me how to run bootstrap_ on my bot, ran a hello world and left :P 09:59:40 :p 09:59:59 hit and run programming 10:00:58 hm... 10:03:39 hm... 10:03:41 ^bf --[>-<-------]>++.----.+.+++.+.-.---. 10:03:41 plmpqpm 10:03:44 argh 10:03:57 ^bf --[>-<-------]>+.----.+.+++.+.-.---. 10:03:57 oklopol 10:04:05 yeah i'm bf-licious 10:04:34 not bad. 10:05:09 a bf/char ratio of 5.14 10:06:31 thutubot might be easy too 10:07:17 well the t and u 10:07:45 also opqrst isn't that much, but yeah i guess there are others 10:09:01 ^bf ++++++++++[>++++++++++<-]>.--.+. 10:09:01 dbc 10:09:14 that's probably far from an optimal 100 10:10:02 it may be optimal without wrapping 10:10:14 (best on the wiki) 10:10:27 ^bf --[>-<+++++]>--.--.+. 10:10:27 dbc 10:13:13 ^bf +++[>+++[>+++++++++++<-]<-]>.--.+. 10:13:40 ^bf +++[>+++[>+++++++++++<-]<-]>>++++.--.+. 10:13:40 gef 10:13:45 ^bf +++[>+++[>+++++++++++<-]<-]>>+.--.+. 10:13:45 dbc 10:13:47 :) 10:14:09 a few chars longer 10:14:31 >_O 10:14:44 >_O? 10:15:02 oerjan, what does that one mean? 10:15:43 fizzie, also what do you think about this: http://www.club.cc.cmu.edu/~ajo/funge/#Fungus 10:16:02 ^bf +++++[>+++[>+++[>++<-]>+<<-]<-]>>>+++++.--.+. 10:16:02 nlm 10:16:05 hmph :P 10:16:17 oklopol, are you anti-golfing? 10:16:24 nahhhhhhhh 10:16:37 i'm trying to convert math expressions to bf. 10:16:38 was supposed to be a raised eyebrow 10:16:56 oklopol, oh ok :P 10:17:12 ^ul (nlm)S 10:17:13 nlm 10:17:54 ^ul (:(oklopol)S^):^ 10:17:54 oklopoloklopoloklopoloklopoloklopoloklopoloklopoloklopoloklopoloklopoloklopoloklopoloklopoloklopoloklopoloklopoloklopoloklopoloklopoloklopoloklopoloklopoloklopoloklopoloklopoloklopoloklopoloklopoloklopoloklopoloklopoloklopoloklopoloklopoloklopoloklopoloklopoloklopoloklopoloklopoloklopoloklopoloklopoloklopoloklopoloklopolok ...too much output! 10:17:58 heh 10:19:13 +ul (:(oklopol)S^):^ 10:19:14 oklopoloklopoloklopoloklopoloklopoloklopoloklopoloklopoloklopoloklopoloklopoloklopoloklopoloklopoloklopoloklopoloklopoloklopoloklopoloklopoloklopoloklopoloklopoloklopoloklopoloklopoloklopoloklopoloklopoloklopoloklopoloklopoloklopoloklopoloklopoloklopoloklopol ...too much output! 10:19:29 ^bf >>++>>++++[<<[-<++>>+<]>[-<+>]>>-]<<[-[>++++++<-]]>++++. 10:19:31 fungot seems to be a lot faster 10:19:34 ...out of time! 10:19:34 AnMaster: how's he do that 10:19:56 ^bf >>++>>++++[<<[-<++>>+<]>[-<+>]>-]<<[-[>++++++<-]]>++++. 10:19:56 . 10:19:58 ... 10:20:01 fungot, by running on cfunge I guess. However referring to yourself using "he" is strange 10:20:01 AnMaster: there are certain " linkers" these days. send the tourists to go a single day without scheme, you wouldn't need to be 10:20:12 huh 10:20:18 ^bf >>++>>++++[<<[-<++>>+<]>[-<+>]>-]<<[>++++++<-]>++++. 10:20:18 10:20:21 :D 10:20:23 okay i suck. 10:20:26 oklopol, um? 10:20:31 ^bf +. 10:20:32 10:20:40 oh ok not that 10:20:44 ^ul ((pol)~^(fok)~^(kok)~^)(~(oklo)~*S::a~*)~^ 10:20:44 ...out of stack! 10:20:48 argh 10:20:59 ^bf --[+>-<]. 10:21:06 ^bf --[+>-<]>. 10:21:06 þ 10:21:11 ^bf ---[++>-<]>. 10:21:14 ^ul ((pol)~^(fok)~^(kok)~^)(~(oklo)~*S::a~*):a~*~^ 10:21:15 ^bf --[++>-<]>. 10:21:16 ...out of time! 10:21:16 oklopoloklofokoklokok 10:21:16 ÿ 10:22:07 ^bf --[+++>-<]>. 10:22:08 ª 10:22:28 ^bf ---[+++>-<]>. 10:22:28 ÿ 10:22:33 ^bf --[++++>-<]>. 10:22:38 ...out of time! 10:22:41 ^bf >++>++++[<[-<++>]<[->+<]>>-]<[->++++++<]>++++. 10:22:41 Ä 10:22:44 ... :D 10:22:50 ^bf --[>++++<-]>. 10:22:50 ø 10:23:04 oklopol, now do that as utf8! 10:23:06 ^bf >++>+++[<[-<++>]<[->+<]>>-]<[->++++++<]>++++. 10:23:06 d 10:23:10 ^bf --[>+++<-]>. 10:23:10 ú 10:23:12 ^bf >++>+++[<[-<++>]<[->+<]>>-]<[->++++++<]>++++.--.+. 10:23:12 dbc 10:23:14 finally 10:23:17 2^4 10:23:30 ^bf --[>+++++<-]>. 10:23:30 ö 10:23:41 hm 10:27:15 ^code 000aa+-p 10:27:16 ^reload 10:27:16 Reloaded. 10:27:44 Added some real spaghetti Befunge code to manage the ignore list; although it still doesn't actually ignore anything. 10:27:47 ^ignore foo 10:27:47 OK. 10:27:49 ^ignore bar 10:27:49 OK. 10:27:52 ^ignore baz 10:27:52 OK. 10:27:53 ^ignore 10:27:53 foo bar baz 10:27:56 ^ignore -bar 10:27:56 OK. 10:27:58 ^ignore 10:27:58 foo baz 10:28:01 ^ignore fizzie 10:28:01 OK. 10:28:09 ^ignore oerjan!!!11 10:28:09 OK. 10:28:23 ^ignore 10:28:23 foo baz fizzie oerjan!!!11 10:29:02 Unlike the command list, I bothered to write that bit which lets me remove items from the middle of the list. 10:29:49 Rather messy code: http://zem.fi/~fis/ignore.txt 10:30:12 fizzie, also what do you think about this: http://www.club.cc.cmu.edu/~ajo/funge/#Fungus 10:30:13 fizzie, ? 10:30:19 I'm not quite sure whether to make ^ignore an owner-only command here. 10:30:28 fizzie, owner only I'd say 10:30:55 definitly 10:30:59 definitely* 10:33:06 Well, you people might want to have some way of making fungot go quiet even if I'm not here. 10:33:07 fizzie: c-a conflicts with emacs and have a look 10:33:20 Fungus looks fungey. 10:33:48 fizzie, well, what about someone abusing it? 10:35:33 if someone would ever write a lint for funge I know the perfect name: flint 10:36:13 -!- rodgort has quit ("Coyote finally caught me"). 10:36:22 -!- rodgort has joined. 10:36:39 I rather doubt that would be very much fun for anyone, but I guess I could make it owner-only. Or I could just make the ^ignore command available for ignored people too. 10:38:00 -!- oerjan has quit ("Cibus"). 10:44:00 err so you can revenge? :P 10:47:21 So you can unignore yourself. 10:47:49 ah. 10:47:54 Maybe I underestimate how much fun you could get out of that. 10:47:56 ^reload 10:47:56 Reloaded. 10:48:05 Should be owner-only now, unless I screwed it up. 10:49:09 ^ignore fizzie 10:49:10 "make the ^ignore command available for ignored people too" does sound more like letting people revenge than letting the unignore themselves, if you don't know ^ignore also unignores 10:49:18 guess it is owner only 10:49:35 hmm 10:49:45 fizzie, added ignore for thutubot? 10:49:46 actually not exactly if it's already allowed for everyone 10:49:48 Owner-only, yes, and it doesn't really do anything, except manipulates that list. 10:49:49 ^ul (S)S 10:49:50 S 10:50:11 The actual ignoring part does not happen yet. 10:50:46 +ul (^ul (hello)S)S 10:50:46 mutable list? that's so cool 10:50:47 ^ul (hello)S 10:50:47 hello 10:51:03 oklopol: Did you check out the implementation: http://zem.fi/~fis/ignore.txt 10:51:07 So very messy. 10:51:07 Single assignment befunge! 10:51:30 fizzie: i can't exactly read befunge anyway; i've only written one program in it, and i haven't read even one program in it 10:51:35 fizzie, run that graph program of yours 10:51:50 well 10:51:58 i can, but not fast. 10:52:07 >$0".derongi toN">61g: 10:52:08 ? 10:52:11 61g? 10:52:43 oklopol, what about malbolge? 10:52:44 (6,1) has the "current output length" value. 10:52:55 "derongi toN" :D god i love s0gnirts 10:53:06 fizzie, ah 10:53:08 So 61g:c+61p3P appends 12 characters of output. 10:53:08 (the plural suffix is a prefix for gnirts right?) 10:53:26 oklopol, um it is string backwards 10:53:26 AnMaster: no, i can't read malbolge 10:53:35 I've seen people write it "gnirtses". 10:53:43 AnMaster: yes that's the joke 10:53:44 fizzie, yeah I end up doing that 10:53:44 But I guess that's a bit boring. 10:54:59 oh and that fungus use elf for file format 10:55:15 http://www.cis.hut.fi/htkallas/ignore.png 10:55:15 so readelf works on it, though it gives strange values 10:56:09 fizzie, that graph seems to be missing parts? 10:56:21 desrever si gnirts eht nehw xiferp a semoceb xiffus larulp eht si ekoj eht 10:56:23 Which parts? 10:56:27 god i'm slow at that 10:56:37 g#v_^ 10:56:38 for example 10:56:52 It's got the 'g'. 10:57:02 The others are not "real instructions", it's implied by the edges. 10:57:15 ah 10:57:48 fizzie, why does it quit in various places? 10:57:58 Because I had to add something there to prevent wrapping. 10:58:09 Those go outside the ignore.txt parts to other areas of fungot. 10:58:09 fizzie: if you write an alternative implementation of syntax-case 10:58:13 ah 10:59:14 There are some silly parts, like the node with only a "0" in it and true/false edges leading out. 10:59:20 It will always take the 'false' edge. 10:59:22 fizzie, it got three exit points? 11:00:18 * AnMaster ponders a fingerprint CAM 11:00:43 as in Content-addressable memory 11:00:58 would actually be searching inside the interpreter 11:01:04 in some list in funge-space 11:01:05 Yes, it's got three @ nodes, which is pretty strange, given that there are only two @s in the program. 11:01:09 fizzie, or would that be cheating? 11:02:45 Oh, right: it has three separate @ nodes, because that one @ is entered both from the left side and from bottom; it won't merge those into same node. Oh-well, a minor issue. 11:03:43 And I guess it might be a bit un-Befungeish to move that sort of stuff into the interpreter. I have some pangs of conscience even now for using STRN so much. 11:04:24 fizzie, so if I made a fingerprint for it, you wouldn't use it? 11:04:45 I'm not sure. I might not use it. 11:04:55 On the other hand I'm pretty lazy. 11:05:06 well I won't write it then 11:21:19 There's a slight bug in cfunge's REXP. 11:22:42 If you specify the REG_NOSUB flag, the 'regexec' function ignores the nmatch and pmatch arguments, and since the 'matches' array is not initialized to be empty, it pushes 128 empty strings on stack. 11:24:15 "PXER"4($$0"oof"4C0"oof"0E.a,@ prints out 128; it should (arguably) print 0. 11:26:57 hm 11:27:38 fizzie, so lets see what you meant... 11:28:25 "is not initialized" <-- valgrind doesn't complain 11:28:40 Well, is not initialized to be empty. 11:28:56 Obviously it's initialized to zeroes, since it's a static thing. 11:29:18 well yeah 11:29:31 But your push_results is testing for rm_so != -1. 11:29:46 grr *reads man page* 11:29:59 Or something like that, anyway; haven't looked at it closely enough, but it does push 128 strings in there. 11:30:12 if (matches[i].rm_so != -1) { 11:30:14 yeah... 11:30:23 My regexec man page says it completely ignores nmatch/pmatch with NOSUB. 11:30:40 fizzie, however it may take a few minutes to fix since I don't remember the API 11:30:53 got to re-read the man page first 11:31:38 Well, I did work-around it already. Actually I'm not quite sure what REXP's "E" *should* do when the expression uses NOSUB; either push a single zero, or alternatively push nothing. 11:32:11 well it shouldn't be 128 strings 11:32:49 * AnMaster adds a static bool compiled_nosub 11:33:19 fizzie, tell me what it should do instead :) 11:34:24 I'm not sure. If you just look at the REXP doc in rcfunge manual, it says "E leaves the results of the match as a series of 0gnirts strings. Each string representing the matched portion of a substring. Top of stack will have the count of these 0gnirts strings." in which case the most logical thing would be to push a zero count. 11:34:39 ^reload 11:34:39 Reloaded. 11:34:42 ^ignore 11:34:51 fizzie, what does rc/funge does on NOSUB? 11:35:41 AnMaster: Uh... pushes 100 empty strings plus the number 100, I think. 11:35:57 well that is because it has 100 static matches 11:35:57 It has a very similar implementation to yours, at least on a quick glance. 11:36:10 fizzie, well iirc rc/funge is per-ip compiled 11:36:27 Okay, there's that. 11:36:55 ^ignore ^thutubot! 11:36:56 OK. 11:37:13 +ul (^ul (hello)S)S 11:37:14 ^ul (hello)S 11:37:20 156 compiled_nosub = (flags & FUNGE_REG_NOSUB); 11:37:20 (gdb) print (flags & FUNGE_REG_NOSUB) 11:37:20 $3 = 0 11:37:21 ^ul (hello)S 11:37:21 hello 11:37:26 (gdb) s 11:37:26 159 stack_freeString(str); 11:37:26 (gdb) print compiled_nosub 11:37:26 $6 = false 11:37:26 wtf 11:37:27 !? 11:37:53 hm 11:37:54 -!- KingOfKarlsruhe has joined. 11:38:00 (gdb) print (flags & FUNGE_REG_NOSUB) == true 11:38:00 $4 = 0 11:38:00 (gdb) print (flags & FUNGE_REG_NOSUB) == false 11:38:00 $5 = 1 11:38:01 so 11:38:04 ^ignore ^optbot! 11:38:05 OK. 11:38:05 fizzie: We can't prove two things to be the same, sure. 11:38:07 what the heck was going on there 11:38:18 fizzie, ? 11:38:34 wait 11:38:38 1 is true, right 11:38:38 duh 11:38:43 * AnMaster coded bash before today 11:38:59 ^ul (opt)(bot)*S 11:39:00 optbot 11:39:00 fungot: http://pastebin.ca/932888 11:39:07 No more fancy-loops. :/ 11:40:05 fizzie, um I get "BAD: TOS was not 1" from Mike's rexp.f98 11:40:07 now 11:40:55 it doesn't say so if I let it push 128 strings instead 11:40:57 fizzie, ! 11:41:09 Checking. 11:41:19 it may be buggy, don't know 11:41:41 fizzie: does java have a join of some sort? 11:42:15 oklopol, java.too.long.namespace.name.os.system.execte("ghci")? 11:42:16 ;P 11:42:42 AnMaster: rexp.f98 seems to compile all regexps with flags==1, so any change in behaviour that affects only REG_NOSUB'd things shouldn't affect it. 11:43:57 oklopol: What, to concatenate strings? 11:44:07 hnuh 11:45:20 fizzie: wait. 11:45:35 My first instinct for string-joining would just be StringBuilder sb = new StringBuilder(); for (String s : collection) sb.append(s); sb.toString(); which looks very Javay. Waiting for further instructions. 11:46:23 wtf 11:46:36 fizzie, either I get your test program working or I get rexp.f98 working 11:46:38 but not both 11:46:52 are you sure your test program is actually correct? 11:46:57 yeah that works. it's just that's a bit ugly; i'll probably inherit a joinable container from arraylist then... 11:47:23 oklopol: It's more than a bit ugly, but I don't think the built-in containers (or String class) has any very helpful methods. 11:47:44 true,. 11:48:02 fizzie, ? 11:48:16 god i wish java had less support for all that io shit, and more algorithmical support i would actually have use for 11:48:27 AnMaster: Of course not, but really, if you change things so that the behaviour differs only when the C command has the NOSUB flag, I don't see how it could alter rexp.f98 results at all. 11:48:42 well 11:48:53 fizzie, well I suspect your program doesn't set NOSUB, I'm not sure 11:48:59 i don't know what the standard classes contain exactly, the list is too long to read 11:48:59 Er. 11:49:09 but most of it seems quite useless 11:49:11 I don't see how 4C *wouldn't*. 11:49:29 ah wait it does 11:49:31 perhaps i'll use a stringbuilder for now, i guess i just need this one join. 11:49:40 I was confusing different types of flags 11:49:46 as in funge flags and system flags 11:50:00 (i do like the idea of SequenceJoinerFactory though) 11:50:29 oklopol: They've got String.split(String regex), that's at least something. 11:50:39 yeah, that was a nice addition 11:52:46 Should save those ignores in the state-file, currently they're non-persistent. 11:52:54 fizzie, fix (push a single 0) pushed 11:53:45 so you can update your copy if you want 11:54:07 now I hope to get around to making that release soonj 11:54:09 soon* 11:54:34 Okay. I'll try to remember to maybe clean out my workaround (not use NOSUB and just discard the useless match-string) when I next update. Although then it won't run correctly on rc/funge with ignores. 11:55:06 fizzie, well report a bug to rc/funge author 11:55:14 and doesn't CCBI have REXP? 11:56:49 Don't know about that. 12:00:17 I think fungot should work just fine even if "PXER"4( reflects, it just won't process the ignores. 12:00:18 fizzie: huh. people are evil. blogs cause conservative canadian governments. i know 2. that's obvious and you did refer to a certain class of bad things/ alone 12:00:46 fungot: You're starting to sound like a paranoid person, you know. 12:00:46 fizzie: it would be 12:02:04 fungot, it is a conspiracy! 12:02:04 AnMaster: but i wanna try debian some day... of course. it's pure space opera. but as i got out 12:02:19 debian.... space opera? 12:02:26 huh 12:03:30 fungot: You're actually running in a Debian system right now. 12:03:30 fizzie: i'm familiar with 12:03:39 fungot: Well, I guess you would be. 12:03:40 fizzie: that's how i managed it, and the oft-cited horror film fnord a wonderful life_, are noteworthy examples of works accidentally committed to the festive mode. 12:04:26 fizzie, how would you interpret THAT? 12:04:42 I don't really know; I guess it got all confused. 12:05:46 Though "fnord a wonderful life" does sound like it'd be a quite scary film. 12:08:32 THAT? a new fingerprint? 12:08:53 pointers for befunge maybe? 12:10:18 (and that was a joke, AnMaster :P stupidity is not my only thing, i also make jokes!) 12:11:32 -!- Corun has joined. 12:13:38 oklopol, hah 12:16:30 -!- optbot has set topic: the entire backlog of #esoteric: http://tunes.org/~nef/logs/esoteric | base-3 is actually fairly feasible... -1, 0, 1 corresponding to the polarity of the volts.. 12:16:43 reverse polarity! 12:19:38 -!- Corun has quit ("This computer has gone to sleep"). 12:22:40 fizzie, I'm preparing 0.3.3 release 12:46:23 (hit a few issues) 13:18:31 Come to think of it, I could've done a lot simpler ignoration thing by simply using a single ignore regex instead of a list of them; and then I wouldn't need to compile all those regexs every time someone speaks to the bot. 13:19:56 fizzie, better change then 13:19:57 :) 13:54:50 ^code 000aa+-p 13:54:51 ^reload 13:54:52 Reloaded. 13:55:13 ^ignore ^(optbot|thutubot)! 13:55:13 Bad re: 112 13:55:13 fizzie: okay whut 13:55:39 "Okay whut" indeed. 13:57:13 Where did it get "112" from? I don't think translate_return_C should be even capable of giving out that number. 13:57:36 ^ignore foo 13:57:36 Bad re: 0 13:57:48 Nor that. 13:57:58 There must be a bug in my code. 13:58:22 Oh, right, it doesn't have REXP loaded because that's done in the initialization now. 13:58:28 ^raw QUIT :let's just restart 13:58:28 -!- fungot has quit ("let's just restart"). 13:58:58 -!- fungot has joined. 13:59:06 ^ignore ^(optbot|thutubot)! 13:59:07 OK. 13:59:07 fizzie: ok 13:59:13 Heh. 13:59:13 fizzie: this new jpeg after ' jpegtran -rotate 270' and switching 1600/ 1200 in gs invocation is actually pretty nice. 13:59:17 Hmm. 13:59:18 H7zzi=: its working over here. :-p so, i load the usb module. 13:59:24 That's not right. 13:59:45 Interesting errors, though. 13:59:46 fizzie: bucket sort... in bancstar. 13:59:51 -!- fungot has quit (Remote closed the connection). 14:00:11 Back to the drawing-board with it. 14:16:06 -!- fungot has joined. 14:16:12 ^ignore ^(optbot|thutubot)! 14:16:12 OK. 14:16:12 fizzie: I can't decide if I should make 0x00 short for current outfile, 0x01 current infile, 0x02 stdout, 0x03 stdin 14:16:22 Now it should be okay. 14:16:28 fungot: Do you feel fine? 14:16:28 fizzie: tell me when you find it. i doubt the basic design on paper in the bus. 14:16:34 ^ignore 14:16:34 ^(optbot|thutubot)! 14:16:35 fungot: (EXPR . EXPR) 14:16:45 It also seems to be doing the ignoration. 14:17:18 +ul (^ul (ok?)S)S 14:17:19 ^ul (ok?)S 14:17:23 Yes. 14:18:25 ^def help ul (^ ; ^def ; ^show [command]; lang=bf/ul, code=text/str:N; ^str 0-9 get/set/add [text]; ^bool)S 14:18:25 Defined. 14:18:48 ^save 14:18:49 OK. 14:19:19 Decided to do the ^help command that way for easier modification. Hopefully no-one will override it with anything silly. 14:21:39 hah 14:21:57 ^talk 14:22:03 fungot, hi? 14:22:03 AnMaster: or someone did 14:22:06 ^show 14:22:06 echo reverb rev rot13 rev2 fib wc ul cho choo pow2 source help 14:22:13 ^echo optbot 14:22:13 optbot optbot 14:22:14 AnMaster: but that's a result of very little operator overloading - most operators do a very specific thing. 14:22:14 fungot: yes 14:27:15 release time! 14:46:33 fizzie, cfunge 0.3.3 is released 14:49:46 I suppose there's not much difference between releases and bzr-pulled copies? 14:50:31 fizzie, not really, there were some typo fixes, updated version number, updated changelog 15:10:42 -!- ehird has joined. 15:10:48 fizzie: 15:10:57 make ^ignore and ^unignore available to all unignored people 15:11:06 but let you only ignore one person every 5 minutes 15:11:17 impossible to take over ze world, but nicely anarchistic 15:12:20 Well, I took the lazy way and just added a single owner-configurable ignoramation regex. 15:12:33 fizzie: what about when you are away and co2games comes here ;-P 15:13:53 hey, fizzie 15:13:56 unignore optbot! 15:13:56 ehird: then I was told you can't use input 15:13:57 :-( 15:14:07 -!- KingOfKarlsruhe has quit (Remote closed the connection). 15:15:19 Bot-loop preventamation was, I think, the whole point of ignorey. 15:15:31 fizzie: But optbot/fungot conversations are wonderful 15:15:31 ehird: i assume you understand i'm just representing a part of the main array python-tsyle? 15:15:32 ehird: gauche now supports wrights matcher, right sarahbot? :p). can i do 15:16:21 -!- ehird has changed nick to optbot_feels_rej. 15:16:26 ... 15:16:27 optbot_feels_rej: Bullshit 15:16:30 LOL :D 15:16:30 optbot_feels_rej: :) 15:16:35 Shut up optbot 15:16:35 optbot_feels_rej: but I think that's kind of cheating :P 15:16:39 fungot: hi 15:16:39 optbot_feels_rej: code got borken somewhere. melody of oblivion? 15:16:39 optbot_feels_rej: If you can mark a function as, say, never being jumped into. 15:16:40 fungot: at least it proves you aren't the first one to make that mistake 15:16:44 Huh. 15:16:45 optbot_feels_rej: HQ9+ is just as fully functional; it's been implemented 15:16:48 ^talk 15:16:48 optbot_feels_rej: my monitor plugs into my USB port... 15:16:59 ^ul (optbot)S 15:17:00 optbot_feels_rej: afternoon. 15:17:00 optbot 15:17:00 fungot: "x", [[:apply, ["x", "y"], []]]]] 15:17:05 -!- optbot_feels_rej has changed nick to ehird. 15:50:42 -!- hakware has joined. 15:50:58 -!- hakware has changed nick to ENKI-][. 16:21:33 -!- oerjan has joined. 16:39:56 -!- ENKI-][ has quit ("WeeChat 0.2.6"). 16:44:15 -!- hakware has joined. 16:44:55 -!- hakware has changed nick to ^dENKI. 16:54:46 -!- KingOfKarlsruhe has joined. 17:03:39 http://www-cs-faculty.stanford.edu/~knuth/news08.html NOOOOOOOOOOOOOOOOOOOOOOOOOOOOO 17:04:08 [[Instead of writing personal checks, I'll write personal certificates of deposit to each awardee's account at the Bank of San Serriffe, which is an offshore institution that has branches in Blefuscu and Elbonia on the planet Pincus.]] 17:04:29 what? huh 17:09:15 Hahaha, "It turns out that only 9 of the first 275 checks that I've sent out since the beginning of 2006 have actually been cashed. The others have apparently been cached." 17:09:19 Well put. 17:09:45 GregorR, what do you get for pointing out a typo on that page ;) 17:09:49 ah wait 17:09:51 double meaning 17:09:56 ... 17:09:57 >_______< 17:10:02 No, single meaning X_X 17:10:05 Just clever :P 17:10:06 ehird, remember I'm not a native speaker 17:10:07 Note to self: Check for AnMaster before linking to any joke. 17:10:15 ehird: http://codu.org/ is now themed by my color matcher! 17:10:37 ehird, any English pun yes 17:10:38 GregorR: when you first said that i thought you meant each load 17:10:39 in the logs 17:10:41 but NO 17:10:42 OH NO 17:10:44 IT WAS A LIE 17:10:47 A DIRTY COMMUNIST LIEFJDSKF 17:10:58 Uh, it is done each load, it's just cached in a cookie. 17:11:05 If you want to refresh it just use the menu. 17:11:08 NO 17:11:11 MAKE IT CHANGE EACH TIME 17:11:11 :_: 17:11:22 I didn't want the theme to change as you navigated the page :P 17:11:27 GregorR, is that box different in colour at the top and the bottom? 17:11:35 also it change on each load for me 17:11:35 AnMaster: Yeah, it has a gradient. 17:11:37 AnMaster: Welcome to 1997 17:11:39 it's called a GRADIENT 17:11:41 but I reject cookies ;P 17:11:43 also, welcome to paranoia 17:11:43 AnMaster: You must have cookies disabled. 17:11:44 Right. 17:12:02 ehird, yes I just didn't remember the English word........ 17:12:13 GregorR: remove the cookies 17:12:14 ;_________: 17:12:20 People who don't have cookies enabled will have a much more ... colorful experience on this site :P 17:12:35 GregorR, and it doesn't work with javascript off ;P 17:12:42 17:13:01 i've given up complaining because you have a vocabulary of 30 things, 29 of which are annoying. 17:13:05 i haven't seen the 30th in person yet. 17:13:38 ehird, and I thank you for your help with funge recently 17:14:01 AnMaster: which consists of "repeating fizzie's and ais523's lines because you didn't listen the first time"? 17:14:21 ehird, ? 17:14:35 static array 17:15:36 well the idea about not using linked-list style hash 17:15:40 that too 17:15:51 AnMaster: With JS off it will now give a good default theme, rather than drab white-on-black. 17:16:00 And by "now" I mean "when I upload it" :P 17:16:01 GregorR, yes and that is nice 17:16:10 GregorR, um it gave me black on black? 17:16:12 with white text 17:16:23 And by "now" I mean "when I upload it" :P 17:16:40 GregorR, oh I thought you meant *white box* on black page 17:16:48 which would have been worse 17:16:50 Nono. 17:16:54 (It is now uploaded) 17:17:02 GregorR, nice yes 17:17:15 "No JavaScript? Click here to disable the JavaScript menu." 17:17:17 GregorR: 5 amelican moneys for no cookie 17:17:19 the click here link is 404 17:17:22 :( 17:17:31 GregorR: I will kill you if you don't then 17:17:43 AnMaster: ORLY? I haven't looked at that in a long time :P 17:18:19 well I usually have a noscript exception for you site, because of your interesting projects 17:18:29 Hahaha 17:18:32 like jsmips 17:18:32 :P 17:18:50 GregorR: ok, here's my idea: 17:18:56 GregorR, how goes jsmips then? 17:19:01 ehird, just turn off cookies? 17:19:03 AnMaster: no 17:19:03 :P 17:19:06 AnMaster: Haven't updated it in a while. 17:19:18 GregorR: all non-main page pages cache in cookie 17:19:20 but the main page doesn't 17:19:30 ehird, well you asked me to turn on cookies/javascripts lots of times, so why can't I ask you to turn it *OFF* for once :P 17:20:03 'cuz having them off is silly 17:20:11 :P 17:20:22 AnMaster: Hmmmm, seems that somethings wonky about the PHP on this server ... 17:20:30 GregorR: http://codu.org/dcvogllmrcmcdp.ogg this would be a good OS startup sound 17:20:34 if it made you listen to it all before starting 17:20:53 Ohhhh kay :P 17:21:23 ehird, I disagree :P 17:21:32 AnMaster: that's because you suck 17:21:57 ehird, well your niceness yesterday didn't last that long. :P 17:22:05 back to usual I gues 17:22:05 you started being annoying again 17:22:06 guess* 17:22:15 ehird, hey wrong, *you* started ;P 17:22:27 /ignore AnMaster. Problem solved. 17:22:40 *shrug* 17:23:04 GregorR: add a 'generate two non-matches' 17:23:13 I got an idea eh 17:23:14 ehird* 17:23:19 that you would like 17:23:22 basically 17:23:39 cookie stored option, that lets you select "random every time" 17:23:45 of different types 17:24:11 GregorR, what do you think about that way? 17:25:55 Fixed nojs. 17:25:57 Now then :P 17:26:13 GregorR: how does nojs.php even work? 17:26:19 some weird transformation? 17:26:20 GregorR, works 17:26:40 AnMaster: Yeah, that'd probably work *shrugs* 17:26:51 GregorR: also, why not just use a