←2008-07-24 2008-07-25 2008-07-26→ ↑2008 ↑all
00:03:49 -!- lilja has quit ("KVIrc 3.2.0 'Realia'").
00:36:38 -!- olsner has quit ("Leaving").
00:42:32 -!- pikhq has joined.
00:43:12 -!- psygnisfive has quit (Client Quit).
00:44:09 -!- tusho has quit.
01:35:38 -!- zEEbe has joined.
01:39:48 -!- psygnisfive has joined.
01:40:08 <psygnisfive> boo!
02:00:28 <zEEbe> booo!
02:00:35 -!- zEEbe has changed nick to zeeb.
02:40:04 -!- zeeb has left (?).
02:51:49 -!- Sgeo has joined.
03:05:17 -!- Sgeo_ has quit (Read error: 110 (Connection timed out)).
04:08:22 -!- Corun has quit ("This computer has gone to sleep").
04:38:15 -!- GreaseMonkey has joined.
04:46:10 -!- pikhq has quit ("leaving").
05:47:30 -!- GreaseMonkey has quit ("saving RAM for nexuiz").
06:35:37 -!- GreaseMonkey has joined.
07:15:31 -!- Sgeo has quit (Read error: 104 (Connection reset by peer)).
07:17:24 -!- Sgeo has joined.
07:40:00 -!- dbc has quit (Client Quit).
07:40:29 -!- dbc has joined.
07:59:59 -!- clog has quit (ended).
08:00:00 -!- clog has joined.
08:12:03 -!- olsner has joined.
08:18:57 <psygnisfive> lalala
09:12:22 -!- olsner has quit ("Leaving").
09:23:52 <AnMaster> morning
09:26:56 <psygnisfive> hey
09:26:59 <psygnisfive> later today
09:27:04 <psygnisfive> im teaching you about formal languages
09:38:58 <AnMaster> psygnisfive, no thanks
09:39:00 <AnMaster> too busy
09:39:32 <psygnisfive> :P
10:23:07 -!- GreaseMonkey has quit ("HydraIRC -> http://google.com <- Go find something better").
10:23:12 -!- jemarch has joined.
10:24:13 <jemarch> oh my god, is the people from esoteric.sange.fi there?
10:25:50 <AnMaster> jemarch, what?
10:25:56 <AnMaster> who are they?
10:26:04 <jemarch> heh :)
10:27:45 <AnMaster> ~/funges/interpreters/rcfunge $ ./funge -Y -md ~/src/cfunge/trunk/mycology/mycology.b98
10:27:45 <AnMaster> Segmentation fault
10:27:47 <AnMaster> hrm
10:28:01 <AnMaster> need to reach Mike Reily
10:28:04 <AnMaster> err spelling
10:37:24 -!- tusho has joined.
10:38:07 <AnMaster> hi tusho!
10:38:23 <tusho> hi AnMaster
10:38:42 <AnMaster> it is a pitty RC Funge doesn't have a public repo, otherwise I might have helped with some patches
10:39:14 <tusho> some people like working undisturbed
10:39:16 -!- tusho has quit (Client Quit).
10:39:26 <AnMaster> why the quit...
10:39:35 -!- tusho has joined.
10:39:38 <AnMaster> wb
10:39:39 <AnMaster> why the quit?
10:40:16 <tusho> testing stuff
10:42:09 <AnMaster> k
11:04:24 <AnMaster> Deewiant, should 0k 7 skip over next instruction or next char?
11:04:30 <AnMaster> next instruction seem sanest
11:13:21 -!- RedDak has joined.
11:25:52 -!- Sgeo has quit (Remote closed the connection).
11:36:54 <AnMaster> Loaded STRN: testing I. Please input: åäö
11:36:55 <AnMaster> UNDEF: got "å"
11:36:55 <AnMaster> hrrm
12:00:20 <Deewiant> instruction seems sanest to me too, although it's bound to be insane in some case ;-)
12:07:00 <AnMaster> Deewiant, indeed
12:07:04 <AnMaster> k is insane
12:07:18 <AnMaster> Deewiant, it is really an optimized version of:
12:07:28 <AnMaster> err
12:07:31 <AnMaster> can't paste that code
12:07:42 <AnMaster> it is an optimized version of a simple loop you could do anyway
14:11:19 -!- pikhq has joined.
14:14:04 -!- RedDak has quit (Read error: 104 (Connection reset by peer)).
14:30:47 <AnMaster> hi pikhq
14:30:54 <pikhq> Hi.
15:12:02 <pikhq> http://en.wikinews.org/wiki/China_to_establish_protest_zones_for_Olympics
15:12:14 <pikhq> How. . . Depressing.
15:12:23 <pikhq> The US and China seem to have similar stances on protests now.
15:12:25 <pikhq> :(
15:44:48 -!- MikeRiley has joined.
16:00:27 <AnMaster> MikeRiley, hi, your last release segfaults when loading mycology for me
16:00:32 <AnMaster> using -md -Y
16:00:40 <AnMaster> want a backtrace?
16:00:43 <MikeRiley> increase the stack size...
16:00:54 <MikeRiley> i am running with -i 500000 and -c 200000
16:00:56 <AnMaster> MikeRiley, um this happens before any output
16:01:36 <MikeRiley> here is my full command line:
16:01:37 <MikeRiley> time ../../funge -i 500000 -c 200000 -Y -md -w mycology.b98
16:01:49 <AnMaster> what does -w do?
16:01:54 <MikeRiley> enables warnings...
16:02:23 <AnMaster> ah right
16:02:57 <MikeRiley> mostly issues warnings when an invalid instruction is specified...
16:03:25 <AnMaster> ah yes your interpreter lacks a --help or such
16:03:28 <AnMaster> (cfunge got -h)
16:03:43 <MikeRiley> yeah, i noticed that the other day!! eheheheheh i should add one....
16:03:51 <AnMaster> would be useful yet
16:03:53 <AnMaster> yes*
16:04:00 <MikeRiley> yep,,,would be usefull...
16:04:15 <MikeRiley> want to add help to the debugger too,,,when i was trying to use it yesterday,,,could not remember the commands!! eheheheheheeh
16:05:36 <AnMaster> well cfunge doesn't have a debugger, I did have some plans for a remote debugging protocol
16:05:43 <AnMaster> so you could use an external debugger
16:05:49 <AnMaster> for several different interpreters
16:06:19 <AnMaster> never got around to implementing it
16:06:41 <MikeRiley> that would be usefull....
16:07:05 <MikeRiley> down to only 4 bads left....and a couple of them might be related to the broken k in mycology,,,,or else should be considered UNDEF rahter than BAD
16:07:30 <MikeRiley> i mainly wrote my debugger in order to help debug the interpreter more so than debugging funge programs...
16:08:22 <AnMaster> MikeRiley, there are no bads due to the k change for me except those few I mentioned early on
16:08:28 <AnMaster> all before the fingerprints
16:08:39 <MikeRiley> i can accept that....
16:08:51 <AnMaster> MikeRiley, oh, for debugging interpreter I use gdb
16:08:53 <MikeRiley> my interest in the moment is in testing FILE
16:08:55 <AnMaster> with some macors
16:08:59 <AnMaster> in the .gdbinit file
16:09:43 <MikeRiley> it appears to test G on EOF, but the interpreter says that G is only run twice,,,getting results each time,,,and does not show a third attempt to use G to get EOF...
16:10:10 <AnMaster> um G as in the fgets()?
16:10:19 <AnMaster> or rather get line
16:10:29 <MikeRiley> yes....
16:10:55 <MikeRiley> i see it execute once and get 'bar\n'
16:10:59 <MikeRiley> then again to get 'baz'
16:11:08 <AnMaster> um you shouldn't include the trailing \n iirc
16:11:10 <MikeRiley> and then the interpreter says that G is not run again after this...
16:11:27 <AnMaster> at least I strip any trailing \n from the string I get
16:12:16 <MikeRiley> not sure if i want to strip line endings or not....
16:12:29 <MikeRiley> technically,,,since the spec says like c fgets, and fgets keeps the line endings....
16:12:37 <MikeRiley> would be correct to specs to keep the line endings...
16:12:45 <MikeRiley> even tho i would prefer them not to be there...
16:13:26 -!- puzzlet_ has quit (Read error: 60 (Operation timed out)).
16:18:26 <MikeRiley> found problem in FILE and fixed,,,,down to only 3 BADs now...
16:19:56 <MikeRiley> did you get Rc/Funge to run??? or is it still giving you a sement fault??
16:24:41 <AnMaster> bbiab food
16:30:39 -!- Hiato has joined.
16:44:04 -!- Corun has joined.
16:54:32 -!- Hiato has quit ("Leaving.").
17:03:33 <AnMaster> MikeRiley, yes I got it running
17:03:35 <AnMaster> but it is rather slow
17:03:45 <AnMaster> maybe due to the huge memory needed to be allocated
17:10:41 -!- Corun has quit ("This computer has gone to sleep").
17:15:09 <MikeRiley> hmmmm,,,,does not run slow for me....
17:17:04 <AnMaster> MikeRiley, it also hangs in IMAP
17:17:09 <AnMaster> GOOD: '5O unmapped 5 to push 5
17:17:14 <AnMaster> MikeRiley, that seems odd :P
17:17:26 <MikeRiley> that does seem odd....
17:17:29 -!- Corun has joined.
17:17:40 <MikeRiley> which version are you actually running???
17:17:54 <AnMaster> v1.08
17:18:21 <AnMaster> MikeRiley, I got no idea if it matters but this is 64-bit
17:18:34 <AnMaster> so if RC/Funge isn't 64-bit clean, maybe?
17:19:05 <MikeRiley> it is not 64-bit clean,,,,i am running 64-bit here as well,,,but the only problem i know of is in FPDP....
17:19:39 <AnMaster> ah
17:20:00 <AnMaster> MikeRiley, um wait, how does "not 64-bit clean" cause an issue in FPDP?
17:20:06 <AnMaster> I can't see how that would happen
17:20:28 <AnMaster> typedef union u_doubleint {
17:20:28 <AnMaster> double d;
17:20:28 <AnMaster> struct { int32_t high; int32_t low; } i;
17:20:28 <AnMaster> } doubleint;
17:20:30 -!- jemarch has quit ("ERC Version 5.0 (CVS) $Revision: $ (IRC client for Emacs)").
17:20:32 <AnMaster> that is my union for it
17:21:13 <MikeRiley> yeah,,,i do not either,,,,but on this machine everythying in FPDP comes up zeros.....my machine at home (32-bit) works correclty...
17:22:43 <AnMaster> MikeRiley, that's strange
17:22:55 <AnMaster> MikeRiley, is the code the same as in 1.08?
17:23:01 <AnMaster> I want to take a look at it
17:23:47 <MikeRiley> hold on a few minutes,,,and will put 1.0.9 where can get at it....
17:23:48 <AnMaster> MikeRiley, not odd at all
17:23:49 <AnMaster> union {
17:23:50 <AnMaster> double f;
17:23:50 <AnMaster> struct { long int l; long int h; } i;
17:23:50 <AnMaster> } FpDp;
17:23:58 <AnMaster> long int == 64-bits on x86_64
17:24:02 <AnMaster> not 32-bits
17:24:15 <AnMaster> on x86 long int is 32-bit and long long is 64-bit
17:24:27 <AnMaster> on x86_64 long int and long long are both 64-bit
17:24:37 <AnMaster> for both those platforms int is 32-bit
17:24:46 <AnMaster> however that may not be true on some other systems
17:25:03 <AnMaster> so using the int32_t from stdint.h in C99 is the most sane way
17:25:17 <AnMaster> but that means you have to depend on C99
17:25:24 <AnMaster> which few compilers implement so far
17:25:29 <AnMaster> GCC does it, MSVC doesn't
17:25:37 <AnMaster> icc may do it, not sure
17:25:41 <AnMaster> pcc probably not
17:26:18 <AnMaster> MikeRiley, on and double is *always* 64-bits
17:26:24 <AnMaster> and float always 32-bits
17:26:55 <AnMaster> union {
17:26:55 <AnMaster> float f;
17:26:55 <AnMaster> long int i;
17:26:55 <AnMaster> } FpSp;
17:27:03 <AnMaster> that wouldn't work on big endian systems
17:27:16 <AnMaster> as you would get wrong part of the long int into the float there
17:27:26 <MikeRiley> you can get the latest sources from: http://www.elf-emulation.com/rcfunge_dev.tgz
17:27:31 <Deewiant> I think only something by Sun, forget what it's called, supports C99 fully
17:27:51 <AnMaster> Deewiant, well GCC aims for it, not complete yet, but will be at some point
17:28:01 <Deewiant> sure, a couple aim for it
17:28:09 <Deewiant> but I think only that one has a full implementation
17:28:11 <AnMaster> tar: ./rcfunge: time stamp 2008-07-25 18:35:27 is 440.78867794 s in the future
17:28:14 <AnMaster> MikeRiley, interesting
17:28:20 <AnMaster> may want to set your clock :)
17:28:34 <AnMaster> I set mine using ntp so it is exact
17:29:12 <MikeRiley> i never did bother about correct time on my machines!! eheheheheheeheheh
17:30:17 <AnMaster> ext/file.c:86: warning: implicit declaration of function 'strlen'
17:30:24 <AnMaster> MikeRiley, that means you forgot to include <string.h>
17:30:44 <AnMaster> same in debugger.c
17:31:25 <AnMaster> mterm.o: In function `Input':
17:31:25 <AnMaster> mterm.c:(.text+0x39): warning: the `gets' function is dangerous and should not be used.
17:31:30 <AnMaster> is another warning I get
17:34:02 <fizzie> Huh? In no fashion is double *always* 64 bits (or float 32). They are for IEEE 754 floating-point arithmetics, but the C standard doesn't require that.
17:34:11 <AnMaster> fizzie, hm
17:34:29 <fizzie> For all reasonable platforms, sure, but that's still different than "always".
17:34:29 <AnMaster> fizzie, so how to you get an integer type of the same size as a float?
17:34:32 <AnMaster> or double
17:34:45 <Deewiant> you don't?
17:34:47 <AnMaster> I guess you need autoconf hell.
17:35:14 <AnMaster> IEEE 754 floating-point <-- right, I'll note down in my README that the platform need such floating point
17:35:26 <AnMaster> fizzie, I think C99 mandates that btw
17:35:28 <AnMaster> not 100% sure
17:35:47 <MikeRiley> yeah, there are a few places where string.h was missing...
17:35:52 <MikeRiley> been adding them as i run into them...
17:36:08 <AnMaster> MikeRiley, what warning flags do you use?
17:36:16 <AnMaster> for cfunge I use a LARGE set of warning flags
17:36:17 <MikeRiley> for what??
17:36:32 <AnMaster> as in -Wall -Wextra -Wformat=2 ...
17:36:38 <MikeRiley> i just use -Wall
17:36:46 <AnMaster> -Wall -Wextra -pedantic -Wpointer-arith -Wimplicit -Wnested-externs -Wcast-align -Wcast-qual -Wbad-function-cast -Wstrict-prototypes -Wmissing-prototypes -Wmissing-declarations -Wparentheses -Wundef -Wpacked -Wstrict-aliasing=2 -Wformat=2 -Winline -Wmissing-noreturn -Wmissing-format-attribute -Wdeclaration-after-statement -Wunused-function -Wunused-label -Wunused-value -Wunused-variable -Wwrite-strings
17:36:49 <AnMaster> for cfunge :P
17:37:13 <Deewiant> and a set of optimization flags which is twice as long?
17:37:17 <fizzie> C99 doesn't require sensible floating-point either, although it does specify a __STDC_IEC_559__ macro that can be used for checking if it's there.
17:37:20 <AnMaster> Deewiant, that I leave to the user
17:37:46 <Deewiant> AnMaster: and besides, with cmake, doesn't specifying those mean that you restrict the platform to GCC?
17:37:49 <fizzie> If you don't mind the holes, I guess you can just use some suitable union of a floating-point and integer type that are likely to be of same size.
17:37:59 <AnMaster> fizzie, all I can find is "Annex F (normative): IEC 60559 floating-point arithmetic"
17:38:16 <AnMaster> Deewiant, I only use them if GCC is detected
17:38:27 <AnMaster> without GCC I only assume that -DFOO will define FOO
17:38:33 <AnMaster> which I think is required
17:38:40 <AnMaster> either by POSIX or the C standard
17:38:41 <Deewiant> isn't there a set_define or some such
17:38:45 <fizzie> Yes, and "An implementation that defines __STDC_IEC_559__ shall conform to the specifications in this annex", which implies that you can have a C99 implementation that does not conform.
17:38:53 <AnMaster> fizzie, ah
17:39:17 <AnMaster> another check with #ifdef and erroring out in other words
17:41:23 <fizzie> I would probably just make it work even if the types aren't exactly the same size, unless that's for some reason not possible. I'm sure there are implementations with correctly sized floating-point types, but which do not define that macro because they don't bother implementing all peculiarities of the actual floating-point ops.
17:41:45 <AnMaster> fizzie, I need an union that matches size exactly
17:41:56 <AnMaster> typedef union u_doubleint {
17:41:56 <AnMaster> double d;
17:41:56 <AnMaster> struct { int32_t high; int32_t low; } i;
17:41:56 <AnMaster> } doubleint;
17:42:03 <AnMaster> need to store that on the funge stack
17:42:15 <AnMaster> so the sizes *need* to match
17:43:56 <fizzie> Why can't you just leave holes in the stack if the sizes happen to be perverse?
17:44:11 <AnMaster> fizzie, what if they happen to be bigger?
17:44:18 <AnMaster> also it needs to fit in exactly two cells
17:44:24 <AnMaster> and one cell for float
17:44:30 <AnMaster> I can accept wasted space
17:44:36 <AnMaster> but not the other way around
17:44:56 <fizzie> Uh, well: the size of that union is the size of the largest member.
17:45:30 <AnMaster> fizzie, anyway you saw what happened to RC/Funge when the integer version was too large
17:45:36 <AnMaster> it used long int by mistake
17:45:44 <AnMaster> so it didn't work on 64-bit platforms
17:45:48 <AnMaster> at all
17:45:51 <fizzie> Actually, I didn't: I just jumped into the conversation without reading the logs.
17:45:57 <AnMaster> ah
17:46:43 <fizzie> Anyway, if you make your stack cells have that union type, they will obviously be big enough. Would be pretty tricky if you need to split it in two cells, though.
17:46:49 <AnMaster> err
17:46:51 <AnMaster> that doesn't work
17:47:00 <AnMaster> it needs to be same size as the funge-space
17:47:04 <AnMaster> they can't be unions
17:47:09 <AnMaster> this is just for a fingerprint duh
17:47:37 <AnMaster> fizzie, and yes the FPDP specs mandates that I need to split a double in 2 cells
17:48:18 <fizzie> Heh. Well, you can't split a double in 2 cells if it's too big, no matter what you do.
17:48:25 <AnMaster> indeed
17:48:36 <AnMaster> fizzie, and the cells need to be a specific size
17:48:40 <AnMaster> in various other places
17:48:59 -!- Corun has quit ("This computer has gone to sleep").
17:49:01 <AnMaster> 32-bit or 64-bit, depending on compile time options
17:49:21 <MikeRiley> could always change cell size to 64-bit...
17:49:37 <AnMaster> MikeRiley, well I have that as compile time option already
17:49:45 <AnMaster> but it is much slower on 32-bit platforms
17:49:53 <AnMaster> and a bit slower on 64-bit platforms too
17:49:57 <MikeRiley> other things in the code might need to change tho...
17:50:00 <AnMaster> less data fit in cache you see :P
17:50:04 <MikeRiley> yep....
17:50:12 <AnMaster> and this hurts on my cpu
17:50:14 <AnMaster> cpu[1 x AMD Sempron(tm) Processor 3300+ (AuthenticAMD) @ 1000MHz w/ 128 KB L2 Cache]
17:50:20 <MikeRiley> may have to redesign how that module is supposed to work...
17:50:21 <AnMaster> (dynamic cpu speed, can run at 2 GHz)
17:50:32 <AnMaster> MikeRiley, anyway it makes sense to have it the way it is now
17:50:38 <AnMaster> just require sane floating point
17:50:45 <MikeRiley> yeah,,,,,
17:50:51 <AnMaster> oh and fix the issue with long int mess up :P
17:50:58 <AnMaster> but you don't depend on C99 do you?
17:51:05 <MikeRiley> no...i do not...
17:51:12 <AnMaster> could be a bit harder to check floating point then, at least just using a plain make file
17:51:27 <AnMaster> with autoconf or cmake you can check size at compile time
17:51:30 <MikeRiley> probalby should stop using things like int, long, etc and go to int32_t, etc...
17:51:40 <AnMaster> MikeRiley, int32_t is C99 though
17:51:44 <AnMaster> defined in stdint.h
17:52:49 <MikeRiley> oh.......
17:52:50 <AnMaster> MikeRiley, but there are ways to find the type yourself using autoconf or cmake or such
17:53:02 <AnMaster> but that requires a configure script or a cmake file
17:53:05 <AnMaster> or similar
17:53:27 <AnMaster> a plain makefile won't work any longer
17:53:50 <AnMaster> (anyway converting rcfunge to either should be pretty easy, I could convert it to cmake if you want
17:54:16 <MikeRiley> what is cmake??? i have never used it....
17:54:40 <AnMaster> MikeRiley, it is a tool to generate a makefile, somewhat like the ./configure stuff, but another software
17:54:40 <MikeRiley> down to 3 Bads now,,,and 1 probably should be UNDEF rather than BAD....
17:54:50 <AnMaster> configure would be autoconf
17:54:53 <MikeRiley> i will have to take a look at it....
17:54:57 <AnMaster> while cmake is an alternative software
17:55:00 <MikeRiley> autoconf i am familiar with...
17:55:02 <AnMaster> cfunge uses cmake
17:55:07 <AnMaster> MikeRiley, well use autoconf then
17:55:12 <AnMaster> use what you like :)
17:55:24 <AnMaster> I find cmake a lot easier to work with though
17:56:41 <MikeRiley> will have to look at cmake,,,never cared much for autoconf...
17:57:11 <AnMaster> just one thing I don't like with cmake
17:57:28 <AnMaster> it's IF/ELSE/ENDIF syntax is too verbose
17:57:38 <AnMaster> OPTION(CONCURRENT_FUNGE "Enable support for concurrent funge." ON)
17:57:38 <AnMaster> IF(CONCURRENT_FUNGE)
17:57:41 <AnMaster> for example
17:57:51 <AnMaster> you have to put the full condition in when closing it too
17:58:03 <AnMaster> that's the only think I don't really like with cmake
17:59:56 <tusho> i like scons because it's like cmake, except it doesn't try and piggyback on the deprecated make-and-friends
18:00:01 <tusho> and it doesn't have that god-awful syntax
18:00:03 <tusho> (uses python)
18:00:07 <tusho> if concurrent_funge:
18:00:09 <tusho> ...
18:00:10 <tusho> vs
18:00:13 <tusho> IF(CONCURRENT_FUNGE)
18:00:13 <tusho> ...
18:00:16 <tusho> ENDIF(CONCURRENT_FUNGE)
18:00:34 <tusho> it'd be nice if someone ported scons to ruby
18:00:39 <AnMaster> tusho, yes right, I don't like scons because it seems less actively maintained
18:00:40 <tusho> shrug
18:00:51 <tusho> AnMaster: why does it need to be?
18:00:55 <tusho> it works great and i've never found a bug
18:00:56 <AnMaster> I don't want to end up depending on on unmaintained software
18:01:07 <AnMaster> tusho, and there is always bitrot
18:01:13 <tusho> it is certainly maintained
18:01:14 <MikeRiley> got to run for a bit,,,,be back later....
18:01:15 <tusho> last release in march
18:01:19 <AnMaster> tusho, nice
18:01:28 <tusho> plus
18:01:31 <tusho> doom3 used it as a build system
18:01:33 -!- MikeRiley has quit ("Leaving").
18:01:37 <AnMaster> so?
18:01:38 <tusho> i'd think they'd be pretty careful about that
18:01:48 <AnMaster> tusho, cmake got KDE
18:01:56 <tusho> *KDE uses cmake
18:01:57 <AnMaster> and some other heavy projects
18:02:02 <tusho> and true, but so has scons
18:02:03 <AnMaster> tusho, yes as users
18:02:06 <tusho> most of them just aren't open source :)
18:02:09 <tusho> from what I can tell
18:02:17 <tusho> but anyway, scons is pretty widely used
18:02:22 <tusho> and actively maintained as far as I can tell
18:02:26 <tusho> and robust in my experience
18:02:31 <AnMaster> tusho, may learn it for future software
18:02:44 <tusho> also, it is crazy about backwards compat. - AnMaster will like that
18:02:46 <tusho> it works in python 1.5
18:02:53 <tusho> (when python 3.0 alpha/beta is out)
18:02:55 <AnMaster> tusho, hah. no I don't really care about that
18:02:56 <tusho> *on
18:03:09 <AnMaster> I do care for python 2.4 and 2.5, maybe 2.3 too
18:03:17 <AnMaster> not longer back than that
18:03:24 <tusho> 2.4 is the minimum worth supporting
18:03:27 <tusho> 2.3 was pretty painful, I hear
18:03:31 <tusho> and 2.2 is just unusable
18:03:32 <AnMaster> oh?
18:03:37 <AnMaster> 1.5 then?
18:03:45 <tusho> AnMaster: yeah, so what
18:03:51 <tusho> scons devs can be masochists if they want
18:03:53 <AnMaster> was it even more unusable?
18:03:57 <tusho> but they apparently have users working on 1.5
18:03:58 <tusho> which is a WTF
18:04:01 <tusho> and 1.5 ...
18:04:04 <tusho> well, that was ages ago
18:04:09 <tusho> like 1996 or something
18:04:12 <AnMaster> tusho, I got 1.2 for Mac OS classic somewhere I think...
18:04:19 <tusho> way before new style classes or anything was even considered
18:04:20 <tusho> *have
18:04:24 <tusho> (that was at you)
18:04:26 <AnMaster> argh right
18:11:10 -!- Sgeo has joined.
18:57:12 -!- lilja has joined.
18:57:20 <lilja> okokokokoko
19:39:13 <psygnisfive> blahblahblah
19:41:52 * tusho is finally getting an #esoteric qdb up
19:42:01 <tusho> we desperately need it, how many awesome quotes have we left to just rot in the logs
19:42:02 <tusho> psht
19:43:05 -!- Corun has joined.
19:44:14 -!- oerjan has joined.
19:52:27 <psygnisfive> awesome quotes?
19:52:58 <tusho> psygnisfive: yes
19:53:11 <tusho> a while ago every day in #esoteric spawned an amazing nugget of a quote.
19:53:21 <tusho> and ais complained that there was nowhere to put them because i'm lazy
19:53:22 <psygnisfive> tusho how long have you been frequenting #esoteric, btw?
19:53:33 <tusho> psygnisfive: sometime 2007
19:53:46 <psygnisfive> ah ok, so only since you were 11.
19:53:50 <tusho> hah
19:53:57 <tusho> dunno if that was before or after my birthday
19:54:03 <tusho> whatever, it feels longer than it is
19:54:20 <psygnisfive> i was envisioning a little(r) tusho
19:54:32 <oerjan> february i think, wasn't that when you posted that regexp language...
19:54:33 <psygnisfive> and it was funny
19:54:36 <tusho> psygnisfive: i'm actually lament.
19:54:46 <tusho> oerjan: that regexp language came way after me coming here
19:54:46 <psygnisfive> :O
19:54:50 <tusho> p.s. february which year
19:55:02 * oerjan goes to check his filestamp
19:55:54 <oerjan> oh, May 30 2007
19:56:26 <psygnisfive> you know what i want on Mac OS X?
19:56:31 <tusho> oerjan: that the regexp time
19:56:32 <psygnisfive> i want a built-in way to make booklets
19:56:33 <tusho> or my first visit
19:56:37 <psygnisfive> why is there no such facility?
19:56:40 <psygnisfive> -_-
19:56:47 <oerjan> tusho: the filestamp on my interpreter for it
19:56:59 <tusho> oerjan: ah
19:57:04 <tusho> then I've been here longer than I thought
19:57:11 <oerjan> also, that's the first thing i remember about you
19:58:22 <tusho> is that regexp language TC after all, oerjan?
19:58:24 <tusho> it seems it
19:58:29 <tusho> it's just string rewriting, albeit bizzare one
19:58:32 <tusho> s/one//
19:59:18 <oerjan> i don't think we figured that out. although with full perl regexps anything is possible :D
19:59:34 <oerjan> oh wait it was python
19:59:35 <tusho> well yeah :)
19:59:40 <oerjan> the perl one was buggy
19:59:40 <tusho> also your impl was broekn
19:59:46 <tusho> it didn't handle escapes properly
20:00:47 <oerjan> hm we may have had a disagreement on the second (python) one.
20:03:27 <tusho> oerjan: it didn't handle \n and co
20:03:28 <oerjan> (they're at oerjan.nvg.org/esoteric/ehird.{pl,py}, i never made a link from anywhere else i think)
20:03:36 <tusho> but you can't do eval('"' + s + '"')
20:03:39 <tusho> because then " is special
20:04:04 <tusho> also
20:04:07 <tusho> while len(prog) == 3 is wrong
20:07:10 <AnMaster> <psygnisfive> i want a built-in way to make booklets
20:07:14 <AnMaster> psygnisfive, you have latex right?
20:07:25 <AnMaster> so just use the booklet document class
20:07:49 <AnMaster> psygnisfive, issue solved :P
20:08:01 <AnMaster> psygnisfive, http://www.ctan.org/pub/tex-archive/macros/latex/contrib/booklet/
20:08:29 <AnMaster> psygnisfive, *warning* doesn't work with custom marginals
20:08:42 <psygnisfive> anmaster: 1) no i dont have latex. 2) i need to print booklets from existing documents not documents ive made.
20:09:06 <AnMaster> psygnisfive, well 1) get it 2) bad luck
20:10:09 <psygnisfive> getting latex is fucking impossible on a mac. not because it doesnt exist but because you open source idiots dont know how to make user friendly anything
20:11:07 <tusho> psygnisfive: texlife.
20:11:09 <tusho> *live
20:11:11 <tusho> texlive. get it.
20:11:46 <psygnisfive> does it require ghostscript?
20:11:56 <tusho> no.
20:12:11 <tusho> and calling a huge group of people idiots as a blanket makes you a complete asshole, btw
20:12:11 <psygnisfive> ill look into it. not that it matters since i dont need (la)tex :P
20:12:19 <psygnisfive> i know im an asshole. ;)
20:12:54 <psygnisfive> but i reserve the right to do so, since all the software i write (or atleast get halfway through writing) is MIT licensed
20:13:19 <AnMaster> har
20:13:29 <AnMaster> psygnisfive, and what do you mean "user friendly"
20:13:38 <AnMaster> I make sure to use getopt correctly
20:13:44 <AnMaster> and to add a -h
20:13:48 <AnMaster> as well as a man page
20:13:57 <AnMaster> if that isn't user friendly I don't know what is
20:14:00 <AnMaster> psygnisfive!
20:14:03 <psygnisfive> well, last time i tried to install a tex viewer i needed to install like forty million dependencies by handing via the command line
20:14:15 <psygnisfive> none of these dependencies were actually listed anywhere in the install instructions
20:14:19 <AnMaster> psygnisfive, the shell is user friendly
20:14:28 <tusho> psygnisfive: texlife is a dmg
20:14:28 <AnMaster> and well I don't know about a tex viewer
20:14:29 <AnMaster> wtf is that
20:14:32 <tusho> which expands to a point-n-click installer
20:14:32 <psygnisfive> i had to find them via the ERRORS the thing spit out
20:14:37 <tusho> and it comes with an ide too
20:14:41 <tusho> which compiles & opens a pdf ready
20:14:42 <tusho> etc
20:14:47 <tusho> it's 100mb, but then it has just about everything
20:14:50 <psygnisfive> awesome tusho. <3
20:14:52 <tusho> including all the document types and other contributed stuff
20:14:58 <tusho> also: it uses xetex
20:14:59 <tusho> so: unicode
20:15:00 <AnMaster> tusho, well not all contributed stuff
20:15:07 <tusho> AnMaster: well yeah
20:15:08 <AnMaster> tusho, and xetex got issues
20:15:15 <psygnisfive> i will look at it tusho, thank you
20:15:16 <tusho> works fine for me and many others
20:15:26 <AnMaster> like not supporting the micro typography that pdftex got
20:15:27 <tusho> psygnisfive: http://thepiratebay.org/tor/3630644/
20:15:27 <AnMaster> tusho, :P
20:15:30 <tusho> is a torrent
20:15:31 <tusho> for it
20:15:32 <tusho> (it's free)
20:15:35 <tusho> official torrent
20:15:43 <AnMaster> tusho, once it has micro typography I may consider xetex
20:15:48 <AnMaster> pdftex got it
20:15:56 <psygnisfive> pdfteX!
20:15:57 <tusho> AnMaster only uses things that have features which he doesn't use
20:16:00 <psygnisfive> also impossible to get installed!
20:16:09 <AnMaster> psygnisfive, I bet texlive contains it...
20:16:10 <AnMaster> duh
20:16:12 <tusho> psygnisfive: or if you don't want a bittorrent
20:16:13 <tusho> ftp://ftp.tex.ac.uk/tex-archive/systems/texlive/Images/texlive2007-live-20070212.iso.zip
20:16:18 <AnMaster> just get a tex distro
20:16:31 <tusho> if you go on like any site for a tex for os x
20:16:32 <AnMaster> <tusho> AnMaster only uses things that have features which he doesn't use
20:16:34 <AnMaster> ...........
20:16:35 <tusho> they all say 'use tex live foo'
20:16:39 <tusho> so yeah.
20:16:40 <AnMaster> I USE microtypography
20:16:42 <psygnisfive> i just cant stand the majority of open source software
20:16:43 <AnMaster> so what
20:16:43 <AnMaster> the
20:16:44 <AnMaster> heck
20:16:46 <AnMaster> do you mean?
20:16:47 <AnMaster> tusho, ?
20:16:49 <AnMaster> tell me
20:16:50 <tusho> AnMaster: and you develop kernels in python
20:16:51 <tusho> :)
20:16:55 <AnMaster> tusho, no I don't
20:16:59 <tusho> no, you don't
20:17:04 <tusho> but you won't use python if you can't develop a kernel in it
20:17:07 <AnMaster> tusho, but the Funge108 standard uses micro typography
20:17:19 <AnMaster> tusho, I might, just to spite you
20:17:22 <tusho> and nobody will use funge108
20:17:27 <tusho> gee, a lot of non-use isn't there
20:17:58 <AnMaster> oh god.... please stop acting like a 12 year old girl....
20:19:07 <tusho> AnMaster can't bear people saying things
20:19:11 <tusho> it disturbs him greatly :-(
20:19:17 <AnMaster> not at all
20:19:31 <AnMaster> I just can't bear people who don't act like grown ups
20:22:25 -!- MikeRiley has joined.
20:40:47 <AnMaster> MikeRiley, hi
20:40:54 <MikeRiley> hello!!
20:41:19 <AnMaster> did you come up with some solution for 64-bits?
20:41:47 <MikeRiley> not yet,,,,been working on getting my TOYS module 100% compliant....
20:41:58 <MikeRiley> which finishes up my BADs....
20:42:06 <MikeRiley> then will see what i can come up with on the 64-bits...
20:43:45 <AnMaster> MikeRiley, you fixed fingerprints?
20:43:55 <AnMaster> I mean fingerprint overloading
20:44:01 <MikeRiley> no.....and may not....
20:44:08 <MikeRiley> i like my way of doing it better,,,,however....
20:44:19 <MikeRiley> i may add to the -Y compatability mode to do it the other way....
20:44:19 <AnMaster> well why not provide a runtime option then?
20:44:29 <MikeRiley> the runtime option is what i will do...
20:44:45 <MikeRiley> probably do it with -Y (maybe rename it) for spec compatability mode...
20:45:20 <AnMaster> MikeRiley, also you don't grow stacks dynamically do you?
20:45:31 <AnMaster> cfunge will allocate in chunks
20:45:51 <AnMaster> while your interpreter segfaults instead if you don't do those -n and such options
20:46:15 <MikeRiley> no,,,,but am going to change that....
20:46:28 <MikeRiley> after mycology wanted to put nearly half a million entries on the stack!!!! eheheheheheh
20:46:34 <MikeRiley> will change it to dynamic allocation...
20:46:42 <AnMaster> cfunge is standard compatible by default, I made a point of not needing any special options to be standard compatible
20:46:44 <MikeRiley> chunks is what i will do as well...quicker that way....
20:46:51 <AnMaster> <MikeRiley> after mycology wanted to put nearly half a million entries on the stack!!!! eheheheheheh
20:46:53 <AnMaster> HRTI right?
20:47:00 <AnMaster> that one is a pain
20:47:02 <MikeRiley> yep!!!!
20:47:05 <MikeRiley> the ff*kyn
20:47:11 <AnMaster> you can't debug it under valgrind
20:47:17 <AnMaster> because valgrind will eat so much memory
20:47:25 <AnMaster> so I have to comment out HRTI to test fingerprints after
20:47:29 <AnMaster> under valgrind
20:47:39 <MikeRiley> for a long time i was getting a stack overflow....
20:47:47 <AnMaster> of course standalone cfunge handles it without overflows
20:47:48 <AnMaster> :)
20:47:51 <MikeRiley> so i put a breakpoint at that section of code to see what it was doing...
20:48:02 <AnMaster> MikeRiley, oh yeah, it is killing time ;P
20:48:07 <MikeRiley> and saw that ff*kyn and just went OH MY GOSH!!!!!!
20:48:13 <Deewiant> :-)
20:48:13 <MikeRiley> i figured that....
20:48:22 <Deewiant> it was just fkyn at first
20:48:24 <MikeRiley> would be better to make a different loop and clear the stack after each y...
20:48:24 <AnMaster> cfunge can be pretty fast at that though
20:48:31 <AnMaster> with -S switch
20:48:44 <AnMaster> because most of the time is taken by env variables
20:48:46 <Deewiant> but the windows timer has such a low granularity (in CCBI, at least) that fkyn wasn't enough to get even a single interval :-P
20:48:50 <MikeRiley> otherwise it is a great way to test for very large stack expansion!!! eheheheheheeheheheh
20:48:51 <AnMaster> and -S restrict that
20:49:03 <AnMaster> Deewiant, cfunge got very good granularity on my system
20:49:09 <AnMaster> thanks to HPET I bet
20:49:09 <MikeRiley> my interval is pretty small....from just m to t will show about 200....
20:49:15 <AnMaster> my system has HPET
20:49:31 <Deewiant> MikeRiley: yeah, it could be done like that but I figured that might as well test for big stacks while we're at it ;-)
20:49:37 <AnMaster> Testing fingerprint HRTI... loaded.
20:49:38 <AnMaster> UNDEF: G gives clock granularity as 2 microseconds
20:49:38 <AnMaster> GOOD: T reflected when called before M
20:49:38 <AnMaster> UNDEF: S pushed 969346
20:49:38 <AnMaster> UNDEF: T after M pushed 30 and a second T, after ff*kyn, pushed 232
20:49:38 <AnMaster> GOOD: ET reflected
20:49:42 <AnMaster> using -S
20:49:42 <MikeRiley> and that certainly does!!!! eheheheheheheeh
20:49:48 <AnMaster> MikeRiley, ^
20:49:53 <Deewiant> AnMaster: CCBI has equivalent granularity on Linux as well
20:49:57 <AnMaster> MikeRiley, but I will have to grow the stack a lot
20:50:01 <AnMaster> Deewiant, I know...
20:50:06 <MikeRiley> yep....a whole lot!!!!!
20:50:16 <AnMaster> you can pre-allocate yours
20:50:20 <AnMaster> a lot faster actually
20:50:25 <AnMaster> than allocating in chunks
20:50:28 <AnMaster> yes I do it in chunks
20:50:45 <AnMaster> an idea I had was to cache size of y and pre-allocate that in one go
20:51:04 <MikeRiley> mine is preallocated,,,,but if you are doing concurrent funge with really large stacks!!!! yikes!!!!
20:51:08 <AnMaster> heck even caching parts of the y request would help
20:51:15 <MikeRiley> yes it would...
20:51:25 <AnMaster> environment mostly
20:51:31 <AnMaster> that is the main time hog
20:52:01 <MikeRiley> yeah,,,same here,,,huge enviornment...
20:52:21 <AnMaster> well I never seen a modern linux system without a huge environment...
20:52:27 <AnMaster> FreeBSD have smaller ones
20:52:30 <AnMaster> at least on my server
20:52:49 <AnMaster> just 13 variables on my FreeBSD server
20:52:53 <MikeRiley> never played with FreeBSD,,,,been hooked on linux too long!!! eheheheheeheh
20:52:59 <MikeRiley> wow!!! that is not very many!!!
20:53:03 <AnMaster> 86 on my Gentoo system
20:53:04 <AnMaster> ...
20:53:07 <AnMaster> most are HUGE too
20:53:20 <AnMaster> ROOTPATH=/usr/kde/3.5/sbin:/usr/kde/3.5/bin:/usr/local/bin:/usr/bin:/bin:/opt/bin:/usr/x86_64-pc-linux-gnu/gcc-bin/4.1.2:/opt/ghc/bin:/opt/blackdown-jdk-
20:53:22 <AnMaster> like that
20:53:26 <Deewiant> 41 62 2598 on WinXP here
20:53:28 <AnMaster> LS_COLORS=no=00:fi=00:di=01;34:ln=01;36:pi=40;33:so=01;35:do=01;35:bd=40;33;01:cd=40;33;01:or=01;05;37;41:mi=01;05;37;41:su=37;41:sg=30;43:tw=30;42:ow=34;42:st=37;44:ex=01;32:*.tar=01;31:*.tgz=01;31:*.svgz=01;31:*.arj=01;31:*.taz=01;31:*.lzh=01;31:*.lzma=01;31:*.zip=01;31:*.z=01;31:*.Z=01;31:*.dz=01;31:*.gz=01;31:*.bz2=01;31:*.bz=01;31:*.tbz2=01;31:*.tz=01;31:*.deb=01;31:*.rpm=01;31:*.jar=01;31:*.rar=
20:53:28 <AnMaster> 01;31:*.ace=01;31:*.zoo=01;31:*.cpio=01;31:*.7z=01;31:*.rz=01;31:*.jpg=01;35:*.jpeg=01;35:*.gif=01;35:*.bmp=01;35:*.pbm=01;35:*.pgm=01;35:*.ppm=01;35:*.tga=01;35:*.xbm=01;35:*.xpm=01;35:*.tif=01;35:*.tiff=01;35:*.png=01;35:*.svg=01;35:*.mng=01;35:*.pcx=01;35:*.mov=01;35:*.mpg=01;35:*.mpeg=01;35:*.m2v=01;35:*.mkv=01
20:53:31 <AnMaster> and that is not all of it
20:53:36 <AnMaster> LS_COLORS is exetreme
20:53:38 <Deewiant> just pipe it through wc
20:53:39 <AnMaster> extreme*
20:53:44 <AnMaster> Deewiant, I did that...
20:54:01 <AnMaster> gentoo $ env | wc -c
20:54:02 <AnMaster> 5696
20:54:02 <Deewiant> then paste that, not uninteresting snippets of your environment variables :-P
20:54:18 <Deewiant> heh, over twice that of Windows
20:54:20 <AnMaster> freebsd $ env | wc -c
20:54:20 <AnMaster> 337
20:54:23 <Deewiant> and they say Windows is bloated ;-)
20:54:27 <AnMaster> Deewiant, a lot less I bet?
20:54:27 <MikeRiley> eheheheheeheh
20:54:37 <AnMaster> mind you the latter is a root shell
20:54:41 <AnMaster> will check as normal user
20:54:42 <Deewiant> AnMaster: I pasted mine above...
20:54:47 <AnMaster> ah yes
20:54:57 <AnMaster> for root on my gentoo:
20:54:59 <AnMaster> # env | wc -c
20:54:59 <AnMaster> 3891
20:55:03 <AnMaster> a bit less than for user
20:55:11 <AnMaster> but blame my .bashrc for some of that
20:55:15 <AnMaster> bbiab
20:55:35 <MikeRiley> wc shows 3525 for my environment...
20:57:00 <tusho> 21 23 1360
20:57:12 <tusho> os x
20:58:27 <fizzie> 1735 in my Debian here.
20:59:17 <fizzie> 2553 on the SuSE workstations at work.
20:59:27 <MikeRiley> in the TOYS test,,,i get some extra characters at begining of a line:
20:59:29 <MikeRiley> GOOD: G works
20:59:29 <MikeRiley> @+#GOOD: 1J moves itself south one row
20:59:33 <MikeRiley> why???
20:59:47 <Deewiant> guess somethings not as GOOD as Mycology thinks ;-)
21:00:01 <Deewiant> s/things/thing's/
21:01:12 -!- oerjan has quit ("Walk").
21:01:13 <MikeRiley> hmmmmm wonder if the G code is strange...
21:02:13 <AnMaster> back
21:02:43 <AnMaster> MikeRiley, probably something wrong on the stack
21:03:01 <MikeRiley> some of those TOYS commands are pretty foggy on how some of those commands are supposed to work...
21:03:13 <MikeRiley> what is your take on G???
21:03:15 <AnMaster> yes, like the butterfly operator
21:03:22 <Deewiant> a lot like RC/Funge-98's fingerprints ;-)
21:03:39 <AnMaster> MikeRiley, and for TOYS I just reverse engineered the CCBI source
21:03:49 <Deewiant> because he's lazy :-P
21:03:50 <AnMaster> did that for a few other vague fingerprints
21:04:04 <AnMaster> Deewiant, no I did it for the vague stuff, like TOYS and TURT
21:04:12 <MikeRiley> spec mentions j,,,,,but not where it comes from...
21:04:15 <Deewiant> doesn't make you any less lazy :-P
21:04:16 -!- RedDak has joined.
21:04:32 <MikeRiley> i figured out what the butterfly operator is supposed to do...
21:04:33 <AnMaster> Deewiant, for TOYS it was the only realistic approach at least
21:04:34 <Deewiant> MikeRiley: yeah, there are a few possible takes on that
21:04:35 <AnMaster> same for k
21:04:43 <MikeRiley> yep....
21:04:51 <Deewiant> MikeRiley: but there was another command which takes j
21:04:55 <Deewiant> where it was said where it comes from
21:05:04 <Deewiant> and it's probably reasonable to just go with that...
21:05:09 <MikeRiley> let me look...
21:05:12 <AnMaster> Deewiant, anyway didn't you simply say you reverse engineered the RC/Funge one above? :P
21:05:16 <AnMaster> and that is harder
21:05:21 <AnMaster> at least CCBI's source is very readable
21:05:30 <Deewiant> where did I say that?
21:05:36 <AnMaster> <Deewiant> a lot like RC/Funge-98's fingerprints ;-)
21:05:53 <Deewiant> AnMaster: in response to 2008-07-25 23:03:01 ( MikeRiley) some of those TOYS commands are pretty foggy on how some of those commands are supposed to work...
21:05:57 <MikeRiley> there is F and G,,,,both mention using j,,,,but not where it comes from...
21:06:04 <AnMaster> ah
21:06:06 <AnMaster> right
21:06:16 <AnMaster> MikeRiley, j being?
21:06:42 <MikeRiley> spec says: G ('counterclockwise') pops a vector, then a value /i/. It then pushes
21:06:42 <MikeRiley> onto the stack /j/ groups of /i/ cells each which it retrieves as a 2D
21:06:42 <MikeRiley> matrix in Funge-space in row-major order, the least point of which being
21:06:42 <MikeRiley> the vector supplied to it.
21:06:58 <MikeRiley> does not mention where j comes from...
21:07:10 <MikeRiley> F is similar,,
21:07:30 <Deewiant> I think I just checked what !Befunge did, and it took it from the top of the stack
21:07:37 <MikeRiley> G takes a 2d array from fungespace and puts it on the stack,,,
21:07:37 <AnMaster> // j's location not in spec...
21:07:37 <AnMaster> j = StackPop(ip->stack);
21:07:37 <AnMaster> i = StackPop(ip->stack);
21:07:39 <MikeRiley> F is the referse...
21:07:41 <AnMaster> that's from cfunge
21:07:49 <AnMaster> I expect it is in CCBI too
21:07:52 <AnMaster> the comment
21:07:54 <MikeRiley> so you are popping j first in both cases????
21:07:57 <Deewiant> yeah
21:08:17 <AnMaster> MikeRiley, well there was a reason I reverse engineered CCBI for TOYS
21:08:26 <MikeRiley> ok,,,,i have it reversed,,,let me try changing it...
21:08:44 <MikeRiley> TOYS was incomplete in Rc/Funge-98....finishing it now!!! eheheheheeheheheh
21:08:47 <AnMaster> MikeRiley, oh btw your FRTH assumes the implementer knows Forth, I had to google for wtf "forth roll" was :P
21:08:57 <AnMaster> and the other ones too
21:09:24 <AnMaster> Deewiant, you should test MODE in combination with FRTH
21:09:30 <MikeRiley> reversing i and j still give me:
21:09:32 <MikeRiley> GOOD: F works
21:09:32 <MikeRiley> GOOD: G works
21:09:32 <MikeRiley> @+#GOOD: 1J moves itself south one row
21:09:41 <Deewiant> no, I very much do not wish to test combinations of fingerprints
21:09:50 <Deewiant> because there are too many possible combinations
21:09:56 <MikeRiley> yeah,,,i should document that one better!!!! just like so many others huh????? eheheheheheheheheeh
21:10:00 <ihope> I suddenly want to write a concurrency monad in Haskell.
21:10:11 <AnMaster> Deewiant, well you have a comment in your FRTH about how it interacts with MODE
21:10:17 <AnMaster> so I suggest you test it
21:10:30 <Deewiant> it's unspecced
21:10:44 <AnMaster> Deewiant, you said "top of stack even when using MODE"
21:10:48 <AnMaster> or something like that iirc
21:10:51 <Deewiant> and just because I have a comment there doesn't mean I shouldn't have a comment everywhere else ;-P
21:10:53 <AnMaster> anyway I don't implement MODE
21:11:08 <AnMaster> thank god for that
21:11:19 <MikeRiley> MODE was kinda fun to implement...
21:11:33 <AnMaster> MikeRiley, doesn't it need working at the bottom end of the stack?
21:11:36 <AnMaster> quite a performance hit
21:11:54 <AnMaster> unless you do something crazy like allocate it in the middle of a heap block
21:12:13 <AnMaster> so basically you need a dequ
21:12:18 <AnMaster> err
21:12:21 <AnMaster> deque*
21:13:15 <MikeRiley> yes it does work on the bottom end of the stack,,,and yes it can be a big performance hit...
21:14:22 <AnMaster> MikeRiley, FRTH can be a pretty big performance hit too
21:14:55 <MikeRiley> not really....
21:15:23 <AnMaster> MikeRiley, roll
21:15:43 <psygnisfive> oh god x_x
21:15:51 <MikeRiley> if rolling deep into a deep stack it could be...
21:15:57 <AnMaster> MikeRiley, try roll of bottom entry after HRTI ;P
21:15:57 <psygnisfive> i just hurt my back T_T
21:16:03 <AnMaster> psygnisfive, ouch
21:16:10 <MikeRiley> like from the bottom of ff*ky!!! eheheheheheeheh
21:17:04 <psygnisfive> its the kind of pain that makes you breath in short bursts that you hold in
21:17:07 <psygnisfive> :(
21:18:03 <AnMaster> psygnisfive, ouch
21:18:10 <AnMaster> psygnisfive, visit a doctor?
21:18:19 <psygnisfive> this has happened before
21:18:25 <psygnisfive> it goes away in a few hours
21:18:31 <AnMaster> psygnisfive, be more careful?
21:18:37 <psygnisfive> all i did was bend over! T_T
21:18:56 <AnMaster> psygnisfive, don't do that then?
21:19:06 <psygnisfive> but.. how.. what..
21:19:07 <psygnisfive> X_X
21:27:17 <AnMaster> psygnisfive, what?
21:27:38 <psygnisfive> how can i not bend over?!
21:27:43 <AnMaster> MikeRiley, btw cfunge grows it stack dynamically but it never shrinks it
21:27:54 <AnMaster> psygnisfive, um... by standing tall?
21:28:15 <AnMaster> psygnisfive, or?
21:28:21 <MikeRiley> i will alloy my dynamic stack to grow and shrink,,,shrinking will be on the same chunk boundary....
21:28:33 <AnMaster> MikeRiley, hm why shrink?
21:28:37 <AnMaster> there is no need really
21:28:38 <MikeRiley> why not???
21:28:43 <AnMaster> and it is slower
21:28:57 <AnMaster> I already allocate in chunks
21:29:11 <AnMaster> 4096 items at a time
21:29:56 <AnMaster> assuming 32-bit Funge and x86 this mean four memory pages at a time I think
21:30:03 <AnMaster> um wait yes
21:30:32 <AnMaster> $ getconf PAGE_SIZE
21:30:33 <AnMaster> 4096
21:30:35 <AnMaster> same on x86_64
21:30:40 <MikeRiley> well,,,,i want rid of mycology's 500,000 item stack!!! ehehehehehehe
21:30:56 <AnMaster> haha
21:31:09 <AnMaster> MikeRiley, it is simple to fix
21:31:16 <AnMaster> env -i ./funge blah blah
21:31:18 <AnMaster> maybe do
21:31:24 <AnMaster> env -i TERM=$TERM ./funge blah blah
21:31:31 <AnMaster> and maybe PATH too
21:31:39 <MikeRiley> yeah,,,could alter my start script i guess...
21:31:42 <AnMaster> but that means almost clean environment
21:31:51 -!- KingOfKarlsruhe has joined.
21:32:10 <AnMaster> MikeRiley, you probably want PATH too, if you want to locate perl
21:32:17 <AnMaster> or do you assume /usr/bin/perl?
21:32:27 <AnMaster> on FreeBSD it is /usr/local/bin/perl
21:32:53 <MikeRiley> it gets perl from the path....
21:32:57 <AnMaster> ah good
21:33:13 <AnMaster> MikeRiley, anyway mycology doesn't like an empty environment
21:33:15 -!- pikhq has left (?).
21:33:21 <AnMaster> it is a bug Deewiant haven't fixed yet
21:33:25 <AnMaster> even though I reported it
21:33:46 <Deewiant> there is literally nothing I can do about it
21:33:50 <AnMaster> That the command-line arguments were: [ "mycology/mycology.b98" null ]
21:33:50 <AnMaster> That the environment variables are:
21:33:57 <AnMaster> that's what it looks like
21:33:59 <Deewiant> because the spec says that they come in a row
21:34:04 <AnMaster> yes and?
21:34:06 <Deewiant> so then you get like 4 zeroes in a row or whatever
21:34:12 <Deewiant> how do you know where args end and env starts
21:34:41 <AnMaster> Deewiant, well you know 2 zeros in a row mark the end of the arguments
21:34:50 <AnMaster> so just end that after two zero
21:35:03 <AnMaster> then you take care of the next two zero in the environment outputting code
21:35:07 <Deewiant> AnMaster: but 3 zeroes in a row mean an empty argument and the end of the arguments
21:35:08 -!- Corun has quit ("This computer has gone to sleep").
21:35:25 <AnMaster> Deewiant, so check for 4 then?
21:35:46 <Deewiant> AnMaster: two empty arguments and the end
21:36:03 <AnMaster> ./cfunge -SF mycology/mycology.b98 "" "" ""
21:36:04 <AnMaster> so...
21:36:10 <AnMaster> it doesn't handle that either
21:36:17 <Deewiant> what's it say
21:36:19 <AnMaster> Deewiant, do you think Funge-108 should fix this in some way?
21:36:26 <MikeRiley> i think it should...
21:36:27 <AnMaster> and do you have a suggestion how?
21:36:31 <Deewiant> well, I think zero-terminated strings are crap in general :-P
21:36:41 <MikeRiley> maybe push a count of the arguments....
21:36:42 <AnMaster> Deewiant, well they are easier to write in Funge
21:36:43 <Deewiant> and in this case, having zero-terminated arrays containing zero-terminated strings.......
21:36:47 <AnMaster> MikeRiley, good idea
21:36:50 <MikeRiley> instead of relying on the double null...
21:36:54 <AnMaster> agreed
21:36:54 <Deewiant> AnMaster: not if we had a reasonable k :-P
21:37:01 <AnMaster> Deewiant, eh?
21:37:06 <Deewiant> easier to write in Funge
21:37:10 <AnMaster> Deewiant, would you prefer something like:
21:37:10 <Deewiant> compare: >:#,_ to k,
21:37:17 <AnMaster> "foo bar"7
21:37:23 <AnMaster> easy yes
21:37:29 <AnMaster> but imagine long strings
21:37:33 <AnMaster> then correcting a typo
21:37:40 <AnMaster> then having to recalculate it
21:38:01 <Deewiant> oh noes, Befunge is hard to use for string manipulation!!
21:38:06 <AnMaster> har
21:38:07 <AnMaster> har
21:38:11 <Deewiant> there's STRN for that
21:38:24 <AnMaster> STRN takes 0"gnirts"
21:38:31 <MikeRiley> almost done with the TOYS module now....
21:39:11 <AnMaster> nice
21:39:11 <MikeRiley> that is whay i defined STRN!!! so that strings would be easier to deal with...
21:39:23 <Deewiant> AnMaster: yeah, but STRN has a strlen
21:39:47 <AnMaster> Deewiant, well but becomes a pain if you are trying to load other fingerprints too
21:39:53 <AnMaster> that conflicts
21:40:00 <AnMaster> and you need that char
21:40:06 <Deewiant> oh noes, Befunge is a pain!!
21:40:09 <AnMaster> of course you could unload parts of STRN
21:40:43 <AnMaster> Deewiant, If we keep it hard, how can we aim for it becoming the defacto standard in enterprise solutions!?
21:41:19 <MikeRiley> or use FNGR!!!! eheheheheheeheh
21:41:31 <AnMaster> MikeRiley, no, it breaks standard ;P
21:41:59 <MikeRiley> eheheheehehehe
21:42:09 <MikeRiley> should T in TOYS reflect on a bad dimension???
21:42:25 <AnMaster> MikeRiley, mine does at least
21:42:37 <AnMaster> TOYS is so vague it is hard to say what it should
21:42:45 <MikeRiley> i agree...
21:42:54 <MikeRiley> ok,,,,if yours does,,,will have mine do that as welll...
21:43:02 <AnMaster> but I think the defacto way of reporting errors in Funge is reflecting
21:43:13 <AnMaster> switch (StackPop(ip->stack)) {
21:43:13 <AnMaster> case 0: IfEastWest(ip); break;
21:43:13 <AnMaster> case 1: IfNorthSouth(ip); break;
21:43:13 <AnMaster> default: ipReverse(ip); break;
21:43:13 <AnMaster> }
21:43:15 <AnMaster> that is my T
21:43:30 <AnMaster> as cfunge doesn't do anything but two dimensions
21:44:50 <AnMaster> anyone got a C compiler to befunge?
21:45:02 <MikeRiley> looks good,,,but i do need to add the case 2,,,since Rc/Funge does support 3 dimensions...
21:45:21 <AnMaster> MikeRiley, indeed
21:45:23 <tusho> MikeRiley: why not N dimensions?
21:45:23 <tusho> :D
21:45:45 <AnMaster> tusho, that is hard because you need to select at execution time
21:45:51 <tusho> yes
21:46:32 <AnMaster> I have some idea for a "n-dimensional on demand" fungeoid
21:47:06 <MikeRiley> may just make it n-dimensional....in some later version...
21:47:42 <MikeRiley> no different than a 4-dimension array,,,so certainly possible....just need a new command to move between dimensions beyond 3....
21:48:11 <AnMaster> MikeRiley, um? x?
21:48:36 <AnMaster> x will allow that
21:49:42 <MikeRiley> yes,,,x will certainly allow for that...
21:50:02 <AnMaster> not very userfriendly for 105 dimensions though
21:50:17 <MikeRiley> ehehehehhe really!!!!
21:50:51 <AnMaster> 0000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000001x
21:51:05 <AnMaster> and then imagine writing that across some other dimension
21:51:18 <AnMaster> as in "not any of the standard 2D"
21:51:27 <AnMaster> MikeRiley, there is one problem however with that
21:51:38 <AnMaster> newline, formfeed
21:51:46 <AnMaster> what to use for the next 115 dimensions?
21:51:49 <AnMaster> in the source file
21:51:56 <AnMaster> MikeRiley, give me a good answer to that please
21:52:52 <MikeRiley> hmmmmmmm,,,,good question....
21:53:09 <MikeRiley> maybe use another control character,,,,followed by a byte for the dimension???
21:53:29 <AnMaster> followed?
21:53:34 <AnMaster> hm yes ok
21:53:38 <AnMaster> right
21:53:48 <AnMaster> MikeRiley, vertical tab!
21:53:53 <AnMaster> no one uses that these days
21:54:25 <AnMaster> MikeRiley, anyway would changing dimension n reset all lower dimensions?
21:55:15 <MikeRiley> i would say yes....
21:55:26 <MikeRiley> just like putting lf into the file resets x and y in trefunge...
21:55:40 <MikeRiley> vt seems applicable...
21:57:13 <MikeRiley> two BADs to go and TOYS will be all done...
21:58:23 <AnMaster> MikeRiley, in what parts of TOYS?
21:58:48 <MikeRiley> BAD: 000p000W reflects
21:58:48 <MikeRiley> BAD: 000p100W reflects
21:59:10 <AnMaster> night
22:09:15 -!- tusho has quit ("And then-").
22:09:34 -!- tusho has joined.
22:11:47 -!- Corun has joined.
22:13:19 <tusho> You're all in serifed fonts.
22:13:20 <tusho> Be afraid.
22:13:22 <tusho> Be very afraid.
22:15:35 * Corun gets out the anti-serif barrier
22:16:34 <tusho> Corun: I think you mean the "sans barrier".
22:16:43 <tusho> Alas, it is to no avail.
22:16:48 <tusho> My serifs will defeat you immediately.
22:16:53 <Corun> I doubt it
22:17:00 <Corun> Can your serifs... _swim_?
22:17:10 <Corun> Through _ink_?
22:17:10 <tusho> Swim? Pah! My serifs can /fly/.
22:17:17 <Corun> I am in space
22:17:20 <tusho> And they evaporate ink *just like that*.
22:17:34 <Corun> Surrounded by a sphere of "sans serif" ink
22:17:39 <Corun> Bring it.
22:18:19 <tusho> Corun: Bring what? My serifs are already behind you...
22:18:32 <Corun> That's odd
22:18:35 * tusho watches as serifs rip away Corun's flesh until there is no more.
22:18:36 <Corun> Because I'm behind me
22:18:44 <tusho> Oh cute, they're nibbling at your bones.
22:18:47 <tusho> Who's the cute skeleton? Aww.
22:18:48 <Corun> Which means that they're behind behind me
22:18:55 <tusho> Shut up. You're dead.
22:19:02 <Corun> Which means they're behind behind behind me, so they must not exist
22:19:07 <Corun> YOU'RE CHEATING
22:19:12 <tusho> I can cheat all I want.
22:19:14 <tusho> And why?
22:19:16 <tusho> Because I have serifs.
22:19:17 <tusho> Fuck yeah.
22:19:20 <Corun> Then I can speak when dead :-P
22:19:25 <tusho> No.
22:19:28 <tusho> Only I can cheat.
22:19:30 <Corun> I am speaking
22:19:31 <tusho> Do you have serifs? NO.
22:19:35 <Corun> How do you know?
22:19:39 <Corun> HOW DO YOU KNOW?
22:19:47 <Corun> I might just keep a few
22:19:51 <Corun> For just such an occasion
22:19:58 <tusho> Because you have anti-serifs.
22:20:06 <tusho> If they saw them - which they would - they would dispose of you.
22:20:14 <Corun> Yeah but I keep them well seperated
22:20:17 <tusho> No.
22:20:20 <tusho> They are omnipotent.
22:20:21 <Corun> Otherwise they'd cancel out
22:20:24 <Corun> And release energy ;-)
22:20:53 <tusho> No.
22:20:55 <tusho> They'd just kill you.
22:21:18 <Corun> They also kill you
22:21:21 <Corun> For talking to me
22:21:22 <tusho> No.
22:21:23 <tusho> They love me.
22:21:26 <tusho> I'm wubbalicious.
22:21:42 <Corun> The ones I know hate you
22:21:49 <tusho> You are imagining them.
22:21:57 <Corun> O
22:21:59 <Corun> Oh
22:22:00 <Corun> Well
22:22:02 <Corun> Shit.
22:22:10 <Corun> I guess you win then
22:22:31 <tusho> How the cocky have fallen.
22:22:46 <Corun> :'-(
22:23:37 <tusho> Aww here.
22:23:40 * tusho gives reviving potion
22:36:50 -!- MikeRiley has quit ("Leaving").
23:17:55 <tusho> http://antitrust.slated.org/www.iowaconsumercase.org/011607/3000/PX03020.pdf fun
23:18:54 <tusho> thank you ichverstehe
23:18:56 <tusho> whoever you are
23:19:00 <tusho> 'Once upon a time it suddenly decided to flip backwards, and there went Time'
23:19:01 <tusho> tee hee
23:19:47 <tusho> wrong chan
23:23:38 -!- bwr has left (?).
23:27:23 -!- CakeProphet has quit (Read error: 104 (Connection reset by peer)).
23:50:02 -!- pikhq has joined.
←2008-07-24 2008-07-25 2008-07-26→ ↑2008 ↑all