| 00:00:02 | * | junland quit (Quit: %ZNC Disconnected%) |
| 00:00:38 | audiofile | thanks will look into fidget |
| 00:01:13 | audiofile | wow that looks real clean m8 |
| 00:01:47 | * | junland joined #nim |
| 00:01:53 | audiofile | just to clarify, if i write a nim script with this library and end up packaging to a binary, it is self-conttained right |
| 00:02:35 | audiofile | ohhhh |
| 00:02:43 | audiofile | i remember fidget, some dude did a demo in nimconf |
| 00:02:55 | FromDiscord | <Elegant Beef> "Some dude" being treeform 😄 |
| 00:03:05 | audiofile | they not here |
| 00:03:20 | FromDiscord | <Elegant Beef> Huh? |
| 00:03:20 | audiofile | but yes, it is treefrog as you say |
| 00:03:28 | FromDiscord | <Elegant Beef> Treeform uses discord |
| 00:03:36 | audiofile | ah ok |
| 00:03:37 | audiofile | irc |
| 00:22:18 | * | krux02_ quit (Remote host closed the connection) |
| 00:24:42 | FromDiscord | <19> shashlick: trying to wrap https://github.com/Kode/Kinc |
| 00:32:43 | * | gmaggior quit (Quit: Leaving) |
| 00:41:00 | FromDiscord | <PizzaFox> best way to generate most random numbers? |
| 00:41:05 | FromDiscord | <PizzaFox> ideally without using OS specific apis |
| 00:41:30 | FromDiscord | <19> https://en.wikipedia.org/wiki/Mersenne_Twister |
| 00:42:37 | FromDiscord | <PizzaFox> is this ture |
| 00:42:38 | FromDiscord | <PizzaFox> (edit) 'ture' => 'true' |
| 00:43:12 | FromDiscord | <PizzaFox> https://nim-lang.org/docs/mersenne.html↵this doesnt explicitly say not to use it for crypto like `random` does, time to go ape mode |
| 00:43:33 | FromDiscord | <Rika> no |
| 00:43:41 | FromDiscord | <Rika> mersenne is even weaker than what random uses |
| 00:43:49 | FromDiscord | <PizzaFox> yeah i figured |
| 00:44:08 | FromDiscord | <PizzaFox> random reddit thread says read `urandom` (only linux), another comment says use libsodium |
| 00:44:21 | FromDiscord | <PizzaFox> surely there is a better method |
| 00:44:41 | FromDiscord | <19> mersenne not so good for crypto? |
| 00:44:54 | FromDiscord | <PizzaFox> > The Mersenne Twister is a pseudorandom number generator |
| 00:44:56 | FromDiscord | <PizzaFox> pseudorandom |
| 00:44:58 | FromDiscord | <PizzaFox> = bad |
| 00:45:04 | FromDiscord | <19> got it |
| 00:45:11 | FromDiscord | <PizzaFox> > Is not cryptographically secure, unless the CryptMT variant (discussed below) is used. The reason is that observing a sufficient number of iterations (624 in the case of MT19937, since this is the size of the state vector from which future iterations are produced) allows one to predict all future iterations. |
| 00:46:03 | FromDiscord | <19> so what is the other classification called other than pseudorandom |
| 00:46:34 | FromDiscord | <PizzaFox> true random |
| 00:46:45 | FromDiscord | <19> cool |
| 00:48:24 | FromDiscord | <Rika> true random is technically os-specific by design |
| 00:51:25 | FromDiscord | <PizzaFox> so nim doesnt have a builtin way to generate random bytes |
| 00:51:35 | * | FromDiscord quit (Remote host closed the connection) |
| 00:51:49 | * | FromDiscord joined #nim |
| 01:15:17 | federico3 | there's no true random generator on most hardware |
| 01:15:22 | federico3 | only CSPRNG |
| 01:30:15 | * | apahl quit (Ping timeout: 272 seconds) |
| 01:31:34 | * | apahl joined #nim |
| 01:44:55 | * | arecacea1 quit (Remote host closed the connection) |
| 01:45:29 | * | arecacea1 joined #nim |
| 01:49:18 | * | bouzu__ quit (Quit: Leaving) |
| 01:51:40 | disruptek | Error: inconsistent typing for reintroduced symbol 'nobj': previous type was: object; new type is: object |
| 02:11:58 | * | a_chou joined #nim |
| 02:30:43 | shashlick | 19: will try later tonight - which os |
| 02:35:01 | * | thomasross quit (Ping timeout: 264 seconds) |
| 03:08:57 | * | r4vi quit (Ping timeout: 246 seconds) |
| 03:09:06 | * | r4vi joined #nim |
| 03:19:02 | * | Zectbumo joined #nim |
| 04:01:40 | * | audiofile quit (Quit: Default Quit Message) |
| 04:06:02 | * | supakeen quit (Quit: WeeChat 2.9) |
| 04:06:46 | * | supakeen joined #nim |
| 04:50:55 | disruptek | i can't even. |
| 05:15:27 | * | narimiran joined #nim |
| 05:53:41 | * | solitudesf joined #nim |
| 06:24:42 | * | waleee-cl quit (Quit: Connection closed for inactivity) |
| 06:31:38 | * | a_chou quit (Quit: a_chou) |
| 06:52:32 | * | gmaggior joined #nim |
| 07:17:14 | * | pietroppeter joined #nim |
| 07:44:35 | * | arecacea1 quit (Remote host closed the connection) |
| 07:45:10 | * | arecacea1 joined #nim |
| 08:01:16 | * | narimiran quit (Ping timeout: 246 seconds) |
| 08:06:48 | * | bouzu_ joined #nim |
| 08:23:32 | * | hasbeen joined #nim |
| 08:24:27 | hasbeen | Hi. In nim, can I ensure that this[1][2] doesn't happen? ie. I can somehow reference the in-class field specifically? [1] https://paste.debian.net/1162826/ [2] https://paste.debian.net/1162827/ |
| 08:28:14 | * | Zectbumo quit (Remote host closed the connection) |
| 08:30:48 | * | antranigv quit (Ping timeout: 256 seconds) |
| 08:31:44 | * | antranigv joined #nim |
| 08:43:52 | * | bouzu__ joined #nim |
| 08:46:22 | * | bouzu_ quit (Ping timeout: 260 seconds) |
| 08:54:44 | * | krux02 joined #nim |
| 08:56:08 | * | hasbeen quit (Quit: Leaving) |
| 09:22:21 | * | hnOsmium0001 quit (Quit: Connection closed for inactivity) |
| 09:53:43 | Oddmonger | can i import only unindent from strutils ? |
| 09:54:02 | Oddmonger | yes i can, but not with «import unindent from strutils» |
| 09:59:57 | Oddmonger | ok: from « strutils import unindent » (i was _so_ sure of the above 🤔 ) |
| 10:14:18 | Zevv | How could I optimize this so that I don't need the intermediate Row type to implement the double [y][x] indexing? http://ix.io/2whM |
| 10:16:43 | * | muffindrake joined #nim |
| 10:17:01 | * | pietroppeter quit (Quit: Connection closed for inactivity) |
| 10:26:06 | * | guelosk[m] quit (*.net *.split) |
| 10:26:06 | * | skrylar[m] quit (*.net *.split) |
| 10:26:06 | * | silvernode[m] quit (*.net *.split) |
| 10:26:08 | * | leorize[m] quit (*.net *.split) |
| 10:26:08 | * | l1x quit (*.net *.split) |
| 10:26:08 | * | stefantalpalaru quit (*.net *.split) |
| 10:26:08 | * | lain quit (*.net *.split) |
| 10:26:08 | * | Cadey quit (*.net *.split) |
| 10:27:53 | * | lain joined #nim |
| 10:32:28 | * | skrylar[m] joined #nim |
| 10:32:28 | * | guelosk[m] joined #nim |
| 10:32:28 | * | silvernode[m] joined #nim |
| 10:32:28 | * | leorize[m] joined #nim |
| 10:32:28 | * | l1x joined #nim |
| 10:32:28 | * | stefantalpalaru joined #nim |
| 10:32:28 | * | Cadey joined #nim |
| 10:43:03 | * | Trustable joined #nim |
| 10:59:20 | * | Vladar joined #nim |
| 11:11:56 | * | krux02_ joined #nim |
| 11:14:25 | * | krux02 quit (Ping timeout: 240 seconds) |
| 11:20:16 | stisa[m] | Zevv any reason you don't want [y, x] for indexing? |
| 11:30:42 | * | shodan45 quit (Quit: No Ping reply in 180 seconds.) |
| 11:35:03 | * | shodan45 joined #nim |
| 11:38:11 | Zevv | because people come and ask for it all the time |
| 11:38:35 | Zevv | it's one of these things that come op on #nim |
| 11:39:12 | Zevv | of course, "just use a 1d array and do the indexing yourself" is a perfectly fine answer. But I was wondering if I could abstract this away for dynamic sized arrays |
| 11:52:39 | * | Hotbees quit () |
| 12:04:34 | stisa[m] | Mmh ok, I've used a concept that matches data is seq[T] , width is int for [y,x] , but for [y][x] I don't know, sorry |
| 12:04:48 | Oddmonger | is it possible to alias a function ? |
| 12:05:04 | Oddmonger | like var shortname = myfunctionwithaverylongname |
| 12:06:02 | * | supakeen quit (Quit: WeeChat 2.9) |
| 12:06:39 | stisa[m] | I think something is wrong with discord bot, irc -> discord works but discord->irc doesn't seem to work |
| 12:06:41 | * | supakeen joined #nim |
| 12:09:41 | * | FromDiscord quit (Remote host closed the connection) |
| 12:09:59 | * | FromDiscord joined #nim |
| 12:12:08 | stisa[m] | @Oddmonger maybe `template short(a: varargs[untyped]) = longname(a)` ? |
| 12:12:56 | FromDiscord | <Hearthstone (LGBTQ+)> I'm assuming that's a no on Matrix- |
| 12:12:57 | Oddmonger | thank you stisa[m] , i was seeing that a bit more easier |
| 12:14:03 | stisa[m] | @Heartstone oh it started working again |
| 12:14:06 | Oddmonger | well in fact |
| 12:14:24 | Oddmonger | var bar = foo # work in inim, but not in my source |
| 12:14:31 | Oddmonger | (foo is a proc) |
| 12:15:27 | stisa[m] | oh you're right, works in the playground too |
| 12:15:28 | Oddmonger | ok it was another error in fact, it's working |
| 12:18:53 | Oddmonger | and while searching that, i have discovered that modules can be renamed easily |
| 12:19:11 | Oddmonger | from mymodule as m import nil |
| 12:19:18 | Oddmonger | m.x() |
| 12:19:33 | FromDiscord | <Rika> you lose ufcs that way |
| 12:19:52 | FromDiscord | <Hearthstone (LGBTQ+)> Oh cool |
| 12:20:26 | * | pietroppeter joined #nim |
| 12:21:08 | Oddmonger | Rika: i don't understand the link with UFCS |
| 12:21:30 | FromDiscord | <Rika> you must qualify all procs that way |
| 12:21:49 | Oddmonger | ah, when you rename module ? |
| 12:21:52 | FromDiscord | <Rika> you cannot use `obj.proc()` anymore since it must be `module.proc(obj)` |
| 12:21:57 | FromDiscord | <Rika> when you use `import nil` |
| 12:22:05 | FromDiscord | <Rika> there is no reason to rename a module without `import nil` |
| 12:22:30 | Oddmonger | import nil , is for importing everything from the module, isn't it ? |
| 12:22:33 | FromDiscord | <Rika> well there is when a clash happens but that is rare enough |
| 12:22:48 | FromDiscord | <Rika> `import nil` is for requiring qualified import |
| 12:23:42 | FromDiscord | <Rika> use `import xxx as yyy` if you want to use unqualified but renamed import |
| 12:23:51 | Oddmonger | ah i understand |
| 12:23:54 | FromDiscord | <Rika> though as ive said its only useful when there is a clash |
| 12:24:11 | Oddmonger | you import nothing by default, and must qualify everything, then |
| 12:24:14 | Oddmonger | so no UFCS |
| 12:24:16 | Oddmonger | ok |
| 12:25:02 | Oddmonger | may be we can do this another way… |
| 12:25:37 | Oddmonger | import strutils as su , is working but as you said no interest |
| 12:25:56 | Oddmonger | because functions are all in the global space |
| 12:26:08 | Oddmonger | it's clearer now, thank you |
| 12:27:11 | Oddmonger | i'm using inim often, to learn nim and test things: is it possible to list the objects imported with the repl ? |
| 12:27:58 | Oddmonger | like in Lua: for i,v in pairs(_G) do print(i,v) end |
| 12:28:39 | FromDiscord | <Rika> i do not know, i dont use inim |
| 12:29:12 | FromDiscord | <Hearthstone (LGBTQ+)> If you can list all the imported modules in Nim normally, you should be able to do the same in Inim |
| 12:29:28 | Oddmonger | but the import is just done for the compilation stage, or is it accessible from the AST ? |
| 12:30:24 | Oddmonger | i have no idea how to coin this |
| 12:32:18 | * | lritter joined #nim |
| 12:41:40 | FromDiscord | <19> sent a code paste, see https://play.nim-lang.org/#ix=2wiz |
| 12:43:20 | FromDiscord | <Rika> yes |
| 12:43:29 | FromDiscord | <Rika> actually |
| 12:43:39 | FromDiscord | <Rika> `pipeline:kinc_g4_pipeline_ptr` -> `pipeline:kinc_g4_pipeline_t_ptr` you missed the t |
| 12:44:39 | FromDiscord | <19> well its giving me SIGSEGV |
| 12:44:45 | FromDiscord | <19> how can i debug it |
| 12:45:22 | FromDiscord | <Rika> im not sure, someone else might know |
| 12:45:44 | FromDiscord | <19> sent a code paste, see https://play.nim-lang.org/#ix=2wiA |
| 12:51:39 | * | gmaggior quit (Ping timeout: 265 seconds) |
| 12:51:49 | FromDiscord | <Rika> well pipeline is nil |
| 12:51:58 | FromDiscord | <Rika> thats why youre getting the sigsegv |
| 12:52:47 | FromDiscord | <19> oh yeah |
| 12:53:50 | ehmry | Zevv: can I pm you with some npeg questions? |
| 12:54:20 | FromDiscord | <Clyybber> or post them here :) |
| 12:55:07 | ehmry | I'm trying to figure out a good strategy to deal with backtracking |
| 13:02:35 | ehmry | Clyybber: are you familiar with peg parsing? |
| 13:04:20 | * | gmaggior joined #nim |
| 13:06:54 | FromDiscord | <haxscramper> What are dependencies for installing nimterop in addition to just nimble? When I install it normally everything works, but inside docker `[email protected]` fails to install. Installation using `nimble install nimterop`, nim 1.2.6 |
| 13:07:41 | FromDiscord | <Clyybber> ehmry: Not too much, I asked Zevv questions yesterday too. |
| 13:08:39 | ehmry | so I have a big grammar https://git.sr.ht/~ehmry/dhall-nim/tree/master/src/dhall/parse.nim |
| 13:09:48 | ehmry | I'm trying to break string matches into "terms", I understand that can npeg can walkback string matches, but I don't know how to correctly walk back the terms that I parse in capture blocks |
| 13:13:33 | ehmry | maybe I just don't understand the lexer example |
| 13:16:45 | * | awe00 joined #nim |
| 13:34:37 | * | gmaggior quit (Quit: Leaving) |
| 13:35:13 | Araq | Clyybber: I'm making cursor inference "scope" aware |
| 13:35:22 | Araq | but here is a nasty problem: |
| 13:35:39 | Araq | var x: seq[string]; x.add "a" |
| 13:36:12 | Araq | how do we know that 'x' is not a cursor? |
| 13:38:09 | FromDiscord | <Clyybber> emhry: Sorry, don't know, probably best to ask Zevv |
| 13:38:31 | FromDiscord | <Clyybber> Araq: Hmm |
| 13:38:46 | FromDiscord | <Clyybber> Why does it matter? |
| 13:39:46 | Araq | well we better destroy this 'x' |
| 13:39:59 | Araq | but I think it's just a bug in my implementation |
| 13:40:11 | FromDiscord | <Clyybber> Araq: Also regardinng semiStmtList, I'm not sure whats an attribute grammar, we simply parse an ifExpr instead of an ifStmt when we are in a semiStmtList context |
| 13:40:17 | Araq | and my design is good |
| 13:40:24 | FromDiscord | <Clyybber> I can move the branching into semiStmtList |
| 13:40:36 | FromDiscord | <Clyybber> Araq: Yeah, I think so too |
| 13:41:06 | Araq | well so move the branching into semiStmtList |
| 13:41:14 | FromDiscord | <Clyybber> Ok, will do. |
| 13:41:40 | Araq | if it avoids the parameter to complexOrSimpleStmt(maybeExpr = true) |
| 13:41:51 | disruptek | is there really no way to achieve static generics? |
| 13:42:08 | Araq | what's a static generic? |
| 13:42:22 | disruptek | a proc that takes static[T] |
| 13:42:54 | FromDiscord | <Clyybber> That is possible. |
| 13:43:03 | disruptek | i finally impl jason properly and discovered bug after bug. |
| 13:43:13 | FromDiscord | <Clyybber> so report them |
| 13:44:33 | disruptek | well, you can think about it and get back to me. 😁 |
| 13:54:45 | * | arecacea1 quit (Remote host closed the connection) |
| 13:55:39 | * | arecacea1 joined #nim |
| 13:59:12 | Araq | disruptek: waiting on your concepts improvement |
| 14:00:09 | * | tane joined #nim |
| 14:02:32 | shashlick | @haxscramper what's the failure you see, can you share a snippet |
| 14:02:37 | * | krux02_ quit (Read error: Connection reset by peer) |
| 14:04:14 | * | krux02_ joined #nim |
| 14:05:31 | FromDiscord | <haxscramper> I don't see any compilation errors. After running `nimble --vebose install nimterop` it just fails and prompts for installation for `#head`. |
| 14:06:27 | FromDiscord | <haxscramper> It is quite possible that I'm broke something on my side/missing something assumed be available default on linux |
| 14:07:36 | FromDiscord | <haxscramper> It fails on `toast.nim` compilation |
| 14:08:45 | FromDiscord | <haxscramper> Full installation log: http://ix.io/2wje |
| 14:12:29 | Zevv | Do not fear, the npeg support man is here! |
| 14:13:10 | Zevv | emhry: whats up |
| 14:13:42 | Zevv | i guess you got the same issue as everyone |
| 14:13:52 | Zevv | code block caputres dont backtrack, right? |
| 14:14:21 | Zevv | Why does no one complain about that with lpeg, which has the same problem, i wonder |
| 14:16:41 | FromDiscord | <Clyybber> sent a code paste, see https://play.nim-lang.org/#ix=2wjg |
| 14:17:30 | * | haxscramper joined #nim |
| 14:22:54 | Zevv | @disruptek: ^^ |
| 14:29:05 | FromDiscord | <Recruit_main707> smart move there |
| 14:30:14 | * | pietroppeter quit (Quit: Connection closed for inactivity) |
| 14:31:42 | Zevv | wha? |
| 14:37:42 | ehmry | Zevv: yea, same as everyone, code block backtrack |
| 14:37:49 | Zevv | bwah |
| 14:38:03 | Zevv | it's one the problems I hope to go away by ignoring it for a long time |
| 14:38:12 | Zevv | I think it'll take another 50 years |
| 14:38:55 | Zevv | problem is that code blocks run at match time and are also used for validation and custom matchers |
| 14:39:12 | Zevv | my usual answer is: if you need backtracking, your syntax is bad. But that doesn't help anyone |
| 14:41:56 | ehmry | yea, my syntax has "minimal" backtracking, but thats different from no backtracking |
| 14:43:10 | ehmry | Zevv: could memoization help? |
| 14:45:22 | ehmry | or rule entry and exit hooks? |
| 14:48:13 | * | theelous3 joined #nim |
| 14:48:57 | * | NimBot joined #nim |
| 14:49:21 | disruptek | Araq: waiting on your feedback for IC. concepts has no deadline. 😉 |
| 14:49:23 | ehmry | idk, I'll play around some more |
| 14:49:37 | Zevv | basically, the call into Nim would need to be postponed until any open choices are comitted |
| 14:49:40 | disruptek | Zevv: what's up? |
| 14:49:54 | Zevv | but that would change order in the calls, and breaks the validation-by-code |
| 14:50:39 | Zevv | alternative would be to allow some syntax to register code for a callback later. So the block itself gets called at match time, but you can add a "onfinal:" block or something similar, which will be stashed away until the match is certain |
| 14:51:07 | Zevv | disruptek: Clyybber posted something for you, but the discord bridge dropped your name it seems. It's put into a silly nim-playground snippet |
| 14:51:32 | disruptek | oh i don't read discord anymore. |
| 14:53:33 | FromDiscord | <mratsim> @Clyybber now Higher Kinded types 😉 |
| 14:53:43 | FromDiscord | <mratsim> they work but only if you don't specify them |
| 14:54:24 | FromDiscord | <mratsim> i.e. proc[A, B: Foo[A]](a: A, b: B) |
| 14:54:51 | disruptek | Error: the macro body cannot be compiled, because the parameter 'j' has a generic type |
| 14:55:28 | FromDiscord | <mratsim> this would work but then you need to check the Foo[A] constraint↵↵proc[A, B](a: A, b: B) = static: doAssert B is Foo[A] |
| 14:56:20 | disruptek | Zevv: i handle that npeg problem in nim. if there's a way to do it in npeg, great, but to me it's no big deal. |
| 14:57:16 | disruptek | anyone on irc know what that error message means? |
| 15:08:02 | Zevv | disruptek: well, how would you know in nim that particular part of your grammar backtracked? There's no signalling for that |
| 15:10:01 | shashlick | @haxscramper you don't have g++ installed |
| 15:10:05 | shashlick | 'cc1plus': execvp: No such file or directory |
| 15:12:00 | disruptek | Zevv: you're right, i think ehmry's use-case may be more demanding than mine. |
| 15:12:41 | disruptek | i don't really understand why they need to know about backtracks that were fruitless, though. |
| 15:13:46 | Zevv | well, npeg *tries* to capture something, and on the fly calls your code block with the match |
| 15:13:55 | Zevv | but only later figures out this was wrong |
| 15:13:58 | Zevv | but your call was already made. |
| 15:14:03 | Zevv | Maybe we can add uncall to nim |
| 15:14:05 | Zevv | or decall |
| 15:14:12 | disruptek | heh |
| 15:14:30 | disruptek | unComeFrom |
| 15:15:06 | Zevv | I was hoping this never has to be solved, but it's actually the only issue that gets reported |
| 15:15:26 | Zevv | the good news is that Npeg has proved to be almost bug free |
| 15:16:28 | disruptek | when you figure out that you want to backtrack, what do you do? |
| 15:16:32 | Zevv | reading back all reported issues, I can say "bug free until proven otherwise" |
| 15:16:46 | Zevv | The idea is that these calls should be postponed until the choice is comitted |
| 15:16:59 | disruptek | they can go into a stack. |
| 15:16:59 | Zevv | but that will change semantics, your calls come in in reverse order, for example |
| 15:17:08 | Zevv | sure |
| 15:17:23 | disruptek | as long as we have a way to identify them... |
| 15:17:27 | Zevv | but then it breaks the current api where code blocks are called for validation or custom matches |
| 15:17:36 | Zevv | if your code block says "no", the match can fail. That's super handy |
| 15:17:47 | Zevv | but that needs to run at match time |
| 15:17:49 | disruptek | you won't believe this, but i'm not too bothered by breaking other people's code. |
| 15:18:15 | Zevv | alternative is to either add some kind of way to indicate this in your code, maybe by adding some sugar to register a callback that gets called on commit |
| 15:18:25 | Zevv | or add another operator ">" is for match time, ">>" is for commit time |
| 15:18:28 | Zevv | something like that |
| 15:18:49 | disruptek | to keep the api you'd need to just add a way to query the mode. |
| 15:19:02 | Zevv | the npeg paring engine was a piece of cake. But captures were a bit of apain |
| 15:19:17 | Zevv | query what mode? |
| 15:19:34 | disruptek | commit vs. offer |
| 15:19:54 | Zevv | well, it's only ran once. If I run it for two modes, I'd have to call the same block twice |
| 15:19:58 | disruptek | Araq: comments would go a long way to soliciting contributions. |
| 15:20:14 | disruptek | Zevv: so what? |
| 15:20:33 | * | waleee-cl joined #nim |
| 15:20:55 | disruptek | don't you know if you're going to handle commits when you compile the code? |
| 15:21:09 | disruptek | it's not like you can't read it, right? |
| 15:21:35 | disruptek | maybe i'm missing something. |
| 15:22:53 | Zevv | yeah well. For now I choose to still ignore it. LPEG got away with it for 10 years :) |
| 15:22:56 | disruptek | the problem with the operator solution, to me, is that it means i cannot reuse the grammar as easily. |
| 15:23:25 | disruptek | it's a logic issue so i think it should be handled in logic. |
| 15:23:36 | Zevv | It seems that everyone reporting this found a way to solve this. So I'm not sure if fixing this in npeg is worth the hassle |
| 15:23:59 | Zevv | But you can not properly handle it in your own logic, because that logic *can not know* the capture was backtracked |
| 15:24:23 | Zevv | I can make up a grammar where your caputure block will be called 10 times |
| 15:24:29 | Zevv | none of which actually should have happened |
| 15:24:56 | disruptek | i get it. |
| 15:26:12 | ehmry | I've found workarounds in the grammer, but then my grammers are misleading |
| 15:26:48 | ehmry | like matching the first half of a rule, and if the optional second half is missing then I backtrack |
| 15:28:17 | Zevv | But it's optional, so why backtrack |
| 15:29:20 | haxscramper | shashlick: thanks, now it is worked as expected. |
| 15:30:23 | ehmry | so I can throw out the first half of the match |
| 15:31:03 | ehmry | anyway, I'm going to read these npeg issues |
| 15:33:35 | * | Trustable quit (Remote host closed the connection) |
| 15:34:32 | * | Trustable joined #nim |
| 15:36:13 | disruptek | Zevv: sealmove's kaitai thing is done, right? |
| 15:36:32 | FromDiscord | <19> i'm having a weird issue with nim binding, i can get the size of the file and the number of characters but i cant retreive the data from the pointer: https://play.nim-lang.org/#ix=2wjR |
| 15:36:46 | ehmry | backtrack hack -> https://git.sr.ht/~ehmry/dhall-nim/tree/master/src/dhall/parse.nim#L164 |
| 15:44:59 | Zevv | disruptek: dont know, why? |
| 15:45:22 | disruptek | !repo kaitai |
| 15:45:23 | disbot | https://github.com/sealmove/nimitai -- 9nimitai: 11A native Nim macro implementation of Kaitai Struct 15 13⭐ 0🍴 7& 3 more... |
| 15:47:47 | * | kkk joined #nim |
| 15:48:09 | * | kkk is now known as Guest33034 |
| 15:48:45 | * | Guest33034 quit (Remote host closed the connection) |
| 16:12:09 | * | bra-ket joined #nim |
| 16:14:16 | * | zyklon quit (Ping timeout: 244 seconds) |
| 16:16:03 | * | narimiran joined #nim |
| 16:21:42 | * | vicfred quit (Ping timeout: 256 seconds) |
| 16:31:24 | stefantalpalaru | This example is wrong: https://nim-lang.org/docs/system.html#add%2Cseq%5BT%5D%2CopenArray%5BT%5D |
| 16:31:55 | stefantalpalaru | It should be: `s.add(@["test"])` |
| 16:32:47 | federico3 | can nimitai output a .nim file? |
| 16:32:49 | Zevv | nope |
| 16:32:50 | Zevv | not wrong |
| 16:33:01 | stisa[m] | That kaitai things looks pretty cool, I might give it a try later |
| 16:33:44 | Zevv | any scalar passed to an openarray argument will implicitly be converted to an array |
| 16:33:51 | Zevv | although I can't find that in the manual |
| 16:33:57 | federico3 | if it could avoid adding more dependencies at build time... |
| 16:44:08 | * | solitudesf quit (Remote host closed the connection) |
| 16:44:32 | * | Zectbumo joined #nim |
| 16:47:41 | disruptek | the example is wrong, but only because it's attached to the wrong proc. |
| 16:47:47 | disruptek | a string is an openarray. |
| 17:00:26 | * | superbia1 joined #nim |
| 17:02:14 | * | superbia quit (Ping timeout: 260 seconds) |
| 17:02:16 | * | superbia1 is now known as superbia |
| 17:06:40 | * | hnOsmium0001 joined #nim |
| 17:07:21 | * | WilhelmVonWeiner joined #nim |
| 17:09:03 | * | kwilczynski quit () |
| 17:11:25 | Zevv | disruptek: that's bull. |
| 17:11:31 | Zevv | the example is perfectly fine |
| 17:11:58 | disruptek | how do you figure? |
| 17:12:03 | Zevv | well, it works, right? |
| 17:12:07 | Zevv | it does what it says it does? |
| 17:12:11 | disruptek | because it's using the other proc. |
| 17:12:38 | Zevv | oh well wait a second now |
| 17:12:42 | disruptek | the example exemplifies the add(seq[T], T) not the add(seq[T], openArray[T]). |
| 17:12:53 | Zevv | how did I make up the bull about passing a single argument to an openarray making it an array implicitly? |
| 17:13:12 | disruptek | hey, i tried to correct you gently. |
| 17:14:05 | Zevv | don't do that |
| 17:14:08 | Zevv | that's so confusing |
| 17:14:20 | Zevv | talking to a dutchie here, right |
| 17:14:40 | disruptek | lol, true. |
| 17:18:38 | disruptek | i'm trying to fix the compiler so dust works to minimize bugs that prevent macro behavior that i need for frosty to work in the compiler for IC. |
| 17:18:53 | disruptek | it's a many-layered onion of regret. |
| 17:19:17 | Zevv | oh man |
| 17:19:42 | Zevv | the world is so full of nicely isolated problems |
| 17:19:47 | Zevv | and you pick this sh*t |
| 17:20:39 | disruptek | i know. i'm a masochist. |
| 17:24:17 | disruptek | what's sad is i think i can't get mangling in until concepts work in nim-1.0, which, y'know, they don't. |
| 17:25:02 | disruptek | we really need to clone araq a few times. i dunno why i have these clyybbers growing in the vat when what i really want is a few dozen araq. |
| 17:36:03 | * | kwilczynski joined #nim |
| 17:36:13 | Zevv | we can have scotty beam him up and not destroy the original |
| 17:37:05 | * | thomasross joined #nim |
| 17:37:22 | * | muffindrake quit (Quit: muffindrake) |
| 17:38:12 | Zevv | I read a cool book once, there was this professional "proletaric mover", who used star-trek like technology for moving himself somewhere, steal the loot and move himself out ago. And then all the ghosts of his killed originals team up to haunt him because they are jealus of the last copy that actually gets to survive. |
| 17:39:05 | FromGitter | <Quelklef> Would somebody mind taking a look at the following and verifying that it's not expected behaviour before I submit a bug report? ⏎ ⏎ ```code paste, see link``` ⏎ ⏎ issue seems to be that `setimpl.nim(29, 10)` is using my `[]` overload for some reaosn. [https://gitter.im/nim-lang/Nim?at=5f551eb99566774dfe5a2b35] |
| 17:39:05 | disruptek | but that means that when he moves, the entity they were haunting becomes a haunter. |
| 17:39:19 | Zevv | right |
| 17:39:48 | Zevv | and they all are guilty, excepty for the last one who is still alive |
| 17:40:04 | disruptek | does the hauntee know what's up? |
| 17:40:12 | Zevv | he went insane |
| 17:41:16 | disruptek | dust is really dumb. i wish someone would help me with it. |
| 17:41:27 | disruptek | i know what has to change i just... i'm tired. |
| 17:41:37 | Zevv | man I can imagine. you just wont stop |
| 17:41:49 | Zevv | concentrate, focus, one thing |
| 17:41:51 | * | solitudesf joined #nim |
| 17:41:57 | disruptek | it is eating all my memory. |
| 17:42:05 | disruptek | 25gig so far. |
| 17:42:46 | disruptek | whaddya mean i won't stop? |
| 17:42:59 | Zevv | you just do too many things |
| 17:43:10 | disruptek | oops, int128 bug. |
| 17:43:22 | disruptek | just too many damned bugs. |
| 17:43:38 | disruptek | dude, there is so much that needs writing. most of it is dogshit, too. |
| 17:43:39 | voidpi | disruptek: what is dust? |
| 17:43:48 | voidpi | nimble search didn't return anythins |
| 17:43:52 | voidpi | anything* |
| 17:44:37 | disruptek | there are people out there that don't use nimble. |
| 17:44:56 | disruptek | !repo dust |
| 17:44:56 | disruptek | disbot: wtf; you okay, buddy? |
| 17:44:57 | disbot | https://github.com/disruptek/dust -- 9dust: 11DUST is Unattended Syntax Truncation 15 3⭐ 0🍴 7& 1 more... |
| 17:44:57 | disbot | yep. 😊 |
| 17:45:13 | disruptek | must be his lunch hour. |
| 17:46:19 | voidpi | heh |
| 17:46:27 | voidpi | disruptek: thanks |
| 17:47:56 | disruptek | that readme isn't up-to-date. |
| 17:48:02 | disruptek | you run it like the compiler. |
| 17:48:10 | disruptek | dust check bug.nim |
| 17:48:22 | voidpi | It will overwrite your source file during operation. |
| 17:48:30 | voidpi | is that up to date? |
| 17:48:33 | disruptek | yes. |
| 17:49:32 | disruptek | it has to do that because i haven't patched the compiler to make it easier to embed. |
| 17:50:11 | disruptek | dust is an example of erosion of ambition in the face of practicality. |
| 17:50:44 | voidpi | looks like the kind of program that benefits of a taunting module |
| 17:50:54 | voidpi | "i erased all your crap!" |
| 17:52:38 | disruptek | that's the best possible scenario. |
| 17:52:52 | Zevv | haha |
| 17:52:55 | Zevv | very true :) |
| 17:54:12 | disruptek | it's working, at least. |
| 17:54:23 | disruptek | i mean, it doesn't always work, but when it does, it's fuckin' sweet. |
| 18:03:18 | FromDiscord | <UNIcodeX> @dom96 any insight on GC'ed memory error when trying to use jester with threadpool? https://www.pastery.net/shhbrc/ |
| 18:03:41 | disruptek | one thing it should do it monitor unused symbols and then remove them. |
| 18:04:18 | disruptek | and also score mutations by how high or large their removals are. |
| 18:04:22 | * | Vladar quit (Quit: Leaving) |
| 18:07:31 | FromDiscord | <dom96> @UNIcodeX {.threadvar.} on the `db` or disable threads |
| 18:07:49 | * | FromDiscord quit (Remote host closed the connection) |
| 18:08:07 | * | FromDiscord joined #nim |
| 18:09:34 | * | thomasross quit (Ping timeout: 246 seconds) |
| 18:11:00 | FromDiscord | <UNIcodeX> > @UNIcodeX {.threadvar.} on the `db` or disable threads↵@dom96 hmm. ok. doing `db {.threadvar.} = db.getNextConnection()` gets me a different error `Error: a thread var cannot be initialized explicitly; this would only run for the main thread` |
| 18:14:44 | FromDiscord | <dom96> yes, you need to initialise it per-thread |
| 18:15:27 | FromDiscord | <dom96> sent a code paste, see https://play.nim-lang.org/#ix=2wlf |
| 18:23:36 | * | muffindrake joined #nim |
| 18:25:04 | * | the-oaks joined #nim |
| 18:29:51 | * | the-oaks quit (Client Quit) |
| 18:30:10 | * | the-oaks joined #nim |
| 18:30:24 | * | the-oaks quit (Client Quit) |
| 18:31:10 | * | the-oaks joined #nim |
| 18:31:22 | * | the-oaks quit (Client Quit) |
| 18:32:23 | disruptek | seems like there are no issues at all mentioning static[T] and macros. |
| 18:32:37 | disruptek | why can't we use these together? |
| 18:33:33 | disruptek | if base.isMetaType and procKind == skMacro: localError(c.config, info, errMacroBodyDependsOnGenericTypes % paramName) |
| 18:36:15 | Zevv | I also want generic macros. macro foo[T](). and a pony. |
| 18:38:28 | FromDiscord | <UNIcodeX> sent a code paste, see https://play.nim-lang.org/#ix=2wlp |
| 18:39:28 | FromDiscord | <dom96> seems like `bson` is using global vars :/ |
| 18:40:29 | FromDiscord | <UNIcodeX> any way I can fix that? |
| 18:40:48 | FromDiscord | <dom96> you'll have to fix bson I guess, or just disable thread for now and ask the maintainer to |
| 18:41:51 | FromDiscord | <UNIcodeX> maintainer of `bson` mentions `nimongo` for threads, but the initialization and the way you interact with the db is different. I seem to remember name collisions with `%*` |
| 18:42:44 | FromDiscord | <UNIcodeX> well... more accurately... maintainer of `mongopool` |
| 18:47:58 | FromDiscord | <UNIcodeX> > you'll have to fix bson I guess, or just disable thread for now and ask the maintainer to↵@dom96 https://github.com/JohnAD/bson/issues/11 |
| 18:48:00 | disbot | ➥ bson module not thread safe (using globals) ; snippet at 12https://play.nim-lang.org/#ix=2wlu |
| 18:51:14 | FromDiscord | <tomck> hello! I want to eval some nim source code at runtime with a custom env - what're my options for that, or is it basically not possible? |
| 18:52:05 | FromDiscord | <tomck> ideally with decent performance for running many short scripts thousands of times |
| 18:53:42 | * | natrys joined #nim |
| 18:59:11 | FromDiscord | <UNIcodeX> The reason I'm working on switching to Nim&Jester for my back-end is that I'm currently using Python&Sanic, which gives me 2200 requests / second when running on 4 cores using uvloop with hypercorn, whereas Jester on ONE core gives me 3150 requests per second. The machine in question is a simple 4 core AWS lightsail machine. |
| 19:01:15 | * | FromDiscord quit (Remote host closed the connection) |
| 19:01:29 | * | FromDiscord joined #nim |
| 19:08:40 | haxscramper | Is it possible to `import` some code inside `.nimble` file for project? I have `module.foo()` function that is a part of source code and want to use it inside nimble too. |
| 19:08:40 | haxscramper | |
| 19:08:40 | haxscramper | I cannot simply do `import module` or `src/module` inside nimble as it won't work during installation. |
| 19:09:01 | * | sacredfrog quit (Quit: ZNC 1.8.1 - https://znc.in) |
| 19:09:19 | FromDiscord | <tomck> @haxscramper brill! Is it suitable for loads of tiny scripts? I'm building a game engine where scripts are used to customise entity behaviour, so there's likely be a *bunch* of tiny scripts that get run every frame |
| 19:10:58 | FromDiscord | <dom96> @UNIcodeX interesting, if you are successful and have some time I'd love to see a write up of the things you ran into (this being one of the things, we need this kind of feedback to make it better :)) |
| 19:11:00 | FromDiscord | <tomck> oh yeah that code example makes more sense, thanks |
| 19:11:14 | FromDiscord | <dom96> actually, even if you're not successful, in fact especially so |
| 19:11:28 | FromDiscord | <dom96> (posting the write up on the forum would be idal) |
| 19:11:29 | FromDiscord | <dom96> (edit) 'idal)' => 'ideal)' |
| 19:12:19 | FromDiscord | <haxscramper> I can't really comment on how performant nimscript is, but for running **everty frame** you might want to look for a solution that is specifically optimized for performance. |
| 19:13:12 | FromDiscord | <haxscramper> There are several projects for making embedding lua much easier |
| 19:13:35 | FromDiscord | <tomck> will lua be much faster than nim? |
| 19:13:52 | FromDiscord | <lqdev> no |
| 19:13:57 | FromDiscord | <tomck> (nimscript) |
| 19:14:01 | FromDiscord | <lqdev> yes |
| 19:14:02 | * | sacredfrog joined #nim |
| 19:14:06 | FromDiscord | <tomck> i see |
| 19:14:09 | FromDiscord | <lqdev> luajit, that is |
| 19:14:24 | FromDiscord | <tomck> yes, i keep hearing about how magic that is |
| 19:14:37 | FromDiscord | <tomck> is nim's `include` literally just text substitution? |
| 19:14:45 | FromDiscord | <tomck> like, can i `include` in the middl eof a random block? |
| 19:15:09 | FromDiscord | <tomck> or will it mess up the indentation? |
| 19:15:18 | FromDiscord | <lqdev> it's AST substitution |
| 19:15:32 | disruptek | haxscramper: yes, but i'd recommend against it. |
| 19:15:33 | FromDiscord | <lqdev> so it won't mess up your indent |
| 19:15:35 | * | mipri joined #nim |
| 19:15:45 | FromDiscord | <lqdev> though i haven't tried using it inside of a block |
| 19:15:53 | FromDiscord | <tomck> ahhhh cool yeah i see the example in the manual now, it works inside a block just fine |
| 19:16:02 | FromDiscord | <lqdev> TIL |
| 19:18:00 | FromDiscord | <UNIcodeX> sent a long message, see http://ix.io/2wlF |
| 19:18:01 | FromDiscord | <tomck> sent a long message, see https://paste.rs/0yA |
| 19:18:54 | FromDiscord | <UNIcodeX> (edit) 'http://ix.io/2wlF' => 'http://ix.io/2wlG' |
| 19:18:58 | FromDiscord | <Idefau> can't you do that with a `when` |
| 19:19:08 | FromDiscord | <haxscramper> There is hot reloading for compiled nim code. I only tested it, haven't done anythin serious though |
| 19:20:46 | FromDiscord | <haxscramper> https://nim-lang.org/docs/hcr.html and also https://github.com/genotrance/plugins |
| 19:29:22 | * | gmaggior joined #nim |
| 19:35:04 | FromDiscord | <UNIcodeX> Why are string conversions not thread safe? I'm sure there's a reason, but I'm at a loss as to what that would be. `.choosenim/toolchains/nim-1.2.6/lib/pure/asyncmacro.nim(278, 31) Error: 'matchIter' is not GC-safe as it calls '$'` |
| 19:35:35 | FromDiscord | <Rika> there are more details if you look up more in the error |
| 19:35:38 | FromDiscord | <Rika> that's just one line |
| 19:35:49 | FromDiscord | <Rika> to get the whole idea we need the whole error block |
| 19:37:06 | FromDiscord | <UNIcodeX> I'm still messing with the `bson` module. It's mentioned in the traceback. I'm guessing it's an overload and that's why it's saying that `$` is not threadsafe. |
| 19:37:31 | FromDiscord | <Rika> no |
| 19:37:37 | FromDiscord | <Rika> overloads are resolved in compile time |
| 19:37:52 | FromDiscord | <Rika> or wdym "its an overload" |
| 19:38:27 | FromDiscord | <UNIcodeX> a string method defined for the `Bson` type in module `bson` |
| 19:38:38 | FromDiscord | <UNIcodeX> sent a code paste, see https://play.nim-lang.org/#ix=2wlJ |
| 19:38:41 | FromDiscord | <UNIcodeX> wall - o - text |
| 19:39:29 | FromDiscord | <dom96> You may want to create a simple repro without Jester/Async. Should make the errors much easier to follow |
| 19:42:15 | FromDiscord | <Rika> `collectionToString` and `genExtendedJson` seem to be the issues? |
| 19:42:17 | FromDiscord | <UNIcodeX> > a string method defined for the `Bson` type in module `bson`↵@Rika is my understanding / terminology incorrect here? |
| 19:42:18 | FromDiscord | <Rika> or maybe just `genExtendedJson` |
| 19:43:54 | FromDiscord | <UNIcodeX> > `collectionToString` and `genExtendedJson` seem to be the issues?↵@Rika ok. noted. I'm gonna take a short break and mull it over. The wife wants to watch a movie. |
| 19:44:10 | FromDiscord | <UNIcodeX> Thank you all for your help. I'm sure I'll need more of it. 😅 |
| 19:44:20 | FromDiscord | <Rika> `genExtendedJson` is in bson.nim |
| 19:44:26 | FromDiscord | <Rika> seems to be the real issue |
| 19:45:13 | FromDiscord | <Rika> @UNIcodeX dw about answering me now, but why bson? why not cbor or some other binary representation? just asking really, no real reason for you to answer |
| 19:45:28 | * | tane quit (Quit: Leaving) |
| 19:48:43 | FromDiscord | <UNIcodeX> > @UNIcodeX dw about answering me now, but why bson? why not cbor or some other binary representation? just asking really, no real reason for you to answer↵@Rika It's what mongopool uses / recommends |
| 19:55:10 | FromDiscord | <Rika> I see, okay, thanks for answering |
| 19:55:16 | disruptek | sucks. |
| 19:56:42 | FromDiscord | <haxscramper> > I cannot simply do `import module` or `src/module` inside nimble as it won't work during installation.↵@haxscramperFigured this one, Just need to check `when fileExists(thisDir(), "src/foo_file.nim"): import src/foo_file else: import foo_file` |
| 19:57:22 | FromDiscord | <haxscramper> having source for all nimble packages turned out to be very useful |
| 19:57:41 | FromDiscord | <haxscramper> Found this one by `fd -e nimble | xargs grep "import src"` |
| 19:59:34 | disruptek | Araq: fwiw, sigv4 has been failing ci for the last couple days on devel. i will move it to github actions and maybe bisect it later. |
| 20:00:05 | disruptek | i thought it was in important packages but maybe not. i don't really pay attention to nim's ci setup; it's a moving target. |
| 20:11:46 | * | lritter quit (Ping timeout: 256 seconds) |
| 20:17:32 | * | lritter joined #nim |
| 20:42:57 | FromDiscord | <Rika> Does nim have an http/2 web server impl? |
| 20:43:20 | FromDiscord | <Rika> Actually wait I can prolly use a c lib |
| 21:04:32 | disruptek | eh, it's my fault. |
| 21:04:54 | disruptek | i can't believe it. i have to run a custom nimcrypto to make this shit work. |
| 21:04:56 | disruptek | madness. |
| 21:18:40 | * | solitudesf quit (Ping timeout: 265 seconds) |
| 21:25:34 | * | narimiran quit (Quit: leaving) |
| 21:32:16 | * | natrys quit (Quit: natrys) |
| 21:36:40 | * | Trustable quit (Remote host closed the connection) |
| 21:37:08 | FromDiscord | <iWonderAboutTuatara> What are some problems that are "good" for metaprogramming solutions? |
| 21:37:26 | FromDiscord | <iWonderAboutTuatara> Like can someone give me an example of something that would be super hard to do normally that macros make a lot easier? |
| 21:37:30 | superbia | 1) create your own language |
| 21:37:44 | disruptek | !repo jason |
| 21:37:44 | disbot | https://github.com/disruptek/jason -- 9jason: 11JSON done right 🤦 15 19⭐ 1🍴 |
| 21:38:22 | disruptek | !repo openapi |
| 21:38:22 | disbot | https://github.com/disruptek/openapi -- 9openapi: 11OpenAPI Code Generator for Nim 15 32⭐ 3🍴 7& 5 more... |
| 21:38:34 | disruptek | !repo testes |
| 21:38:37 | disbot | https://github.com/disruptek/testes -- 9testes: 11a small unittest framework with decent support 🔴🟡🟢 15 12⭐ 0🍴 7& 29 more... |
| 21:38:51 | disruptek | !repo cps |
| 21:38:52 | disbot | https://github.com/disruptek/cps -- 9cps: 11Continuation-Passing Style for Nim 🔗 15 32⭐ 1🍴 |
| 21:42:28 | * | krux02__ joined #nim |
| 21:43:09 | * | vicfred joined #nim |
| 21:44:49 | * | krux02_ quit (Ping timeout: 246 seconds) |
| 21:49:54 | * | awe001 joined #nim |
| 21:52:11 | * | awe00 quit (Ping timeout: 240 seconds) |
| 21:53:57 | haxscramper | How can I prevent nimble from deleting temporary nimscript file? It dies with `invalid indentation` error, on some line and I don't understand which line this corresponds to in my `.nimble` file |
| 21:54:31 | disruptek | too funny. if you only knew how much grief not deleting those files has caused others... |
| 21:56:26 | haxscramper | So there is no way to figure this out except for just carefully examining every possible place where indentation error can occur? |
| 21:57:10 | disruptek | set the sticky bit on tmp and/or cwd so it cannot be deleted? |
| 21:57:17 | disruptek | hack nimble? |
| 21:57:28 | disruptek | i dunno. i don't use nimble. |
| 21:57:28 | federico3 | people invented directories... |
| 21:58:18 | disruptek | church. |
| 22:00:16 | FromDiscord | <Clyybber> disruptek: please minimize your concept bugs, if you can |
| 22:00:26 | FromDiscord | <Clyybber> otherwise we can't help you |
| 22:06:27 | * | wiml joined #nim |
| 22:17:29 | haxscramper | There is no other way to check if package is installed other than `when compiles(import X)`? |
| 22:17:48 | haxscramper | Because it also fails if there is some error in package too |
| 22:17:58 | disruptek | yeah, it's pretty cool, right? |
| 22:18:13 | haxscramper | Yeah, makes your life interesting |
| 22:18:21 | disruptek | confucious say... |
| 22:22:17 | * | haxscramper quit (Remote host closed the connection) |
| 22:23:02 | wiml | Is there a list of the .experimental pragmata somewhere and what they do? I'm updating some old code ... |
| 22:23:15 | disruptek | ~concepts |
| 22:23:15 | disbot | concepts: 11user-defined type classes; documented in the experimental section of the manual: https://nim-lang.org/docs/manual_experimental.html#concepts -- disruptek |
| 22:23:22 | disruptek | check that page. |
| 22:25:34 | wiml | I found that but it seems incomplete? Maybe I need a list of old experimental pragmata and what they did? |
| 22:26:08 | disruptek | only if you have old pragmas in your code. |
| 22:26:53 | wiml | well like I said, I'm updating old code. |
| 22:27:08 | * | gmaggior quit (Ping timeout: 260 seconds) |
| 22:27:11 | disruptek | so, do you have an old pragma in your code? |
| 22:27:24 | wiml | ... |
| 22:28:26 | wiml | what do yuo think. yes, I do. that's why I'm looking for documentation. |
| 22:28:36 | disruptek | well, good luck, buddy. |
| 22:29:18 | wiml | so there's no documentation? |
| 22:30:05 | wiml | i mean, the current compiler still accepts this code, so it's not completely historical. it's just not documented, apparently. |
| 22:30:21 | FromDiscord | <Rika> wiml: what version was it |
| 22:30:31 | shashlick | @haxscramper are you using Nim devel |
| 22:30:43 | FromDiscord | <Rika> can you tell us what pragmas you're unfamiliar with |
| 22:32:24 | disruptek | wiml: your guess is as good as mine. maybe if you hold the code closer to your router, i'll be able to read it. |
| 22:34:13 | wiml | Rika: I'm not sure what version of nim was current when it was written. It's 'experimental: "notnil"'. But I guess I have to go muck through the compiler sources |
| 22:34:30 | wiml | I think there are others in this codebase as well, which is why I'm looking for documentation |
| 22:37:09 | FromDiscord | <Rika> wiml: i found it in the nim devel docs |
| 22:37:39 | FromDiscord | <Rika> `All types for which nil is a valid value can be annotated with the not nil annotation to exclude nil as a valid value:` |
| 22:38:23 | FromDiscord | <Rika> install nim devel and search its docs maybe |
| 22:38:31 | FromDiscord | <Rika> usually in manual_experimental |
| 22:39:02 | * | gmaggior joined #nim |
| 22:43:36 | wiml | Aha. Just cloned the nim-lang repo and found it in there. I guess it's just not in the version that's on nim-lang.org. |
| 22:43:57 | FromDiscord | <Rika> so its not an old pragma, its a new one |
| 22:44:01 | wiml | Grubbing around, it was originally documented in th echangelog for 0.19, but changelogs aren't part of the documentation any more |
| 22:44:09 | FromDiscord | <Rika> oh really? |
| 22:44:27 | FromDiscord | <Rika> so it was only recently documented |
| 22:44:49 | wiml | yeah: https://github.com/nim-lang/Nim/blob/devel/changelogs/changelog_0_19_0.md |
| 22:46:47 | wiml | well, now I know what it's doing in there. thanks for the pointers, they were helpful |
| 22:50:39 | * | SenasOzys joined #nim |
| 22:53:47 | * | SenasOzys quit (Client Quit) |
| 22:53:51 | * | lritter quit (Quit: Leaving) |
| 22:54:27 | * | SenasOzys joined #nim |
| 22:58:45 | * | krux02__ quit (Remote host closed the connection) |
| 23:09:21 | * | theelous3 quit (Ping timeout: 265 seconds) |
| 23:34:42 | * | abm joined #nim |
| 23:35:01 | * | gmaggior quit (Ping timeout: 264 seconds) |
| 23:47:20 | * | awe001 quit (Ping timeout: 258 seconds) |
| 23:48:27 | * | gmaggior joined #nim |