00:10:38 though it works better if you ask specific questions or hints 00:10:55 tell what you're stuck with or what doesn't work 00:11:40 I'm tired not going to solve it now 00:11:50 I need to take a rest 00:12:07 See you today but around twelve 00:18:01 sure 00:22:03 -!- arseniiv has quit (Ping timeout: 268 seconds). 00:27:02 I want to minimize dependencies of TeXnicard; it already uses SQLite, Ghostscript, and PCRE; so if I add further dependencies I should add ones which are already used by Ghostscript, if possible, I think. 00:38:01 -!- Frater_EST has joined. 00:48:24 I want to support PK format fonts in TeXnicard, and it seem that FreeType doesn't implement that format. Well, I can implement my own; I have done it before. 01:03:27 -!- j-bot has joined. 01:13:18 -!- Frater_EST has left. 01:40:33 -!- nico_nico has joined. 01:41:54 [[User:DmilkaSTD]] https://esolangs.org/w/index.php?diff=68804&oldid=67931 * DmilkaSTD * (-125) 01:42:15 [[User:DmilkaSTD]] https://esolangs.org/w/index.php?diff=68805&oldid=68804 * DmilkaSTD * (+4) 01:45:01 -!- nico_nico has quit (Client Quit). 02:18:07 [[NewFuck]] N https://esolangs.org/w/index.php?oldid=68806 * DmilkaSTD * (+415) Created page with "NewFuck is another brainfuck derivative. The operations changes everytime you make a new project. == Possible operations == {| class="wikitable" |- ! Brainfuck operation !! Po..." 02:19:11 [[NewFuck]] https://esolangs.org/w/index.php?diff=68807&oldid=68806 * DmilkaSTD * (+1) 02:26:46 [[NewFuck]] https://esolangs.org/w/index.php?diff=68808&oldid=68807 * DmilkaSTD * (+463) 02:54:52 -!- oerjan has joined. 03:10:42 -!- Lord_of_Life_ has joined. 03:11:38 -!- Lord_of_Life has quit (Ping timeout: 268 seconds). 03:12:03 -!- Lord_of_Life_ has changed nick to Lord_of_Life. 03:17:42 -!- sftp has quit (Ping timeout: 252 seconds). 03:20:11 -!- sftp has joined. 04:31:08 -!- MDude has quit (Quit: Going offline, see ya! (www.adiirc.com)). 04:34:03 -!- ^[ has quit (Ping timeout: 245 seconds). 04:35:07 -!- ^[ has joined. 04:37:48 -!- dog_star has quit (Ping timeout: 245 seconds). 04:39:12 -!- dog_star has joined. 04:40:50 -!- kspalaiologos has quit (Ping timeout: 240 seconds). 04:47:45 has anyone made a brainfuck variant which allows direct system calls? 04:48:08 for example say the ! command executes a syscall where the syscall number and arguments are given by the value under the head and the next six 04:48:13 and puts the return value under the head 04:48:44 to be actually useful you would also need a way to get real machine-level pointers to brainfuck cells, or some other way to use syscalls which have pointer arguments 04:51:40 you could just say that the address of the first memory cell is in the first memory cell when the program starts 04:51:45 or something like that 04:58:53 Or you could translate 0 into a pointer to the first cell, 1 into a pointer to the second, and so on 04:59:16 except that some arguments are pointers and some aren't 04:59:24 and it would need to know about each syscall in that case 04:59:37 or there may even be cases where it could be a pointer or not according to the semantics of the call 04:59:51 I think the interpreter would have to know about the syscalls anyway 05:00:17 why? 05:01:00 So it knows how many inputs to take and how many outputs to provide. Unless you always have it do 6 arguments and ignore the extra ones 05:01:22 yeah 05:01:28 on linux anyway, I think it will work fine that way 05:02:01 the syscall ABI is pretty simple 05:02:05 And you'd need some way to deal with structs 05:02:49 yeah the brainfuck code would need to manually construct the appropriate struct layout on the stack 05:02:52 and then get a pointer to it 05:02:55 s/stack/tape/ 05:03:10 it'd be a pain in the ass but 05:03:14 this is esoprogramming 05:06:16 what's a _good_ alternative to syscalls, anyway. 05:09:37 You can also use calls for some kind of VM. I invented (but did not implement) a way to use Glk with Befunge, for example. 05:15:34 what is Glk? 05:18:53 -!- atslash has quit (Quit: Leaving). 05:19:29 It is a API for setting up text windows for I/O especially in text adventure games. It also supports displaying pictures, sounds, music, file I/O, and date/time. 05:26:35 ah, cool 05:26:37 It is a C API, although there is also a dispatch layer which uses a standardized format for arguments and uses numbers to identify each function (although a few functions cannot be dispatched, including the dispatching function itself). 05:32:40 Glulx is a VM which is mainly used with Glk (although the design allows implementations to be made with other I/O systems too), but also has some other functions included, such as the ability to save the entire VM memory and stack to a file, and built-in instructions for searching (linear, binary, or linked). 05:38:38 -!- FreeFull has quit. 05:44:17 -!- xkapastel has joined. 09:03:01 Now loading font metric files is implemented in TeXnicard. 09:06:37 It might be useful to add some extensions beyond what TeX uses, such as a margin character (like the boundary character it is not displayed; it can be used in kerning for hanging punctuation). Possibly some others too, but I don't know which ones. Possibly a table of accented characters, and maybe large character set support, maybe 09:42:33 -!- kspalaiologos has joined. 09:44:12 up for tasks 09:44:36 [[Special:Log/upload]] upload * Hakerh400 * uploaded "[[File:Quine common program structure.png]]": Quine - common program structure 09:44:39 I need to solve the last one 09:45:35 [ s 09:45:36 kspalaiologos: 0 0 1 1 0 0 1 1 0 0 1 1 0 0 1 1 0 0 1 1 0 0 1 1 0 0 1 1 0 0 1 1 09:45:41 wait 09:45:43 what's this 09:45:44 [ set 09:45:45 kspalaiologos: |value error: set 09:51:25 What task is that you will need to solve? 09:54:04 b_jonas, can you look at #jsoftware and help me? 09:54:14 split the vector into three element lists 09:54:17 multiply them 09:54:20 and put it in another list 10:11:41 Recovering Cards {3} Instant ;; Put target non-basic non-creature non-planeswalker non-Equipment non-Vehicle card from your graveyard on top of your library. ;; Retrace ;; Cycling {3} 10:50:05 what is this 10:52:03 `? this 10:52:08 This is something people on the channel like to talk about. We're often unsure what this is, though. Nobody likes this. 10:52:13 hth 10:58:29 -!- kritixilithos has joined. 11:11:11 -!- xkapastel has quit (Quit: Connection closed for inactivity). 11:17:33 -!- oerjan has quit (Quit: leaving). 11:23:16 `? that 11:23:17 that? ¯\(°​_o)/¯ 11:23:24 That's good. 11:33:10 -!- imode has quit (Ping timeout: 268 seconds). 11:37:18 -!- arseniiv has joined. 12:22:55 -!- Frater_EST has joined. 12:23:20 -!- Frater_EST has left. 13:08:19 [[User talk:*berlinquin]] N https://esolangs.org/w/index.php?oldid=68810 * Hex96 * (+101) Created page with "hello! ~~~~" 13:19:22 -!- xkapastel has joined. 13:28:02 Is Taneb actually Tony Hawk? 13:28:35 Just because they both start with "T", I don't think that's enough evidence. 13:33:42 -!- kritixilithos has quit (Remote host closed the connection). 13:35:22 b_jonas: why do you ask? 13:36:43 Maybe they saw you do some sick [INSERT SKATEBOARDING TERMINLOGOY]s. 13:57:02 -!- Frater_EST has joined. 13:57:36 -!- Frater_EST has left. 14:09:08 -!- MDude has joined. 14:12:18 Taneb: I was watching GDQ and it has a skateboarding game that has a cutscene that shows a likeness of Tony Hawk 14:12:36 I probably connected it to you because of that list 14:12:54 ``` cat wisdom/*people*taneb* 14:12:55 cat: 'wisdom/*people*taneb*': No such file or directory 14:12:57 uh 14:13:01 ``` cat wisdom/*taneb*not* 14:13:02 cat: 'wisdom/*taneb*not*': No such file or directory 14:13:07 oh 14:13:15 ``` cat /hackenv/wisdom/*taneb*not* 14:13:16 Taneb is not elliott, a rabbi, Mark Zuckerberg, James Bond, Queen Elizabeth the first, or anyone older than Queen Elizabeth the Second. Pending approval: Shigeru Miyamoto. 14:13:18 right 14:13:24 anyway, that list 14:18:20 Anyway, I am not Tony Hawk, I can't even skateboard 14:37:03 good to know 14:37:19 the mystery thickens, who is Taneb then? 14:37:40 I am 14:54:43 could one person be different people simultaneously? 15:09:24 `? dokalf 15:09:25 dokalf? ¯\(°​_o)/¯ 15:11:55 -!- Lord_of_Life_ has joined. 15:13:21 -!- Lord_of_Life has quit (Ping timeout: 258 seconds). 15:14:48 -!- Lord_of_Life_ has changed nick to Lord_of_Life. 15:31:11 -!- xkapastel has quit (Quit: Connection closed for inactivity). 15:47:01 I think quite often these things turn out to have been some sort of a collective, like Nicolas Bourbaki. 15:47:35 fizzie: which things? Taneb? or Tony Hawk? 15:47:57 or a rabbi? or Queen Elizabeth the Second? that would explain how she reigns for so long. 15:48:15 I was thinking of Taneb, but maybe some of those others too. 15:49:02 Tony Hawk can't be a very large collective since he fits on a single skateboard and in a single helmet 15:59:30 If I was a collective I'd probably be more active 16:00:11 Taneb: maybe you're a collective representing a secret organization, which is why you're not too active? 16:00:32 b_jonas: I do not think that quite works out 16:01:19 `8ball is Taneb James Bond? 16:01:20 You may rely on it. 16:01:58 thats not a bad fortune, you can work with that 16:03:47 kspalaiologos: ok, that's a better guess, because many people think that James Bond is a collective, or an inheritable title 16:17:34 I'm confused 16:17:43 fungot, what just happened 16:17:43 kspalaiologos: the variable " foobar" 17:00:30 win 6 17:08:47 -!- FreeFull has joined. 18:28:13 I saw one program has the following license terms: This program is free and open source under the MIT license. Additionally, any discordian pope may automatically grant themselves a license under the WTFPL, as desired. 18:35:16 Since WTFPL allows any relicensing, and since there are many discordian popes ("every man, woman and child on this Earth is a genuine and authorized pope"), it probably means it's not hard to get a copy of that program under any other license either. 18:44:55 Is Taneb a discordian pope? 18:50:22 I assume so 18:50:27 it's opt-out 18:55:52 I haven't knowingly opted out 18:56:03 who isn"t? 18:56:25 That said, I haven't knowingly opted out of any other sort of papity 18:57:51 I think most of them are opt-in 18:58:03 I'm also not sure if there is a formal procedure to opt out of Discordianism popehood 18:58:31 I suppose I did that and became a mome instead 18:58:44 I had the official card in my wallet for a long time 18:58:53 along with a card listing the axioms of ZFC in Metamath notation 18:59:05 (which also included propositional and predicate logic on the backside as a prerequisite) 18:59:07 > map (pred . pred) "pope" 18:59:10 "nmnc" 18:59:11 I used to be such a nerd 18:59:14 I still am, but I used to, too 19:00:21 that was around the time of my GEB phase 19:00:25 everyone has a GEB phase, right? 19:02:44 don't remind me 19:02:49 technically, you need to opt-in to discordianism, or you don't 19:04:14 [[UnoScript]] N https://esolangs.org/w/index.php?oldid=68811 * *berlinquin * (+2508) Created page with "'''UnoScript''' is a stack-based, interpreted language inspired by the [https://en.wikipedia.org/wiki/Uno_(card_game) UNO card game]. It aims to be as general-purpose and usab..." 19:13:05 [[UnoScript]] https://esolangs.org/w/index.php?diff=68812&oldid=68811 * *berlinquin * (+99) Add categories 19:14:53 [[Language list]] https://esolangs.org/w/index.php?diff=68813&oldid=68716 * *berlinquin * (+16) /* U */ 19:21:38 -!- stux has quit (Quit: Aloha!). 19:21:59 -!- stux has joined. 19:33:29 kmc: yes. and a Smullyan phase too. the Smullyan phase lasts longer. 19:37:36 Smullyan is cool 19:37:46 and I still think GEB is a good book, it's just that the speculative parts are... quite speculative 19:37:49 was, sadly 19:37:58 it's interesting but it doesn't explain everything about the world like I thought it did when I was 16 19:38:25 and the part which teaches metamathematics is both accessable and precise. it's not dumbed down 19:40:40 everyone has a GEB phase, right? => mmmaybe. My one was pretty shallow and maybe not a phase at all. I was even lazy to solve any question from there as I was more or less learned on logic and lazy in general. I read the book, found stories amusing, took something from the dual part but I suppose not at a face value. Complex questions have complex answers, would I expect different even some eight (I think?) years ago?.. 19:41:33 oh, Smullyan phase should be cooler maybe yes maybe definitely but I hadn’t read his books at the right time, only various excerpts 19:41:51 and I still think GEB is a good book, it's just that the speculative parts are... quite speculative => nailed it 19:48:28 I’d like an accessible book on spinors and some simplified quantum-theoretical things, but essentially they should be quite separable. Spinors are interesting in their own way, and in a sense they are even applicable to rare mundane problems, though their transformations, i. e. spin groups (and there are sometimes miscommunication in people occasionally calling elements of a spin group, a spinor, which it isn’t), as I was to say, spin groups are m 19:48:28 ore applicable as they act on vectors too and quaternions in rotating 3D are precisely the 3-dimensional spin group, as are complex numbers a 2D one and split-quaternions a 4D one etc. etc. clifford algebras blah blah 19:49:33 along with a card listing the axioms of ZFC in Metamath notation => whyy? 19:50:29 why not 19:56:34 -!- kspalaiologos has quit (Quit: Leaving). 20:01:39 reasonable! 20:03:24 . o O ( Yay, the moment your program has been running for half an hour and you realise you forgot to compile it with optimizations enabled ) 20:03:55 (it was supposed to finish in 15 minutes ;) ) 20:12:21 :D rofoldl 20:19:53 -!- imode has joined. 20:32:31 -!- kspalaiologos has joined. 20:49:30 -!- stux has quit (Quit: Aloha!). 20:57:18 -!- kspalaiologos has quit (Quit: Leaving). 20:58:18 -!- imode has quit (Ping timeout: 265 seconds). 20:58:53 -!- imode has joined. 21:02:28 -!- stux has joined. 21:33:12 -!- imode has quit (Ping timeout: 260 seconds). 21:43:29 Such intuitive code... http://paste.debian.net/1125756/ 21:50:16 And I should use one more variable rather than using the second half of one of those registers. 21:53:25 (But it's also so much faster than the naive code it replaced, a factor of 30 for what I've been doing.) 21:53:51 SSE, fun 21:54:17 (fortunately I don't have to worry about portability :) ) 21:54:41 requirement: "runs on my PC" 21:55:18 yay :) 21:58:12 [[NewFuck]] https://esolangs.org/w/index.php?diff=68814&oldid=68808 * DmilkaSTD * (+96) 22:00:23 -!- kspalaiologos has joined. 22:03:37 int-e: my way to handle portability is to just make it give a reasonable compile time #error when there's no implementation for how you're compiling 22:03:58 or, if that's hard, a runtime error with a reasonable message 22:07:07 it's also not hard to use that gcc feature I forgot the name of 22:07:19 which selects between multiple versions of a function at load time 22:07:25 based on cpu flags 22:07:31 it might be more of a glibc feature, but anyway 22:07:39 it's good to have a portable fallback anyway for testing 22:08:01 it does introduce a function pointer indirection though 22:08:09 you can do the same thing with dynamic code patching but that's more complicated 22:10:37 -!- Lord_of_Life has quit (Ping timeout: 268 seconds). 22:10:48 'GNU indirect functions are an extension of ELF that allows you to make a decision about which implementation of a function to use at [dynamic] link time.' 22:11:09 -!- Lord_of_Life has joined. 22:11:20 I didn't know what it was called either but I did come across the concept at some point. 22:14:29 kmc: sure but you usually don't want to actualy write the other implementation 22:15:12 sounds like int-e already did 22:15:24 typically you would do a portable version first, then the optimized 22:15:26 and I think it's good to have a reference implementation anyway 22:15:29 for testing 22:15:41 I wouldn't. I would do premature optimization. whether it's a part I need to optimize or not. 22:15:44 if you have both then it's very easy to test the fancy vector code just by throwing random inputs and comparing outputs 22:15:47 lol 22:15:50 -!- stux has quit (Read error: Connection reset by peer). 22:15:59 on top of that the portable code is probably easier to read if someone wants to understand what's going on 22:16:07 if it doesn't have to be fast at all, it can be very clear 22:16:14 -!- stux has joined. 22:16:54 kmc: it's still around as a reference implementation anyway 22:18:57 which is deliberately simple and not efficient, even for processors without fancy carryless multiplication instructions: http://paste.debian.net/1125758/ (and still not quite portable; it's missing the UINT64_C thingy) 22:22:14 isn't uint64_t standard in C99 and later? 22:22:24 what is this code for? 22:22:39 kmc: UINT64_C is the macro for making a uint64_t literal. 22:22:44 huh 22:22:58 is it something like #define UINT64_C(x) x ## ull 22:22:59 ? 22:23:07 I never knew that was needed 22:23:08 Because you can't really know whether the suffix should be empty, ul or ull... 22:23:11 mhm 22:23:14 yay for C :eyeroll: 22:23:26 I was just going to ask if you need a ull actually 22:23:38 better put ullllll just in case 22:23:50 s/empty/U/ (those are capital, aren't they? 22:24:00 hm 22:24:01 maybe 22:24:19 UINT64_C makes a uint_least64_t constant, actually. 22:24:32 I've always found that a little odd, but that's how it is. 22:24:34 good enough? 22:24:41 you could cast it 22:24:51 I guess there are corner cases where that becomes an issue. 22:25:03 I assume it's mostly because uint64_t isn't required to exist. 22:25:10 But uint_least64_t is. 22:25:14 oh really 22:25:17 interesting 22:25:37 It's required to be defined, but only if the implementation provides a suitable type. 22:26:50 C18 7.20.1.1 Exact-width integer types: "These types are optional. However, if an implementation provides integer types with widths of 8, 16, 32, or 64 bits, no padding bits, and (for the signed types) that have a two's complement representation, it shall define the corresponding typedef names." 22:28:56 Yeah that sounds like C. 22:29:48 I'm not really sure what UINT64_C and friends are for, I guess they're just supposed to look nicer than the corresponding cast. 22:29:48 -!- kspalaiologos has quit (Read error: Connection reset by peer). 22:30:31 does POSIX amend this? 22:30:45 I think POSIX might require 8, 16, 32 and 64, yes. 22:31:29 Oh, heh. Actually, only 8, 16 and 32. 22:32:31 https://pubs.opengroup.org/onlinepubs/9699919799/basedefs/stdint.h.html "[CX] The following types are required: int8_t, int16_t, int32_t, uint8_t, uint16_t, uint32_t. If an implementation provides integer types with width 64 that meet these requirements, then the following types are required: int64_t, uint64_t" 22:33:19 Seems sane enough, honestly... 32 bit platforms are not that long gone, probably still lingering in may corners :) 22:33:26 *many 22:33:36 Thing is, uint_least64_t and uint_fast64_t are still mandatory. 22:33:55 Yeah that is strange :) 22:34:28 . o O ( diverging realities are fun ) 22:35:01 I guess maybe the reasoning was, the implementation might have to do something weird to support at-least-64-bit types, and might have a harder time providing the exact-width type. 22:35:21 I mean the Intel 8051 8-bit architecture from 1980 is still around 22:35:29 so yeah 32 bit is still common 22:35:45 32 bit is only starting to become common in embedded 22:36:02 Hmm I suspect POSIX never worried about 8 bit architectures :) 22:36:12 yeah but they do have C compilers 22:36:14 to varying degrees 22:36:19 16 bit though, hmm. Tricky :) 22:36:33 the PIC XC8 compiler has a "short long" type :) 22:36:54 Is that a 24-bit type? 22:36:56 yes 22:37:07 -!- imode has joined. 22:37:10 Yes, C accomodates all sorts of niches, which goes a long way towards explaining their reluctance to mandate types. I'm really surprised to hear about uint_least64_t being madatory. 22:37:24 avr-gcc has it too, but calls it something sensible: __uint24 22:37:52 I suppose short long isn't worse than long double. 22:37:59 The Motorola DSP56K C compiler has a 24-bit 'int' and a 48-bit 'long'. 22:38:00 gcc is more principled than most embedded compilers about how it handles extensions, syntactically 22:38:01 (intel's 10 byte version) 22:38:08 int-e: ah 22:38:09 (And a 24-bit 'char' as well.) 22:38:13 is it 128 bit on ppc? 22:38:17 (now mostly extinct because of SSE) 22:40:02 I think GCC on PPC did the double-double trick for long double. 22:40:40 fancy 22:41:19 Where your extended-precision type is actually just two double-precision floats, organized so that you get about the same range as a regular double, but more or less double the precision. 22:41:26 (IIRC Knuth covers this topic but I skipped that part.) 22:43:18 I was thinking about that some day but wasn’t very sharp to think it in any detail 22:44:52 I didn't know about GNU indirect functions. 22:46:47 Fun fact: the typing of (non-U-suffixed) octal/hexadecimal literals is different than those of decimal constants. On (e.g.) x86-64, 0x80000000 is an `unsigned int`, while 2147483648 is a `long`. 22:47:43 https://ideone.com/aLGNHI as a corollary. 22:47:46 ugh 22:48:41 I didn't know that either; why is that? Isn't there a U and L suffix so that you can use that to indicate it? 22:49:10 Yes, 2147483648U would be an unsigned int too. 22:49:46 (And, you can add a dot if you want a floating number.) 22:49:51 The way it works, the type (decimal, vs. octal/hexadecimal) and suffix (none, U, L, UL, LL, ULL) determine the possible list of types, and then the constant gets the first type on the list that it can be represented in. 22:50:39 `int`, `long` and `long long` is the list for unsuffixed decimal, but the list for unsuffixed octal/hexadecimal is `int`, `unsigned`, `long`, `unsigned long`, `long long`, `unsigned long long`. 22:50:41 int`,? No such file or directory 22:50:47 HackEso: Sorry. 22:52:08 O, that is how it works. 22:53:35 The official reason is: "Unlike decimal constants, octal and hexadecimal constants too large to be `int`s are typed as `unsigned int` if within range of that type, since it is more likely that they represent bit patterns or masks, which are generally best treated as unsigned, rather than 'real' numbers." 22:54:34 OK, although that still look like strange to me, and even then perhaps it might should depend how many leading zeros too (does it count the leading zeros?) 22:55:36 Any (nonzero) amount of leading zeros makes a constant octal, so... kinda? But not really. 22:56:20 No, I mean additional leading zeros for a octal or hexadecimal constant (after the initial 0 or 0x) 22:57:22 Oh. No, those don't matter. 23:39:05 -!- imode has quit (Ping timeout: 246 seconds). 23:40:08 -!- imode has joined. 23:53:07 -!- xkapastel has joined.