00:03:31 -!- sebbu has quit (Read error: 110 (Connection timed out)). 00:03:38 -!- calamari has quit ("Leaving"). 00:23:44 -!- OkloThePol has joined. 00:23:44 -!- oklofok has quit (Read error: 104 (Connection reset by peer)). 00:51:49 Anyone want to try out my drainfuck parser? 00:54:18 If my langauge has all of the brainfuck commands working the same way, is my language turing-complete as well? 00:57:01 Hey I thought brainfuck was intended to have the smallest compiler ever. 00:57:41 200 kb my ass 00:58:01 My interpreter is only 88 kb 00:58:59 yeah but theirs is tested :o) 00:59:05 True 00:59:06 (a joke, but you never know) 00:59:09 But mine can be. 00:59:18 A joke? Theirs aren't tested? 00:59:54 I'm afraid I try to stay away from bf-alikes, so would be hard pressed to see any problems with it, but hopefully someone else will give you a hand 01:00:09 :D 01:00:12 D:* 01:01:40 Wow 01:02:02 Why is my interpreter smaller than their compilers? 01:02:35 well, compilation and interpretation are different beasts for a start, perhaps that has an effect. 01:02:43 err 01:02:54 i think the brainfuck compiler was 200b 01:02:58 umm, actually 01:03:03 Oh...bytes... 01:03:04 what compiler are you talking about? 01:03:05 ...eheh.... 01:03:08 the original? 01:03:17 The smallest working 01:03:33 well the original was near 200b 01:03:34 Sorry I was thinking kilobytes ahah 01:03:57 there's a 3d game in 98kb 01:04:07 .kkrieger or something 01:37:32 -!- oklofok has joined. 01:37:33 -!- OkloThePol has quit (Read error: 104 (Connection reset by peer)). 01:40:07 -!- M0ny has quit ("À vaincre sans péril on triomphe sans gloire..."). 01:42:18 -!- olsner has quit ("Leaving"). 02:24:45 -!- bsmntbombdood_ has joined. 02:25:19 -!- bsmntbombdood has quit (Read error: 110 (Connection timed out)). 03:40:10 -!- GreaseMonkey has joined. 03:42:27 -!- GreyKnight has quit ("fell into a bucket of sleep"). 03:45:36 -!- optbot has set topic: the entire backlog of #esoteric: http://tunes.org/~nef/logs/esoteric | you could easily have like 10 buttons that way. 04:13:35 -!- ihope has quit (Read error: 60 (Operation timed out)). 04:31:59 -!- psygnisfive has joined. 04:32:10 so 04:32:15 http://wellnowwhat.net/transfers/prettyboy.jpg 04:32:16 hot 04:32:18 want him 04:32:21 love his cock 04:32:23 D: 04:34:38 Note to self: Read commentary BEFORE clicking random links. 06:10:12 :) 06:10:18 TOO MUCH FUN? 06:18:54 hey guys- want to try out a new stack-based language I've cooked up? 06:18:55 http://rodger.nonlogic.org/games/GraphScript/ 06:19:23 the interpreter is still a little brittle, but it can deal with most syntax errors somewhat gracefully 06:20:24 running either of the frontends I provide from the console will provide you with more helpful debugging information in the case of an error than just running the applet in a browser. 06:27:03 GregorR, pikhq? 06:31:37 i dislike your symbol choices. :( 06:36:38 how so? Not a fan of square brackets? 06:47:22 -!- Sgeo has quit (Read error: 110 (Connection timed out)). 06:47:33 not a fan of not using + - etc 06:49:03 >.< 06:50:16 -!- psygnisfive has quit ("http://www.mibbit.com ajax IRC Client"). 07:28:42 well, 'night, folks 07:50:05 RodgerTheGreat: I'll check it out in the morning. 07:59:59 -!- clog has quit (ended). 08:00:00 -!- clog has joined. 08:12:01 -!- oerjan has joined. 08:18:21 i'll check it out in the now. 08:20:18 these days, efficiency requires you to check it out in the past 08:20:39 always keep a time machine handy 08:20:47 well, in the past, i was checking out your mum. 08:21:21 how far in the past? 08:24:08 -!- tusho has joined. 08:24:41 -!- kar8nga has joined. 08:24:45 Whew. 08:24:47 I am here early. 08:25:11 yay i drew a line 08:25:14 surprisingly, some people are awake 08:26:45 i was woken up by a friends asking me to go to some whacky martial arts thing 08:28:40 oerjan: yea, but i doubt there's anyone british in here 08:28:41 :) 08:31:19 * oerjan watches tusho doubt himself 08:31:38 i'm too tired to doubt. :3 08:31:53 if anyone is 1. awake 2. on IRC 3. british say "aaaaaaaaaafjkahsiuebfbuaishdiuqh312h3893y*&*&£Y" 08:32:19 i really thought you were british 08:34:13 oerjan: that was a command to others 08:34:14 not to myself 08:35:08 but still, you seem not to include yourself in "anyone" 08:35:09 -!- puzzlet has quit (Remote closed the connection). 08:35:18 -!- puzzlet has joined. 08:35:43 conclusion: tusho is a british robot 08:40:39 or, he is two 08:41:51 impeccable logic 08:43:36 oerjan: i am fuzzy 08:43:38 and i like hugs. 08:44:05 * oerjan hugs tusho 08:44:15 eww, that fuzz is mold! 08:44:17 * tusho 's fuzzy state continues unchanged 08:44:21 no it's not :( 08:46:13 i read too slowly 08:46:49 me too. must be a sign of genius. 08:47:56 :D 08:47:57 * oerjan investigates the fuzz more carefully with a microscope 08:48:11 hey, there are little people in there! and skyscrapers! 08:48:16 8| 08:48:37 and a starbucks! 08:50:51 as a plus side to my slow reading, i do usually remember every little detail about the book 08:51:10 except for names and years of course, for some reason i cannot remember those 08:51:24 easier for me to convert the name to meaningless numbers, then remember them 08:51:35 or, well, think of the word as a random string 08:51:39 that doesn't explain the years 08:51:43 -!- oklofok has changed nick to oklopol. 08:52:07 basically, at some point in my life i've decided names and times are not important 08:52:22 and now i just cannot remember them. 08:53:20 -!- OkloThePol has joined. 08:53:21 -!- oklopol has quit (Read error: 104 (Connection reset by peer)). 08:53:52 i can't remember any one of my friends' birthday, even though i've put quite a lot of effort into remembering them; i do remember about 20 long passwords though, even the passwords of a few of my friends, i remember all phone numbers, and much more useless trivia though 08:54:10 though though 08:54:25 i do remember the names of my friends though, that's not as bad :) 08:55:07 -!- OkloThePol has changed nick to oklopol. 08:55:34 fucking mirc remembers my jokes and keeps repeating them 08:56:37 i read really fast. 08:56:59 i don't really remember many details post-book but I could give you a summary. it's a pleasure for while I read it, at least 08:57:19 I also don't type anywhere near the speed at which I can for some reason. 08:57:25 must be a sign of .. erm 08:59:02 i don't read books where it's useful to give summaries 08:59:12 i guess that's one reason why i read slow, i don't read fiction 09:02:44 Deewiant, now I'm around 09:03:01 well, with a lot of hacking (removing the mmap check as well as a bunch of fingerprints) I managed to get cfunge to compile under cygwin 09:03:01 crashes in o: 09:03:01 Trying to write to it with o... 09:03:01 Segmentation fault (core dumped) 09:03:02 guess that testing cfunge is for tomorrow then 09:03:05 got a backtrace? 09:03:39 it works perfectly under Linux, FreeBSD and Mac OS X 09:03:42 I know that 09:06:42 and well those changes you made, they could have caused the issue, potentially 09:08:21 how do I get a backtrace 09:09:00 you got gdb? 09:09:05 yes 09:09:23 gdb --args path/to/cfunge path/to/mycology.b98 09:09:40 type "r" (without quotes) to run the program 09:09:50 wait till it crashes and you get a gdb prompt 09:09:54 then type: 09:09:56 bt 09:10:02 useless stacktrace 09:10:08 Deewiant, useless as in? 09:10:13 as in, useless 09:10:17 I can paste it for you 09:10:20 you mean lot of question marks? 09:10:25 did you build with debug info? 09:10:40 I just did "cmake . -G 'Unix Makefiles'" 09:10:44 DEBUG for build type in cmake 09:11:05 -DCMAKE_BUILD_TYPE=DEBUG 09:12:12 try that Deewiant :) 09:12:18 the other question I had was how to remove fingerprints from the build 09:12:19 Deewiant: i wanna contribute a ccbi patch, are there any neat fingerprints i could implement for ccbi2? 09:12:41 I went hacking in the CMake-generated makefiles to do it 09:12:51 but surely there's an easy way 09:13:18 Deewiant, remove the directory and rerun tools/gen_fprint_list.sh 09:13:21 tusho: I don't know about "neat", but sure, there's a bunch of stuff of RC/Funge's that I haven't done 09:13:25 to rebuild fingerprints.h 09:13:32 remove what directory 09:13:38 Deewiant: yay. any up-to-date ccbi2 source tarballs? 09:13:41 Deewiant, src/fingerprints/FILE or whatever 09:13:52 and the file src/fingerprints/FILE.spec 09:14:04 tusho: would be kinda pointless as the fingerprint stuff isn't done in ccbi2 (currently blocked on a compiler bug indefinitely) 09:14:13 Deewiant: ouch. what's the bug? 09:14:15 any hacky workarounds? 09:14:18 none 09:14:25 I already have dozens of bug workarounds in my code 09:14:29 couldn't think of one for this 09:14:34 Deewiant, I didn't have any need to disable some in an easy way. sorry 09:14:35 Deewiant: what's the bug? 09:14:48 http://d.puremagic.com/issues/show_bug.cgi?id=2339 09:15:09 Deewiant, anyway got any useful stack trace now? 09:15:21 AnMaster: I am removing fingerprints so that I can get this to build 09:15:26 Deewiant, ah 09:16:01 Deewiant: couldn't you like, restructure how your stuff works and sacrifice neatness for workingness 09:16:01 :P 09:16:09 mixin magic is nice, but when it doesn't work... 09:16:13 tusho: that was 1.0 09:16:14 tusho, you mean make it work like ccbi 1? 09:16:22 true. :p 09:16:30 Deewiant: write it in something else. =P 09:17:41 Deewiant, but I don't have too high hopes for this since I know cygwin's gdb can be whacky, and you said you edited funge-space.c to remove mmap... well that is where the actually file writing happens too, though that bit doesn't use mmap() 09:20:06 AnMaster: no help anyway 09:20:07 http://rafb.net/p/fzo95417.html 09:20:52 Deewiant, yeah, cygwin's gdb is whacky, not much I can do except note that using locally changed source is unsupported :/ 09:20:59 ues 09:21:00 yes 09:21:03 you cannot get paid phone support 09:21:11 or use any of the Unofficial cfunge Community Support Channels 09:21:23 for assistance with The Product. 09:21:28 (As specified by the EULA.) 09:21:31 tusho, very funny, but point is I need a good backtrace to be able to debug it 09:21:44 and eula is GPL so *shrug* 09:21:51 AnMaster: so you need a backtrace to do it but you won't do it anyway as it was locally changed? 09:21:52 i see. 09:21:58 tusho, the backtrace is broken 09:22:01 "give me a backtrace then proceed to watch me do nothing with it!" 09:22:10 and I would help if there was a good backtrace 09:22:15 even if it was locally changed 09:22:18 ...contradicting what you just said 09:22:50 but, now that it is locally changed and there is no good backtrace... well hard to debug, hard to reproduce 09:23:06 Deewiant, if you got a diff of your local changes I could take a look at them 09:24:29 if I could reproduce a bug with that here maybe I could help you make it work 09:24:37 AnMaster: just defining M_PI somewhere where it wasn't, removing the check for _POSIX_mmap_works or whatever, and removing 3DSP,FIXP,PERL,REXP,TERM 09:25:03 Deewiant, so you use mmap() still? just removing the sanity check for it? 09:25:07 yes 09:25:27 not like I'm going to rewrite your file loading code to get it to work :-P 09:25:48 Deewiant, then I'm totally lost. If you had removed mmap() and replaced it with something else then I could see that something could have caused it... but... now, huh? 09:26:03 I don't know 09:26:12 as for where was M_PI not defined? 09:26:18 can't remember, somewhere 09:26:21 here math.h define it 09:26:22 so :/ 09:26:32 not in cygwin evidently 09:26:42 Deewiant, think that is due to windows in fact 09:26:47 anyway will take a look at that 09:26:52 yes, it is 09:27:21 TURT.c I guess 09:28:19 Deewiant, but the bug you describe doesn't happen on any OS I have access to, so I'm sad to say that I'm unable to reproduce it :/ 09:28:33 hard to see what I could do :/ 09:29:28 Deewiant, you could run with -S to drop =, o and i 09:29:32 but... 09:29:51 Deewiant, as for PERL, why wouldn't it work under cygwin? 09:29:56 cygwin does have fork() 09:30:38 AnMaster: Is this where I get revenge for you whining about my blog design? 09:30:43 AnMaster: link error to __Exit 09:30:43 AnMaster: Portability!!!!!! 09:30:51 POOOOOOOOORTABILITY 09:30:57 Deewiant, ah 09:31:06 oh wait, you're portable to anything that supports C99 and POSIX right? 09:31:18 my design is portable to anything that supports CSS2 and some elements of CSS3. 09:31:25 it works without the CSS3 stuff too. 09:31:26 tusho, well that is fine 09:31:32 if you say that in the README :P 09:31:45 I make sure to not use GCC or glibc extensions 09:31:46 AnMaster: but 100% standards compliance is universal! 09:31:47 and so on 09:31:53 is it not?! you are shattering my world here 09:32:05 tusho, no it isn't, I never claimed to work on Mac OS 7 09:32:24 Deewiant: does ccbi run on macos7 09:32:34 tusho: haven't seen anyone try 09:32:34 Deewiant, does cygwin have _exit() then? 09:32:36 in theory, why not 09:32:39 if it doesn't have _Exit() 09:32:42 AnMaster: I don't know 09:33:09 Deewiant: what, but everyone who asks about something is an expert on it.[1][2] 09:33:09 References 09:33:09 --------------- 09:33:10 [1]: ais523 09:33:10 [2]: tusho 09:33:12 NAME 09:33:12 _exit, _Exit - terminate the calling process 09:33:15 CONFORMING TO 09:33:15 SVr4, POSIX.1-2001, 4.3BSD. The function _Exit() was introduced by C99. 09:33:15 okay, I'll stop irritating AnMaster now 09:33:25 wait no i won't 09:33:37 AnMaster: but you just said that standards-compliance wasn't universal... 09:33:39 so replacing it with _exit() could work 09:33:48 so why are you citing one 09:33:50 tusho, that is sadly the truth 09:33:57 tusho, I'm saying that you could use either 09:34:11 if you want it to work under cygwin 09:34:53 okay well i said that before you said that. 09:34:54 type faster :p 09:34:56 Deewiant, anyway, o uses normal file streams so I'm not sure what is the cause of stuff going wrong there. I'm unable to reproduce that type of error 09:35:37 Deewiant, cygwin use / not \ right? 09:35:59 tusho: if you really want to implement stuff for ccbi you can do it against the current version and I can update it myself for version 2 later 09:36:06 AnMaster: yes, of course 09:36:09 Deewiant: alright. any juicy fingerprints? 09:36:12 and Windows NT supports / as well anyway 09:36:23 i don't know a good list of them, so 09:36:28 and i think rcfunge98.com is down 09:36:34 it isn't 09:36:34 ah 09:36:35 'sback 09:36:36 not for me, anyway 09:36:57 Deewiant, well, I got no idea what could be wrong, if I had a bugzilla I would probably close with "WORKSFORME" or "NEEDMOREINFO" at this point :( 09:37:36 tusho: EXEC, FING, LONG, MACR, SETS, STCK, TRGR, from a quick look 09:37:45 EXEC sounds sexalicious. 09:37:47 * tusho checks 09:37:56 tusho, EXEC is just "variations of k in C maj" 09:38:07 ;P 09:38:11 ah. not very interesting from the looks either 09:38:14 tusho: you want to clone a mercurial branch to play with? 09:38:22 Deewiant: oh, sure 09:38:30 i was just going to sync the tarball manually with git but that sounds better :^) 09:38:47 Deewiant, if you manage to get an usable backtrace, or even better a nice patch, then I shall handle fix it as fast as I can (probably the weekend after) 09:38:58 tusho: hg clone http://tar.us.to:8000/ should work 09:38:58 until then: NEEDMOREINFO and WORKSFORME 09:39:00 :( 09:39:07 AnMaster: it works for you on cygwin? 09:39:11 the problem is "doesn't work on cygwin" 09:39:14 tusho, I don't have cygwin 09:39:15 WORKSFORME is inappropriate. 09:39:21 because you haven't even tested 09:39:33 tusho, right, NEEDMOREINFO LACKNEEDEDRESOURCESTOTESTONCYGWIN 09:39:37 I agree, more like NEEDMOREINFO or WONTFIX 09:39:46 Deewiant, WILLFIXWITHMOREINFO 09:39:49 AnMaster: couldn't you, ahh, just run qemu.. 09:39:59 tusho, I don't have any windows cd or license 09:40:04 AnMaster: pirate it. 09:40:11 tusho, I don't do that... 09:40:35 I don't break the law like that 09:40:36 AnMaster: is windows worth the money it costs. 09:40:36 really 09:40:39 :p 09:40:43 tusho, no, but I don't break the law 09:40:47 Deewiant: operation timed out 09:40:58 Deewiant, same as tusho here 09:41:02 meh 09:41:02 AnMaster: ok, i'll call you back post-copyright-reform. :D 09:41:15 tusho, hm? 09:41:28 Deewiant: any other location for the repo? 09:41:29 i guess not 09:41:44 tusho: I can give you the IP address instead or you can wait while I update the DNS record :-) 09:41:56 Deewiant, ip is fine 09:41:58 Deewiant: that sounds nice 09:42:02 88.114.230.95 09:42:16 alright, somebody worked 09:42:20 me 09:42:22 me 09:42:24 I did a clone 09:42:24 and tusho as well 09:42:37 oh, STCK looks incredibly trivial 09:42:38 i'll do that 09:42:42 heh 09:42:44 not that I can build it due to gdc and tango mess 09:43:08 Deewiant: can i install rebuild with dsss? 09:43:09 :P 09:43:10 just get DMD, what's so wrong with that :-P 09:43:16 tusho: rebuild is part of dsss 09:43:20 Deewiant: oh 09:43:22 if you have dsss, you have rebuild 09:43:23 i don't have dsss, it seems 09:43:24 lulz 09:43:26 heh 09:43:41 Deewiant, anyway as it stands now: cygwin is unsupported and NEEDMOREINFO | CANTFIXCURRENTLY 09:44:01 # dsss-0.75-gdc-mac-10.4.dmg <-- oh yeah baby 09:44:03 i hated compiling dsss 09:44:18 so cygwin issues is not valid for mycology results page 09:44:25 http://dsss.codu.org/ wow, gregorr uses the cherokee http server 09:44:25 since I don't currently support cygwin 09:44:29 however I accept patches 09:44:32 that's...even more obscure than nginx and lighttpd. 09:44:37 and good backtraces 09:44:37 and thttpd. 09:45:07 tusho, agreed, I used both lighttpd and thttpd (and apache of course) 09:45:15 I heard about nginx 09:45:20 but cherokee? huh? 09:45:29 cherokee looks like a neat project but meh 09:45:36 -!- optbot has set topic: the entire backlog of #esoteric: http://tunes.org/~nef/logs/esoteric | OK. 09:45:39 it seems neat but ... no reason to switch atm i think 09:45:55 AnMaster: the problem is cf_putc_unlocked 09:45:59 fputc works fine 09:46:02 rutian uses apache 'cause it's easy to spread out domains in different directories (the way we do it is: domain X administered by user Y goes in ~Y/www/X) 09:46:08 and it has Passangers and nifty stuff 09:46:27 Deewiant, interesting. I do call the needed locking and unlocking functions before and after 09:46:39 Deewiant, you should probably report this bug to the cygwin project I suspect 09:46:53 rutian is a pretty nifty little machine. 09:46:54 for now if you give me some thing to ifdef on I can work around it for cygwin 09:46:56 nicely organized and such 09:46:58 Deewiant, :) 09:47:09 and always zippy 09:47:33 % rebuild -rfccbi.rf 09:47:33 ccbi.d(13): module Exception cannot read file 'tango/core/Exception.d' 09:47:39 what's the dsss command to tangoify again 09:47:47 beats me, I don't use dsss 09:47:51 Deewiant, #ifdef what? to detect cygwin? 09:47:57 Deewiant: To make things easiest, get Rebuild and run rebuild -rfccbi.rf, passing any other options you wish (e.g. for optimization, -O -inline -release). If not, you’ll need to manually pass each source file to the compiler. 09:48:10 tusho: you don't have tango 09:48:20 Deewiant: i know 09:48:26 but there's a dsss command to download&install tango 09:48:29 so why paste that :-P 09:48:32 dsss net install tango? 09:48:34 ah 09:48:38 # define cf_putc_unlocked(x, y) putc_unlocked((x), (y)) 09:48:38 or 09:48:41 # define cf_putc_unlocked(x, y) putc((x), (y)) 09:48:42 Deewiant, ^ 09:48:53 depending on if the system claims to support unlocked IO 09:49:13 so does normal putc, not fputc work? 09:49:15 Deewiant: are those gdc problems still relevant 09:49:22 fputc() writes the character c, cast to an unsigned char, to stream. 09:49:22 tusho: probably yes 09:49:23 putc() is equivalent to fputc() except that it may be implemented as a macro which evaluates stream more than once. 09:49:26 GDC hasn't been updated in ages 09:49:27 Deewiant: i can't use dmd on os x :\ 09:49:46 tusho: well, like said, it's fine as long as you don't use 'o' :-P 09:49:49 lib/common/tango/core/Exception.d:71: internal compiler error: Abort trap 09:49:51 Deewiant, anyway, give me something to #ifdef on and I shall special case cygwin 09:49:54 /opt/local/var/db/dports/build/_opt_local_var_db_dports_sources_rsync.rsync.darwinports.org_dpupdate_dports_lang_gdc/work/gcc-4.1.2/gcc/d/dmd/toobj.c:417: failed assertion `classinfo->structsize == CLASSINFO_SIZE' 09:49:54 lib/common/tango/core/Exception.d:71: internal compiler error: Abort trap 09:49:56 the fuckkkkkkkkkkkkkkkkkkkkkkkkkkkk 09:50:03 tusho: yeah, hooray for GDC 09:50:12 ah: 09:50:12 gdc (GCC) 4.1.2 20070214 ( (gdc 0.23, using dmd 1.007)) 09:50:12 Copyright (C) 2006 Free Software Foundation, Inc. 09:50:13 tusho: do you have the latest SVN? 09:50:17 I AM EVER SO SLIGHTLY OUT OF DATE 09:50:17 evidently not 09:50:36 tusho, surely that is bad, you must of course use the last! ;P 09:50:46 tusho, use cvs head of your glibc! 09:50:48 * AnMaster ducks 09:51:01 AnMaster: django is worse 09:51:09 well 09:51:09 was 09:51:12 they released 1.0 recently 09:51:13 ah 09:51:18 but before that they were on 0.96 09:51:24 and the trunk was like 3 major releases ahead 09:51:29 production enterprise websites used the trunk 09:51:31 tusho, yes most projects tend to get a bit more stable by 1.0 09:51:38 and the actual experimental development was on other branches scattered about 09:51:59 Deewiant: will i have to reinstall dsss for the new gdc 09:52:01 AnMaster: I can't reproduce it trivially 09:52:04 tusho: I don't know 09:52:07 Deewiant: :D 09:52:20 probably not 09:52:24 Deewiant, hm... well give me an #ifdef for cygwin then, and I shall use normal putc for it 09:52:32 i hope gdc compiles easy... 09:52:33 unless it needs fputc not putc? 09:52:53 http://sourceforge.net/projects/gdcmac 09:52:55 june 1 2008 09:52:58 woop woop 09:53:43 Deewiant, other places in the code also use these functions so... well I need *something to use in #ifdef to detect cygwin* 09:54:02 Deewiant, want the GCC command to list predefined #defines? 09:54:22 echo | gcc -std=c99 -E -x c -dM - -o - 09:54:38 see if there is anything about cygwin there 09:54:41 outputs nothing :-D 09:54:48 #ifdef _CYGWIN_ 09:54:49 i think... 09:54:49 ok that's strange.... 09:54:51 osmething like that 09:55:00 but yeah, __CYGWIN__ 09:55:03 right 09:55:54 Deewiant, support.h, line 138 09:55:58 change it to 09:56:01 #if !defined(__CYGWIN__) && defined(_POSIX_THREAD_SAFE_FUNCTIONS) && (_POSIX_THREAD_SAFE_FUNCTIONS > 0) 09:56:09 AnMaster: btw, easier, echo | cpp -dM 09:56:12 that is, add "!defined(__CYGWIN__) &&" 09:56:15 Deewiant: what the fuck i have to switch gcc for gdc? 09:56:22 tusho: ? 09:56:32 Deewiant, hm ok 09:56:40 Deewiant: i hate how gcc language stuff requires compiling another gcc 09:56:44 instead of plugging it into your existing one 09:56:49 ah, yes, quite 09:57:02 is there a dmd for os x yet... 09:57:04 goddamn 09:57:06 no. 09:57:07 doubtful 09:57:18 this was easy the last time i did it 09:57:18 :| 09:57:53 AnMaster: btw, I get many warnings building cfunge 09:57:59 Deewiant, like? 09:58:04 mostly implicit declarations 09:58:11 and "warning: #warning is a GCC extension" 09:58:13 Deewiant, hm, I do include all headers I think 09:58:23 Deewiant, oh right.. that one 09:58:28 also redundant redeclarations 09:58:33 Deewiant, like? 09:58:33 warning: warning: warning: warning: warning: warning: warning: warning: warning: warning: warning: warning: warning: warning: warning: warning: warning: warning: warning: warning: warning: warning: warning: warning: warning: warning: warning: warning: warning: warning: warning: warning: warning: warning: warning: warning: warning: warning: warning: warning: stack overflow 09:58:34 and nested extern declarations 09:58:38 care to pastebin some 09:59:06 Deewiant, because I suspect many of those are in system headers, since cfunge turns on lots of warnings, that could lead to warnings about stuff in system headers 09:59:40 I certainly don't get those warnings here you see 09:59:56 I get *two* warnings, both in genx that tusho recommended 09:59:58 baah, the paste broke, http://rafb.net/p/RMegHR33.html 10:00:17 AnMaster: oooh so sorrrryyy! 10:00:21 wow. cool breakage 10:00:25 of the pastye 10:00:27 paste* 10:00:28 I resized the window 10:00:38 and expected that it wouldn't break copying from it 10:00:41 evidently I was wrong 10:00:57 I'll just redirect make to a file 10:01:01 D:/Progging/funge/others/cfunge-0.3.1/src/fingerprints/TURT/TURT.c: In function `GenerateCircle': 10:01:01 D:/Progging/funge/others/cfunge-0.3.1/src/fingerprints/TURT/TURT.c:475: warning: nested extern declaration of `snprintf' 10:01:01 :0: warning: redundant redeclaration of 'snprintf' 10:01:07 to me, that makes no sense 10:01:13 I do include the right header before 10:01:28 so how could it be nested extern decl 10:01:51 and if it was, it couldn't be redundant redeclaration 10:02:14 those two warnings are mutually exclusive 10:02:27 at least if you trust what the gcc man page says about them 10:02:58 Deewiant, somehow I suspect cygwin is doing something strange that breaks when you turn on lots of warning flags in GCC 10:03:23 more legible: http://rafb.net/p/RPxTt679.html 10:03:52 Deewiant, lets see here: 10:03:53 D:/Progging/funge/others/cfunge-0.3.1/lib/genx/genx.c: In function `storePrefix': 10:03:53 D:/Progging/funge/others/cfunge-0.3.1/lib/genx/genx.c:345: warning: implicit declaration of function `snprintf' 10:03:53 D:/Progging/funge/others/cfunge-0.3.1/lib/genx/genx.c:345: warning: nested extern declaration of `snprintf' 10:03:53 :0: warning: redundant redeclaration of 'snprintf' 10:04:02 I don't know, I'm just reporting 10:04:02 doesn't that seem rather strange to you? 10:04:10 just lets see if you agree with me 10:04:10 I don't know what "nested extern" means 10:04:15 Deewiant, means putting: 10:04:19 extern foo(); 10:04:22 inside a function 10:04:35 yeah, so snprintf is probably a macro 10:05:11 Deewiant, it would also happen if you haven't included the right header. *but* I do include the right header. and it can't also be "redundant redeclaration" then 10:05:21 some of those warnings are mutually exclusive 10:05:56 snprintf needs #include 10:05:59 according to man page 10:06:03 AnMaster: some SOCK/SCKE fails in mycology 10:06:04 i talk in a row a lot 10:06:08 i wonder if it annoys anyone 10:06:13 does it annoy any of you? 10:06:14 Deewiant, not here hm 10:06:19 i've gone 10-15 interrupted i think 10:06:21 Deewiant, with what does it fail on cygwin? 10:06:23 average 3-4 10:06:24 BAD: couldn't close original socket with K 10:06:29 and then in SCKE 10:06:29 BAD: need working B to test P 10:06:49 Deewiant, oh? that's odd... let me see 10:07:04 if (shutdown(sockets[s]->fd, SHUT_RDWR) == -1) { 10:07:04 goto error; 10:07:04 } 10:07:09 AnMaster: what Mycology version do you have? 10:07:09 if (close(sockets[s]->fd) == -1) { 10:07:09 goto error; 10:07:09 } 10:07:42 Deewiant, the one after you fixed the lock up in rc/funge and cfunge 10:07:49 so what version 10:07:56 Deewiant, where does it say? 10:08:03 just the latest date in readme.txt 10:08:07 a sec 10:08:25 2008-08-30 - Fixed the case where SCKE is included in SOCK. 10:08:33 alright 10:08:38 too old? 10:08:42 new enough 10:08:50 newest released, I think 10:09:42 Deewiant, src/fingerprints/SOCK/SOCK.c, try adding something perror("K failed:" in FingerSOCKkill() after line 252 10:09:47 error: 10:09:47 FreeHandle(s); 10:09:49 between those lines 10:09:50 that is 10:10:12 would say what went wrong at least with K 10:10:51 oh and you have to add an include for stdio.h too to that file Deewiant 10:11:09 K failed:: Transport endpoint is not connected 10:11:40 hm 10:12:07 Deewiant, ok, that's strange... you see there is both a shutdown() and a close() call above 10:12:16 B failed:: Address already in use 10:12:24 Deewiant, try adding some code to print to see which of those that fails 10:12:26 that makes sense, because the K failed 10:12:33 yes sure does 10:13:08 possibly something doesn't like shutdown on a socket that didn't connect 10:13:08 AnMaster: shutdown fails 10:13:22 Deewiant, try to make it keep going to close after shutdown instead 10:13:30 I think that may be the bug 10:13:52 so just remove the if bit over shutdown 10:13:59 and ignore the return value of shutdown() 10:14:00 works fine if I comment out the goto error after shutdown 10:14:31 AnMaster: like the man page says 10:14:31 ENOTCONN 10:14:32 The specified socket is not connected. 10:14:37 Deewiant, right, *fixes that*, seems like linux, *bsd and OS X accepts shutdown on a non-connected socket though 10:14:41 since it doesn't fail here 10:14:47 * AnMaster fixes 10:14:51 so it looks like cygwin is more POSIX-compliant here ;-) 10:15:03 Deewiant, and yes I fixed it locally, will commit soon 10:15:12 pwnt 10:15:14 :D 10:15:18 tusho, not really 10:15:28 if it is possible to track down a bug I'm fine with doing so 10:15:33 AnMaster: note the ":D" 10:15:55 AnMaster: append a "&& errno != ENOTCONN" or what? 10:15:59 or just call the close always 10:16:21 Deewiant, I'm looking at man page atm 10:16:34 just calling close anyway probably 10:17:09 EBADF, ENOTCONN and ENOTSOCK will not cause any kind of horrible failure as far as I can see 10:17:49 well whatever, that's that 10:18:34 Deewiant, right. and what about the compile failure in 3DSP? Any idea of cause? 10:18:55 I think it was *l functions being used 10:18:57 sinl etc 10:19:03 link errors resulted 10:19:18 Deewiant, ah it doesn't implement those? 10:19:29 evidently not if they don't link 10:19:50 well, they are part of C99. But since older FreeBSD also miss them I was planning to add some sort of detection and fallback to non-l versions 10:19:57 so that will be in next version too 10:21:49 "We are happy to announce that W3C has integrated a version of HTML 5 conformance checker into a beta instance of the W3C Markup validator. That will help us to detect bugs, improve the user interface, and benefit from the large W3C communities" 10:21:50 woop woop 10:22:47 oh, nice 10:23:17 http://qa-dev.w3.org/wmvs/HEAD/check?uri=http%3A%2F%2Feso-std.org%2F&charset=(detect+automatically)&doctype=Inline&group=0 10:23:18 woop woop 10:23:29 no ugly button yet, though :P 10:24:35 Deewiant, ok, those fixes are in, the l vs. non-l math ones will probably be done later today 10:24:48 Deewiant, anyway now you can put that on the mycology results page :) 10:25:02 today I think I will finally buy tusho.net 10:25:11 and go "woop woop" again 10:25:17 Deewiant, these were in r406. btw 10:25:21 trunk 10:25:58 revision-id: anmaster@envbot.org-20080907092316-eczkdchbs861mm1s 10:26:43 -!- oerjan has quit ("meep meep"). 10:26:48 AnMaster: nah, I need to run it in an environment where everything works 10:27:11 Deewiant, well, some of those you may never be able to get working under cygwin 10:27:16 at least some of the fingerprints 10:27:23 exactly, so I need to run it on a linux machine 10:27:29 right 10:27:31 just like RC/Funge, so that I can get TURT results 10:27:58 Deewiant, what ones was it you had disabled on cygwin now again? 10:28:16 can't remember, see the scrollback 10:28:45 3DSP,FIXP,PERL,REXP,TERM right 10:29:17 3DSP, FIXP, well those use sinl and such. PERL was _Exit. REXP and TERM.... not sure what could go wrong on cygwin in those 10:31:12 actually, 3DSP does not use sinl? 10:32:00 Deewiant, I'm at loss what was the issue in 3DSP 10:32:48 -!- Judofyr has joined. 10:33:10 ok wtf happened to http://eso-std.org/ :| 10:33:32 Parse error: syntax error, unexpected T_STRING, expecting ';' in /home/tusho/www/eso-std.org/user/themes/tusho-blogs/header.php on line 1 10:33:38 looks like a php error 10:33:45 thank you AnMaster, I kind of knew that. 10:33:49 :| 10:33:56 * AnMaster is now known as CaptainObvious 10:34:31 -!- GreaseMonkey has quit ("Jesus loves you"). 10:34:33 tusho, possibly: echo $a "foo" could cause it, that is a missing . between variable and string 10:34:49 nope, no concatenation going on here 10:34:51 it's just a template file 10:34:53 not a code file 10:34:57 so just simple function calls and loops etc 10:35:09 here, I can even paste the whole thing 10:35:31 (the reason it is 1 line is because I automatically compact it before uploading, probably a placebo but it feels more compact... might stop doing that :P) 10:35:33 <?php if ($request->display_entry && isset($post)) { echo $post->title;} else { Options::out('title');} ?>
10:36:14 inserting newlines would help locate the error 10:36:21 yes, yes it would :p 10:36:21
10:36:23 this time 10:36:26 yes 10:36:28 that's the full thing then 10:36:31 anyway 10:36:35 I have a non-compacted version here 10:36:39 and no obvious causes for the error 10:36:44 anyway, try the non-compacted one on server 10:36:48 see what line it fails on 10:36:54 yea probably a good idea 10:36:59 site::out_url() doesn't have a ";", does the next statement just concatenate after it with multiple 's? 10:37:31 no 10:37:41 right of course not that would make no sense 10:37:51 oh 10:37:53 duh 10:37:53 10:37:59 tusho, lets see, assuming \n newlines, you save 1 byte / line per file by removing newlines? 10:38:00 should be foreach ($pages as $page) 10:38:02 >.< 10:38:10 AnMaster: well, no, much more 10:38:14 since, newlines in loops duplicate 10:38:19 and also it strips out general whitespace to 10:38:23 tusho, ah. hm 10:38:27 before the stuff was like 3 pages of lines not longer than 30 chars 10:38:34 now it's 3 lines of ... well, a lot, but quite a bit less than before 10:38:54 i'm gzipping the output on there too, it's still not as fast as some sites but it's very zippy compared to most 10:39:11 and it's just a 25 line shell script to compress&upload 10:39:18 tusho, for 1100 baud modems? ;) 10:39:23 :D 10:39:28 I suspect that it really doesn't matter 10:39:30 hmm 10:39:34 $pages isn't predefined...odd 10:39:35 you are over optimising 10:39:35 :P 10:39:53 AnMaster: actually i tested it when i made that script and it was a little faster 10:39:54 tusho, so if you ever claim that I am, you are just a hypocrite :P 10:39:58 you should have seen the whitespace it generated without it 10:40:01 i mean, literally, tons 10:40:07 newlines and like 5 spaces in a row all over the place 10:40:11 tusho, did you do profiling? 10:40:14 it's not over-optimizing, it's making the output sane ;) 10:40:21 AnMaster: yes. i reloaded a page and reloaded after compression. 10:40:24 it was faster. 10:40:31 tusho, how did you measure? 10:40:35 by wall clock? 10:40:38 by feeling? 10:40:39 AnMaster: human perception. 10:40:39 :) 10:40:46 not very exact 10:40:48 it's 24 lines of shell script i don't really care if it's placebo 10:40:57 hahah 10:41:21 it's just some trivial sed incantations that 1. makes the output of the templates look nicer (that whitespace abuse is really ugly) 2. save me a miniscule amount of bandwidth & make the site a miniscule amount faster & save a miniscule amount of disc 10:41:23 *disk 10:41:30 sounds like a good deal to me 10:42:07 hehe 10:42:28 tusho, and make finding cause of error a lot harder 10:42:38 "Notice: Undefined variable: pages in user/themes/tusho-blogs/header.php line 1" yes 10:42:45 uh, that one was easy 10:42:52 i thought $pages was predefined for themes 10:42:53 turns not 10:43:00 AnMaster: i just have to comment out the compression 10:43:09 agreed 10:43:14 besides, the php calls are trivial that the only time i get it wrong are when i do things like that for thing 10:43:22 because php's syntax is stupid and my brain wants it to not be :D 10:43:27 AnMaster: i meant for debugging 10:43:36 tusho, what do you think about the source of http://rage.kuonet.org/~anmaster/cfunge/ 10:43:36 ? 10:43:44 AnMaster: it's xhtml 1.1 10:43:45 :) 10:43:48 or rather, :( 10:43:49 tusho, apart from that? 10:43:57 yeah, it's fine 10:44:08 tusho, do you think it is generated or hand written? 10:44:09 though 10:44:11 why "? 10:44:18 AnMaster: knowing you hand written 10:44:31 tusho, well hand written since it is a rather simple page 10:44:34 I don't need php 10:44:36 or such 10:44:39 for such a siumple 10:44:48 tusho, why not "? 10:45:01 AnMaster: it makes reading&editing the source more obscure and is unneeded when you coudl just use " 10:45:13 hm true 10:45:13 although " isn't a quotation mark. 10:45:19 “ and ” are :P 10:45:40 tusho, if I want good typography I would use LaTeX not html :P 10:46:18 AnMaster: another comment on that page is that I'm not a fan of the background colour 10:46:20 but that's it 10:46:32 tusho, you mean off-white color? 10:46:38 AnMaster: yes, it makes it harder to read the text for me 10:46:55 tusho, well I need non-white since it is hard to read for some if it is white 10:47:01 forgot the name of the condition 10:47:02 AnMaster: why specify it at all 10:47:10 those people will have it set in their preferences 10:47:24 additionally your link colours are almost the browsers default, the only thing you really need is: 10:47:29 10:47:35 well 10:47:41 img{border:0} 10:47:41 too 10:47:42 you might want a