| 00:02:48 | ldlework | Hmm don't think I've seen this one before, dadren/meta.nim(113, 13) Error: attempting to call undeclared routine: 'type_name=' |
| 00:03:09 | ldlework | oh |
| 00:06:12 | dom96 | Xe: Please make an issue about install.sh if you haven't already. |
| 00:07:49 | Xe | dom96: i did |
| 00:08:16 | dom96 | thanks |
| 00:09:34 | ldlework | Error: interpretation requires too many iterations |
| 00:09:36 | ldlework | doh |
| 00:10:55 | Xe | compiler/plugins/active.nim(13, 18) Error: cannot open 'itersgen' |
| 00:12:59 | * | Xe fell back to a tarball install for now |
| 00:16:30 | Araq_ | uh oh |
| 00:17:00 | Trustable | Nim 0.13.0 - yay :) - thank you guys |
| 00:19:33 | * | brson quit (Read error: Connection reset by peer) |
| 00:20:35 | * | brson joined #nim |
| 00:20:45 | derka | Guys |
| 00:21:13 | derka | proc foo(x: var seq[int] = @[]) . You can't pass a literal to a 'var T' type. ? |
| 00:21:16 | derka | can someone ellaborate please |
| 00:21:36 | derka | why isnt it allowed |
| 00:22:54 | ldlework | derka: think about it |
| 00:22:57 | dom96 | Maybe this makes it clearer https://gist.github.com/dom96/a540df86ca4435e9e4b6 |
| 00:23:05 | ldlework | there you go |
| 00:23:43 | ldlework | Not sure why those semantics apply to default values |
| 00:23:48 | ldlework | Since you do have a way to access it |
| 00:23:49 | ldlework | `x` |
| 00:23:50 | derka | ok |
| 00:23:52 | * | Matthias247 quit (Read error: Connection reset by peer) |
| 00:23:53 | ldlework | inside the scope of the proc |
| 00:24:08 | * | ephja quit (Ping timeout: 272 seconds) |
| 00:24:18 | ldlework | that doesn't seem to make sense |
| 00:24:48 | * | pregressive quit (Remote host closed the connection) |
| 00:24:53 | ldlework | Do we have a quick nim-snippet web thing yet? |
| 00:25:01 | ldlework | to actually run it and see the output, etc? |
| 00:27:44 | derka | but this is allowed: |
| 00:27:45 | derka | proc foo(x: seq[int] = @[]) = |
| 00:27:46 | derka | echo $x |
| 00:27:47 | derka | foo() |
| 00:28:29 | ldlework | right |
| 00:30:00 | derka | i see the point now |
| 00:30:20 | derka | i was thinking like we do in python |
| 00:30:34 | derka | sometime we can use a default val list and accumulate stuff on it |
| 00:30:46 | ldlework | Does anyone have any interest in helping me draft an outline for a high-level AST library? |
| 00:30:59 | ldlework | One that speaks in terms of Nim-language constructs and enforces their semantics? |
| 00:31:18 | ldlework | I am a good ways along, but I don't really know the best boundries to set. |
| 00:31:27 | ldlework | Basically design help, not implementation help. |
| 00:31:50 | * | Trustable quit (Remote host closed the connection) |
| 00:36:16 | Araq_ | ldlework: I'm interested in the sense that I will add it to the core once it's ready. |
| 00:36:33 | Araq_ | because it's definitely a wise thing to do. |
| 00:39:49 | ldlework | Its hard |
| 00:39:56 | ldlework | Ident's are used for everything. lol |
| 00:40:32 | ldlework | Its hard to tell where, where an Ident is accepted, something else is accepted |
| 00:40:44 | ldlework | like a prefix, or a bracket, and so on |
| 00:41:11 | dom96 | ldlework: What do you envision it would look like? |
| 00:41:39 | ldlework | dom96: you'd have types like ASTProc and ASTType and ASTField and ASTSymbol and so on |
| 00:41:46 | ldlework | And they'd have fields like .exported and so on |
| 00:41:56 | ldlework | And they'd have fields for optional generic discriminators and so on |
| 00:42:00 | ldlework | IE, model the actual things in the language |
| 00:42:16 | ldlework | And then give them all render procs that produce the full nested raw AST. |
| 00:42:36 | ldlework | They'd have parametric initializers for natural use, and also initializers that take a single NimNode |
| 00:42:42 | ldlework | For parsing uses. |
| 00:42:50 | dom96 | That would be nice |
| 00:43:07 | dom96 | I think you should look at what's available in other languages too. |
| 00:43:35 | dom96 | There must be some nice libraries for working with ASTs |
| 00:44:01 | ldlework | Well I think that Nim's AST api is okay, it seems to represent what the actual AST is. |
| 00:44:19 | ldlework | The problem is if you want to generate valid Nim, its hard to get right. |
| 00:44:34 | ldlework | Unless you resort to parsing string-templates |
| 00:44:40 | ldlework | which seems... eh we can do better |
| 00:45:10 | ldlework | I think I just need to clarifiy, in my mental model, how the various things that Ident is used for, how they are distinct |
| 00:46:25 | ldlework | Somethings I've identified as being implemented with Idents: proc names, type names, field names, field types, descriminators |
| 00:46:58 | ldlework | proc return types |
| 00:47:04 | ldlework | They can all be exported, except for field types and return types |
| 00:47:49 | ldlework | proc names, type names, return types and field types can all optionally exist as Brackets or BracketExprs (I'm not sure which cases use which yet) |
| 00:48:29 | Araq_ | er that's quite wrong ;-) |
| 00:48:36 | ldlework | hah, no doubt! |
| 00:48:44 | Araq_ | maybe the grammar helps you. |
| 00:49:09 | ldlework | Araq_: where can I look at that |
| 00:49:22 | ldlework | I don't know if I know how to read any grammar formats |
| 00:49:44 | Araq_ | http://nim-lang.org/docs/manual.html#syntax-grammar |
| 00:50:27 | Araq_ | but yeah, doesnt help |
| 00:50:35 | Araq_ | we need an AST grammar :-) |
| 00:50:53 | Araq_ | that would be cool. great idea. |
| 00:51:25 | ldlework | great idea, let me get on that? |
| 00:51:30 | ldlework | or great idea, let me know when there is a PR? |
| 00:51:32 | ldlework | haha |
| 00:52:35 | ldlework | I don't think I'm even going to attempt to parse proc bodys... |
| 00:53:13 | Araq_ | ldlework: the AST is quirky but also in a good sense |
| 00:53:42 | Araq_ | you can very often get away with a basic distinction between atoms and compound trees |
| 00:53:46 | * | kniteli quit (Ping timeout: 240 seconds) |
| 00:54:06 | ldlework | Araq_: in my current model, types are specified as a recursive list of types yeah. |
| 00:54:22 | ldlework | And if they are len == 1, they get rendered as an ident, >1 they get rendered as a BracketExpr |
| 00:55:03 | ldlework | Basically I'm just writing various procs and types and seeing how they parse and trying to fit all the different results I see under some unified abstraction |
| 00:55:54 | ldlework | But I've rewritten meta.nim like 3 times now... this would be the 4th rewrite :D |
| 00:56:20 | ldlework | Another tricky thing |
| 00:56:30 | ldlework | Is sometimes where an nnkIdent goes, an nnkEmpty can go O_O |
| 00:57:08 | Araq_ | that's only possible for anon procs I think |
| 00:57:29 | Araq_ | and it makes sense, it's anonymous proc after all |
| 00:57:42 | ldlework | also for various places where the compiler will infer the type |
| 00:57:52 | ldlework | like in proc parameters |
| 00:58:21 | ldlework | Like if you pass a default, you don't need to pass a type |
| 00:58:27 | ldlework | But default can also be empty |
| 00:58:40 | ldlework | That might be the only place I had to support it |
| 00:58:53 | ldlework | oh my god I haven't even considered what the proctypes look like |
| 00:59:17 | ldlework | like a proc param that is a proctype |
| 01:00:46 | ldlework | Araq_: Should I start a google-doc maybe where the high-level types could be described |
| 01:00:55 | ldlework | then once it makes sense, I can implement it? |
| 01:01:39 | dom96 | Xe: hey, still around? Could you try the new and improved installation instructions for Nimble? (In readme) :) |
| 01:05:27 | Xe | dom96: checking |
| 01:05:44 | dom96 | thanks |
| 01:06:02 | ldlework | Is the proper name for the type of a variable, or the type of a parameter, or the return type of a proc a "type description" ? |
| 01:06:12 | ldlework | like |
| 01:06:14 | ldlework | x: int |
| 01:06:15 | Xe | dom96: that worked |
| 01:06:16 | ldlework | the `int` part |
| 01:06:18 | Araq_ | ldlework: ok but I'm sleeping already |
| 01:06:24 | dom96 | Xe: yay, crisis averted :D |
| 01:06:28 | ldlework | Araq_: no problem, I'll come up with a base |
| 01:06:44 | Xe | dom96: yeah lol |
| 01:07:09 | ldlework | Like could I use "type description" and "type definition" to distinguish between type specifiers and type definition blocks? |
| 01:07:23 | Araq_ | ldlework: yes |
| 01:07:32 | ldlework | ok |
| 01:07:56 | ldlework | Maybe a good name is "type descriptor" ? |
| 01:08:03 | ldlework | or just "descriptor" ? |
| 01:08:33 | ldlework | I like descriptor. |
| 01:08:52 | ldlework | Because a "data descriptor" is a structure containing information that describes data. |
| 01:08:56 | ldlework | That's what a type is, right? |
| 01:09:08 | ldlework | a type is a "struct descriptor"? |
| 01:09:43 | ldlework | I'm gonna go with "descriptor" because it sounds cool. |
| 01:15:12 | Araq_ | scriptor, inquisitor |
| 01:16:52 | * | pregressive joined #nim |
| 01:17:03 | dom96 | good night guys |
| 01:17:08 | ldlework | Araq_: haha |
| 01:24:17 | * | brson quit (Ping timeout: 255 seconds) |
| 01:27:12 | * | derka quit (Quit: derka) |
| 01:27:41 | * | bozaloshtsh joined #nim |
| 01:28:49 | bozaloshtsh | I updated to nim 0.13 after not using it since like 0.10, and now I'm getting a strange error: Error: internal error: environment misses: world |
| 01:29:15 | Xe | what file are you trying to compile? |
| 01:29:16 | bozaloshtsh | it raises during compile, on a proc declaration |
| 01:29:31 | bozaloshtsh | it's my own program, actually I'm going to make a pastebin hang on |
| 01:30:38 | bozaloshtsh | http://pastebin.com/qbtnsMNX |
| 01:35:26 | Araq_ | bozaloshtsh: sorry, this was never officially supported, I think |
| 01:35:58 | Araq_ | are you sure that worked with 0.10 ? |
| 01:36:27 | bozaloshtsh | yep, it worked... but yeah I see the issue now. Can't access other parameter in default value. |
| 01:37:53 | Araq_ | error message could be better of course. and yeah eventually we will support default parameters refering to other parameters cause they are quite useful |
| 01:38:06 | Araq_ | good night. |
| 01:39:14 | ldlework | night Araq_ |
| 01:41:47 | * | pregressive quit () |
| 01:42:39 | * | pregressive joined #nim |
| 01:47:58 | ldlework | ARGH TUPLE TYPES |
| 01:48:03 | ldlework | shit on a stick |
| 01:51:30 | * | vendethiel joined #nim |
| 01:59:31 | * | strcmp1 joined #nim |
| 01:59:51 | ldlework | maybe just ignore tuple types for now |
| 02:00:11 | * | strcmp1 quit (Client Quit) |
| 02:00:13 | ldlework | ugh this is a hard task |
| 02:00:26 | * | strcmp1 joined #nim |
| 02:14:44 | * | vendethiel quit (Ping timeout: 245 seconds) |
| 02:18:29 | ldlework | If I have an object type with say a string field, is it okay to sometimes initialize that field to nil? |
| 02:19:17 | * | vendethiel joined #nim |
| 02:27:49 | ldlework | object variants are crazy |
| 02:29:55 | * | lankanmon joined #nim |
| 02:41:11 | ldlework | Can you use multiple variant fields? |
| 02:41:20 | ldlework | like multiple variant discrirminators? |
| 02:42:31 | * | vendethiel quit (Ping timeout: 260 seconds) |
| 03:23:28 | ldlework | dom96: def- how is this looking? https://gist.github.com/dustinlacewell/957646128a24cd6c8141 |
| 03:23:42 | ldlework | this is purely for just type definitions, no proc support |
| 03:24:39 | * | maikklein_ joined #nim |
| 03:25:05 | maikklein_ | Does nim have ownership semantics? |
| 03:25:19 | maikklein_ | like in c++ with move? |
| 03:32:03 | * | maikklein_ quit (Ping timeout: 252 seconds) |
| 03:58:32 | * | pregressive quit (Remote host closed the connection) |
| 04:14:08 | * | brson joined #nim |
| 04:27:18 | vegansk | Arrrr, sorry, no time until february :-( |
| 04:55:01 | * | strcmp1 quit (Quit: peace) |
| 05:09:06 | mountaingoat | sorry if this is a nonsensical question, but is Seq[T] a reference type? |
| 05:09:17 | mountaingoat | how can i pass one as a reference to a function? |
| 05:16:56 | Xe | stupid question of the day: what would be the best way to package nimble as a distribution package? |
| 05:31:47 | * | s4 joined #nim |
| 05:33:19 | * | kniteli joined #nim |
| 05:54:37 | * | xtagon joined #nim |
| 06:13:04 | Xe | reactormonk: it was intended for it to print 4 every time? |
| 06:14:31 | reactormonk | Xe, nope, the other way. But apparently now we go all JS |
| 06:14:48 | Xe | we javascript now? |
| 06:15:11 | * | brson quit (Ping timeout: 264 seconds) |
| 06:25:11 | mountaingoat | is there a way to clear/fill/memset a Seq[T]? |
| 06:25:23 | mountaingoat | i mean a standard library proc for it |
| 06:25:51 | mountaingoat | reset[T]()? |
| 06:59:42 | * | derka joined #nim |
| 07:02:07 | * | Demon_Fox quit (Ping timeout: 276 seconds) |
| 07:04:37 | * | derka quit (Quit: derka) |
| 07:05:40 | * | kniteli quit (Ping timeout: 272 seconds) |
| 07:07:43 | * | darkf joined #nim |
| 07:14:03 | * | vendethiel joined #nim |
| 07:16:26 | * | bjz joined #nim |
| 07:21:38 | * | bjz quit (Ping timeout: 276 seconds) |
| 07:37:59 | * | vendethiel quit (Ping timeout: 264 seconds) |
| 07:41:40 | * | bjz joined #nim |
| 07:42:11 | * | satbyy joined #nim |
| 07:42:46 | * | Matthias247 joined #nim |
| 07:45:41 | * | sjums joined #nim |
| 07:48:13 | * | nsf quit (Quit: WeeChat 1.3) |
| 07:49:57 | * | nsf joined #nim |
| 07:52:52 | * | Matthias247 quit (Read error: Connection reset by peer) |
| 07:53:22 | wuehlmaus | dom96: hmm, "our" bug is consistent in 0.13 |
| 08:04:46 | * | gokr joined #nim |
| 08:08:55 | Araq_ | mountaingoat: use x.setLen(0) ? and read some tutorial please ;-) |
| 08:10:42 | * | kniteli joined #nim |
| 08:11:45 | * | vendethiel joined #nim |
| 08:14:03 | * | yglukhov joined #nim |
| 08:15:59 | * | bjz_ joined #nim |
| 08:16:06 | * | bjz quit (Ping timeout: 240 seconds) |
| 08:28:50 | * | Trustable joined #nim |
| 08:33:18 | * | xtagon quit (Quit: Connection closed for inactivity) |
| 08:34:26 | * | vendethiel quit (Ping timeout: 256 seconds) |
| 08:39:10 | mountaingoat | Araq_: i wasn't looking to truncate it, but to zero it |
| 08:40:14 | mountaingoat | right now i'm just iterating over it and manually zeroing each element, but i was wondering if there was an easier way |
| 08:41:05 | Araq_ | well you can use zeroMem() but that's unsafe and will screw up RCs |
| 08:41:16 | Araq_ | so better do it manually |
| 08:41:41 | mountaingoat | okay |
| 08:42:02 | mountaingoat | thanks |
| 08:50:59 | * | desophos quit (Read error: Connection reset by peer) |
| 08:57:15 | yglukhov | mountaingoat: mySeq.applyIt 0 |
| 09:03:29 | * | kniteli quit (Ping timeout: 245 seconds) |
| 09:09:49 | Araq_ | hmm maybe I should start using sequtils |
| 09:17:29 | * | kniteli joined #nim |
| 09:21:24 | * | ephja joined #nim |
| 09:22:32 | * | vqrs quit (Ping timeout: 276 seconds) |
| 09:24:00 | * | vqrs joined #nim |
| 09:26:23 | * | Arrrr joined #nim |
| 09:40:28 | * | vendethiel joined #nim |
| 09:55:33 | * | derka joined #nim |
| 09:55:34 | * | derka quit (Client Quit) |
| 10:01:24 | * | vendethiel quit (Ping timeout: 245 seconds) |
| 10:06:53 | gokr | Araq_: Great work man! :) |
| 10:07:08 | gokr | (just skimmed release notes) |
| 10:07:16 | Araq_ | well apparently the installer doesn't work properly ... -.- |
| 10:08:46 | wuehlmaus | hmmm, when i do `` let t = (0,"hello",'a') and then try echo t[^1] it throws an error |
| 10:10:18 | wuehlmaus | this is with nim 0.13 |
| 10:13:02 | allan0_ | could someone make a github "release" for the updated csources repo? would make packaging easier. |
| 10:13:57 | Araq_ | allan0_: how do I do that? just tag it? |
| 10:15:22 | allan0_ | i think so. but there is also some webgui stuff when you click on "release" just above the clone url. |
| 10:15:40 | Araq_ | nice let me try |
| 10:22:51 | * | derka joined #nim |
| 10:28:01 | * | vangroan joined #nim |
| 10:29:14 | * | vqrs quit (Ping timeout: 255 seconds) |
| 10:29:55 | * | derka_ joined #nim |
| 10:31:40 | * | vqrs joined #nim |
| 10:32:27 | * | derka quit (Ping timeout: 260 seconds) |
| 10:32:27 | * | derka_ is now known as derka |
| 10:33:16 | * | strcmp1 joined #nim |
| 10:38:17 | dom96 | allan0_: Should be tagged now :) |
| 10:44:01 | allan0_ | thanks |
| 10:44:59 | * | kniteli quit (Ping timeout: 255 seconds) |
| 11:01:21 | * | strcmp1 quit (Quit: peace) |
| 11:11:15 | * | exebook quit (Remote host closed the connection) |
| 11:15:48 | * | derka quit (Quit: derka) |
| 11:18:30 | * | vendethiel joined #nim |
| 11:20:46 | * | derka joined #nim |
| 11:42:28 | * | vendethiel quit (Ping timeout: 265 seconds) |
| 11:45:22 | * | vendethiel joined #nim |
| 11:45:46 | * | vangroan_ joined #nim |
| 11:46:25 | * | vangroan quit (Read error: Connection reset by peer) |
| 11:48:46 | * | vangroan__ joined #nim |
| 11:50:41 | * | vqrs quit (Ping timeout: 255 seconds) |
| 11:52:12 | * | vangroan_ quit (Ping timeout: 256 seconds) |
| 11:54:11 | * | vqrs joined #nim |
| 12:06:47 | * | vendethiel quit (Ping timeout: 264 seconds) |
| 12:11:32 | * | derka quit (Ping timeout: 276 seconds) |
| 12:12:02 | * | derka joined #nim |
| 12:30:22 | * | vendethiel joined #nim |
| 12:33:13 | * | kniteli joined #nim |
| 12:41:47 | * | derka quit (Quit: derka) |
| 12:44:18 | * | derka joined #nim |
| 12:51:35 | * | vendethiel quit (Ping timeout: 265 seconds) |
| 13:00:47 | * | askatasuna quit (Ping timeout: 264 seconds) |
| 13:01:37 | * | askatasuna joined #nim |
| 13:07:13 | yglukhov | Has anyone seen a template that locks closure scope to be used in loops? I'm pretty sure I've seen a snippet somewhere on the forum or github, but cant exactly remember... |
| 13:08:26 | * | BitPuffin joined #nim |
| 13:11:26 | * | nchambers joined #nim |
| 13:15:22 | * | vendethiel joined #nim |
| 13:18:36 | * | ekarlso quit (Changing host) |
| 13:18:36 | * | ekarlso joined #nim |
| 13:21:44 | * | askatasuna quit (Ping timeout: 256 seconds) |
| 13:24:28 | yglukhov | ok, wrote my own. in case anyone needs it: template closureScope(body: untyped): stmt = (proc() = body)() |
| 13:34:30 | * | s4 quit (Quit: Konversation terminated!) |
| 13:36:20 | * | vendethiel quit (Ping timeout: 260 seconds) |
| 13:39:31 | * | vendethiel joined #nim |
| 13:53:45 | * | vangroan__ is now known as vangroan |
| 14:00:06 | * | vendethiel quit (Ping timeout: 240 seconds) |
| 14:07:32 | * | vendethiel joined #nim |
| 14:20:06 | * | vqrs quit (Ping timeout: 256 seconds) |
| 14:20:52 | * | vqrs joined #nim |
| 14:21:53 | * | arnetheduck joined #nim |
| 14:23:57 | * | arnetheduck left #nim (#nim) |
| 14:24:06 | * | arnetheduck_ joined #nim |
| 14:24:16 | * | arnetheduck_ quit (Client Quit) |
| 14:24:43 | * | arnetheduck joined #nim |
| 14:30:18 | * | vendethiel quit (Ping timeout: 256 seconds) |
| 14:39:32 | * | NickMane joined #nim |
| 14:40:33 | * | askatasuna joined #nim |
| 14:40:45 | * | saml joined #nim |
| 14:41:26 | NickMane | can someone tell me how xmlCheckedTag(http://nim-lang.org/docs/htmlgen.html#xmlCheckedTag,NimNode,string,string,string) is supposed to be called? Specifically, what should I pass as the first argument? |
| 14:43:56 | arnetheduck | a little toy I came up with: http://forum.nim-lang.org/t/1955 |
| 14:44:17 | arnetheduck | comments welcome ;) |
| 14:45:17 | dom96 | hello NickMane, take a look at the source of the other html elements, for example <a> https://github.com/nim-lang/Nim/blob/master/lib/pure/htmlgen.nim#L98 |
| 14:45:23 | dom96 | They show how to use xmlCheckedTag() |
| 14:46:59 | NickMane | Oh... right. I actually tried that, but that was before importing 'macros'. Thanks! |
| 14:47:41 | dom96 | arnetheduck: nice! Hope this results in a practical alternative Nim backend :) |
| 14:49:34 | dom96 | arnetheduck: seems like you've actually developed quite a lot of it already, how much of Nim does it support? |
| 14:49:52 | derka | yglukhov what is the difference between what you did and using block |
| 14:50:52 | arnetheduck | well, I did get hello world to link and compile at some point, but I think I broke it again later |
| 14:52:12 | arnetheduck | it reuses the usual nim compiler for all the heavy lifting, so all it has to do is spit out IR.. all the parsing, lambda stuff etc is already taken care of.. |
| 14:52:13 | arnetheduck | it |
| 14:52:27 | arnetheduck | 's very similar to the c & js backends in that sense |
| 14:54:46 | yglukhov | derka: the difference is noticable when youre creating a closure in a loop. with my template loop locals will be "copied" to env of every created closure. |
| 14:55:32 | yglukhov | otherwise all closures will refer the same locals |
| 14:55:38 | derka | ok interesting |
| 14:56:44 | derka | sorry for asking but in which use case do you need such a behavior ? |
| 14:57:20 | wuehlmaus | hmm, nobody has replied to my using of var[^1] on a tuple. was that a stupid way? |
| 14:59:06 | yglukhov | derka: almost always when creating closure in a loop =) |
| 14:59:13 | derka | :) |
| 15:00:40 | yglukhov | for reference this is the same way as javascript works. so you can read up more on that in regards to js. e.g. http://stackoverflow.com/questions/750486/javascript-closure-inside-loops-simple-practical-example |
| 15:02:29 | * | vangroan quit (Ping timeout: 276 seconds) |
| 15:03:14 | derka | i see your point now |
| 15:03:22 | derka | thx member:yglukhov |
| 15:06:02 | NickMane | is it possible in nim to use procs as infix operators? |
| 15:06:18 | Xe | NickMane: like backticking in haskell? |
| 15:06:26 | * | bozaloshtsh quit (Ping timeout: 265 seconds) |
| 15:06:35 | NickMane | Xe:yeah |
| 15:06:41 | Xe | NickMane: unsure |
| 15:06:51 | yglukhov | NickMane: afaik only operators can be used as infix operators =) |
| 15:06:51 | * | Senketsu joined #nim |
| 15:08:20 | NickMane | yglukhov: aha, and does proc `somesymbol` count as a new operator, or only the inbuilt operators? |
| 15:08:31 | * | Jesin quit (Remote host closed the connection) |
| 15:09:03 | yglukhov | NickMane: you can define custom opertors, but dont use letters in its name. |
| 15:09:07 | yglukhov | proc `%!@`(a, b: int): int = a + b |
| 15:09:11 | yglukhov | let f = 5 %!@ 7 |
| 15:10:04 | yglukhov | operators are not only procs, but templates and macros as well |
| 15:10:53 | NickMane | yglukhov: ah, thanks! :) I tried that, but didn't omit the `` when using the operator |
| 15:11:10 | NickMane | great that you cna us |
| 15:11:22 | NickMane | great that you can use the operator without ``* |
| 15:11:32 | yglukhov | `` are used for function call syntax: let f = `%!@`(5, 7) |
| 15:11:55 | NickMane | that makes sense |
| 15:12:59 | dom96 | wuehlmaus: if the compiler crashes then please report it as a bug |
| 15:13:33 | dom96 | wuehlmaus: As for whether that should work: I think it should. |
| 15:13:45 | wuehlmaus | it does not crash, it just doesn't output the last element |
| 15:13:54 | wuehlmaus | and gives a confusing error message :) |
| 15:14:04 | Araq_ | I disagree, it does not have to work |
| 15:14:16 | wuehlmaus | var t = (0,"one",'a'); echo t[^1] doesn't work |
| 15:14:36 | Araq_ | ^1 is rewritten to x.len-1 |
| 15:14:46 | Araq_ | and tuples have no length |
| 15:14:50 | dom96 | Araq_: makes sense |
| 15:15:12 | yglukhov | tuples dont have len? |
| 15:15:14 | wuehlmaus | that's what i wanted to know, so i was stupid, learned something :) |
| 15:17:20 | yglukhov | im pretty sure its possible to write a library template/macro len for tuples |
| 15:17:26 | yglukhov | then ^1 should work |
| 15:18:35 | yglukhov | even c++ can has len for tuples! :D |
| 15:19:16 | * | NickMane quit (Quit: Page closed) |
| 15:19:31 | wuehlmaus | okay, then i was not stupid to expect it ;) |
| 15:21:09 | dom96 | arnetheduck: awesome, hope you stick with it! |
| 15:21:42 | Araq_ | arnetheduck: cool! at one point we could adopt it as the standard Nim backend. |
| 15:22:25 | Araq_ | yglukhov: yeah it's possible but I'm not sure it's wise |
| 15:22:48 | Araq_ | the compiler used to rewrite len(3) to len([3]) for example |
| 15:22:57 | Araq_ | so len(3) produced 1. |
| 15:23:09 | Araq_ | produced nasty bugs in my code. |
| 15:23:32 | Araq_ | as a result Nim distinguishes between 'openarray' and 'varargs' ;-) |
| 15:28:58 | * | gokr quit (Quit: Leaving.) |
| 15:30:22 | * | coffeepot joined #nim |
| 15:30:31 | coffeepot | Hey guys! |
| 15:30:39 | coffeepot | congratulations again on the release :) |
| 15:31:23 | dom96 | hello coffeepot, thanks :) |
| 15:31:58 | dom96 | ooh, we're on HN now too https://news.ycombinator.com/item?id=10929426 |
| 15:32:14 | coffeepot | I got a question regarding templates and generics. Can I make this work? It sigsegv's the compiler. https://gist.github.com/coffeepots/53b17b976f6561c27fbf |
| 15:32:18 | coffeepot | ooo cool! |
| 15:32:20 | * | bozaloshtsh joined #nim |
| 15:32:47 | coffeepot | I tried to make a HN account once and the captcha would not produce anything I could read, so I guess that proves I'm a robot |
| 15:33:28 | Xe | dom96: where is an example of the nimscript method for nimble package definition? |
| 15:34:04 | dom96 | Xe: https://github.com/nim-lang/nimble#the-new-nimscript-format |
| 15:34:11 | dom96 | Xe: or you can just run `nimble init` |
| 15:34:12 | Araq_ | coffeepot: get rid of the .immediate |
| 15:34:19 | Araq_ | and replace 'stmt' with 'untyped' |
| 15:34:55 | Xe | dom96: i figured it out, the `name` statement is now gone lol |
| 15:35:01 | coffeepot | rrrooger wilco Araq_ :D |
| 15:36:09 | dom96 | Xe: Yeah, it's called "packageName" now... for some reason. Perhaps I should rename it to 'name'. |
| 15:36:16 | Xe | dom96: or alias it |
| 15:36:18 | dom96 | It is optional now though |
| 15:36:21 | coffeepot | damn, I get Error: internal error: expr(skType); unknown symbol at line 13 |
| 15:36:24 | Xe | ah |
| 15:36:33 | dom96 | since the name of the file contains the package name |
| 15:38:18 | dom96 | coffeepot: They must have noticed that you're a coffee pot heh. |
| 15:39:19 | coffeepot | dom96: I think I'm a relative of that toaster on red dwarf |
| 15:40:18 | dom96 | coffeepot: haven't seen a single episode of Red Dwarf sadly |
| 15:40:31 | coffeepot | :-o |
| 15:42:47 | coffeepot | Araq_ any ideas how to get round this expr(skType) thingie? |
| 15:43:24 | yglukhov | Araq: hypothetically. how much time would you need to make closure iters work in js? ;) |
| 15:43:49 | derka | ok |
| 15:43:54 | derka | sorry wrong window :) |
| 15:44:56 | * | pregressive joined #nim |
| 15:45:46 | Araq_ | coffeepot: no, sorry. you can always remove the parametrization of the template though |
| 15:46:39 | Araq_ | yglukhov: I told you. you need an alternative implementation for every control flow construct (if/case/while/try) |
| 15:46:56 | Araq_ | or maybe wait until JS supports 'yield' natively? |
| 15:50:41 | coffeepot | hmm, I tried to remove the parametrisation, but as far as I can see I need the generic types for the table |
| 15:51:34 | Araq_ | usually you can access the type with the 'type()' operator |
| 15:51:54 | coffeepot | okay! I'll give it a go :) |
| 15:55:48 | coffeepot | haha I was being a dingleberry |
| 15:55:58 | coffeepot | there's a rather obvious mistake in my code |
| 15:56:17 | coffeepot | I was passing the A type into the table.key proc... d'oh! |
| 15:56:18 | Xe | dom96: can you access nimble metadata for a given package from the nimscript integration? |
| 15:56:49 | dom96 | Xe: not right now |
| 15:57:03 | coffeepot | check it out, works :D https://gist.github.com/coffeepots/4fef19f7e36f400b64c4 |
| 15:57:04 | Xe | ah |
| 15:57:07 | dom96 | Xe: What would you like to do with it? |
| 15:57:16 | Xe | i had an idea to make something like |
| 15:57:23 | Xe | `nimble packageListGen` |
| 15:57:31 | Xe | that would generate the corect json |
| 15:57:57 | Xe | but it would mean adding tags to nimble definitions |
| 15:59:01 | dom96 | `nimble publish` kinda does that already |
| 15:59:18 | coffeepot | ... and with that code, the reason for *not* having a 'tryGet' in tables is weakened - the reason to not have one was that it offered no simplistic advantage to doing getDetault and comparing it to the default value. But now you can do it as a one liner with the template creating the return value var! Mwuahhahahaa! |
| 15:59:35 | * | vendethiel joined #nim |
| 16:00:26 | coffeepot | of course, a "real" version of tryGet would be implemented at the lower level of tables to avoid the hasKey call, meaning only one hash required to simultaneously checking a value exists and retrieving it :) |
| 16:01:01 | Xe | also `nimble publish` doesn't work if you have 2FA :P |
| 16:01:30 | derka | What is the max size of a string? |
| 16:01:33 | * | Jesin joined #nim |
| 16:01:44 | Xe | derka: how much ram do you have? |
| 16:02:12 | coffeepot | derka that probably depends on if you're running 32 or 64 bit |
| 16:02:19 | derka | ok |
| 16:02:27 | derka | so the size is stored in 32 or 64 bits |
| 16:02:33 | derka | depending on arch |
| 16:02:42 | coffeepot | don't trust me :) |
| 16:02:45 | derka | :) |
| 16:03:36 | derka | Araq_ or anyone able to answer, how can we have Nim use jemalloc instead of linux malloc |
| 16:09:45 | Araq_ | Xe: dom96 huh? I implemented 'nimble dump' for this |
| 16:10:13 | Araq_ | or maybe 'dumb' ... for some reason I never know if it's with a 'p' or a 'b' |
| 16:10:31 | Xe | ah |
| 16:10:33 | Xe | okay i see |
| 16:10:36 | Xe | thanks |
| 16:11:00 | Araq_ | derka: since it doesn't use malloc to begin with, it cannot use jemalloc either |
| 16:11:16 | Xe | also sigh, it doesn't seem like nim parsecfg correctly reads systemd unit files |
| 16:11:48 | derka | oh ok, so Araq_ how is memory managed |
| 16:11:49 | Araq_ | the GC uses its own memory manager. |
| 16:11:51 | derka | is there a doc about it |
| 16:12:18 | * | kniteli quit (Ping timeout: 256 seconds) |
| 16:12:26 | Araq_ | no, but you can read lib/system/alloc.nim |
| 16:13:19 | Xe | does nimscript work on machines that don't have the nim compiler installed? |
| 16:13:51 | Araq_ | it needs the core of the stdlib (system.nim) |
| 16:14:03 | Araq_ | it cannot work without |
| 16:14:12 | Xe | ah, hmm |
| 16:16:57 | * | irrequietus joined #nim |
| 16:19:50 | * | bozaloshtsh quit (Ping timeout: 272 seconds) |
| 16:21:24 | * | vendethiel quit (Ping timeout: 250 seconds) |
| 16:24:14 | dom96 | Xe: oh really? I have 2FA and it works :P |
| 16:24:35 | Xe | dom96: it didn't work for me! |
| 16:25:50 | dom96 | Xe: did you get an error or what? |
| 16:26:04 | Xe | idk, haven't tested again |
| 16:26:19 | Xe | oh |
| 16:26:21 | Xe | it was a while ago |
| 16:26:24 | Xe | migth have been fixed |
| 16:26:27 | Xe | ignore me |
| 16:29:44 | * | vendethiel joined #nim |
| 16:30:51 | dom96 | Lobsters likes us too https://lobste.rs/s/cbxwkx/nim_v0_13_0_released |
| 16:30:52 | * | Salewski joined #nim |
| 16:34:04 | Salewski | In C we have: pango_layout_set_text (layout, "\u2654", -1); # or in bash echo -e "\U2654" |
| 16:34:34 | Salewski | How can we use that unicode char (chess king) in a Nim string? |
| 16:37:44 | * | satbyy_ joined #nim |
| 16:40:15 | Araq_ | Salewski: write the chess king as utf-8 with your editor |
| 16:40:54 | Araq_ | or you have to use the multibyte encoding: \xff\xff |
| 16:42:46 | Xe | I am liking how the osproc module works |
| 16:42:50 | Salewski | So above is "\x26\54" in a Nim string? |
| 16:43:07 | Araq_ | that's not how UTF-8 works |
| 16:43:35 | * | darkf quit (Ping timeout: 260 seconds) |
| 16:44:35 | * | askatasuna quit (Ping timeout: 264 seconds) |
| 16:47:08 | * | yglukhov quit (Ping timeout: 276 seconds) |
| 16:51:18 | * | vendethiel quit (Ping timeout: 250 seconds) |
| 16:52:26 | Xe | Araq_: is there a way to embed the system.nim file into something that uses nimscript? |
| 16:54:34 | * | darkf joined #nim |
| 16:58:22 | * | ikay joined #nim |
| 17:04:01 | Araq_ | currently that's really hard |
| 17:04:19 | Araq_ | eventually you can slurp it into your exe. |
| 17:04:24 | Xe | ah |
| 17:04:47 | Xe | can you change where it looks for system.nim? |
| 17:04:55 | Araq_ | sure |
| 17:06:00 | * | vendethiel joined #nim |
| 17:12:32 | * | askatasuna joined #nim |
| 17:14:41 | * | strcmp1 joined #nim |
| 17:18:38 | * | brson joined #nim |
| 17:19:16 | * | coffeepot quit (Quit: http://www.kiwiirc.com/ - A hand crafted IRC client) |
| 17:21:05 | Salewski | Araq: SHIFT CONTROL U 2654 works in gedit and gvim for input -- and hexdump gave me the corresponding four byte string "\xe2\x99\x94\x0a". Both works fine in Nim. |
| 17:21:37 | * | ikay quit (Quit: Page closed) |
| 17:22:31 | * | Salewski quit () |
| 17:24:35 | * | jsudlow quit (Ping timeout: 255 seconds) |
| 17:25:20 | * | jsudlow joined #nim |
| 17:29:22 | * | vendethiel quit (Ping timeout: 276 seconds) |
| 17:34:21 | * | yglukhov joined #nim |
| 17:39:00 | * | yglukhov quit (Ping timeout: 272 seconds) |
| 17:47:49 | * | Senketsu quit (Quit: Leaving) |
| 17:50:33 | * | yglukhov joined #nim |
| 17:51:16 | * | yglukhov quit (Remote host closed the connection) |
| 17:54:19 | * | kniteli joined #nim |
| 17:54:58 | * | vendethiel joined #nim |
| 17:56:34 | * | jsudlow quit (Read error: Connection reset by peer) |
| 17:57:16 | * | jsudlow joined #nim |
| 17:57:59 | * | brson_ joined #nim |
| 17:59:17 | * | brson quit (Ping timeout: 276 seconds) |
| 17:59:37 | * | Salewski joined #nim |
| 18:00:06 | * | satbyy_ quit (Ping timeout: 240 seconds) |
| 18:03:41 | Salewski | We have array constants like: const H = ["a", "b"]; echo H[1]. Is the start index always 0 for this case, or can we have an arbitrary index range? (Of course I can assign the constant to a var and then have an arbitrary index.) |
| 18:05:06 | * | brson_ quit (Quit: leaving) |
| 18:05:13 | * | brson joined #nim |
| 18:11:47 | * | gokr joined #nim |
| 18:15:05 | Arrrr | yep, you can do va const H: array[char, char] |
| 18:15:08 | * | stefantalpalaru joined #nim |
| 18:15:57 | stefantalpalaru | Hi. Can somebody please regenerate the makefile and help me fix this issue? https://github.com/nim-lang/csources/issues/19 |
| 18:17:38 | stefantalpalaru | In case you're wondering, I can parallelize the compilation using the makefile instead of build.sh and the speed gain is worth it. |
| 18:17:41 | * | vendethiel quit (Ping timeout: 255 seconds) |
| 18:21:37 | * | vendethiel joined #nim |
| 18:24:58 | * | pregressive quit (Read error: Connection reset by peer) |
| 18:25:22 | * | pregressive joined #nim |
| 18:25:36 | * | Jesin quit (Quit: Leaving) |
| 18:30:26 | Araq_ | stefantalpalaru: I don't know how the makefile was generated. ping def- ? |
| 18:31:13 | stefantalpalaru | the comment at the top says "To regenerate run ``niminst csource`` or ``koch csource``" |
| 18:31:25 | federico3 | dom96 or others: does anyone wants to elaborate on https://news.ycombinator.com/item?id=10931611 ? |
| 18:31:42 | * | askatasuna quit (Ping timeout: 256 seconds) |
| 18:32:45 | * | buMPnet quit (Remote host closed the connection) |
| 18:33:06 | * | buMPnet joined #nim |
| 18:33:58 | stefantalpalaru | federico3: I don't think Nim offers the kind of concurrency and parallelism guarantees that Pony does. Maybe the commenter confuses the 2 languages. |
| 18:34:22 | federico3 | ...pony? |
| 18:35:00 | stefantalpalaru | http://tutorial.ponylang.org/ |
| 18:35:14 | stefantalpalaru | they were in the news recently for a release |
| 18:35:50 | federico3 | *_* |
| 18:35:59 | stefantalpalaru | "the type system ensures at compile time that your concurrent program can never have data races" - I don't think even Haskell does that |
| 18:36:16 | Xe | IIRC nim has that |
| 18:38:40 | * | Senketsu joined #nim |
| 18:39:46 | ldlework | Yeah it does |
| 18:39:55 | ldlework | I've never pushed it to the limit and tried to break it though |
| 18:40:16 | federico3 | people really don't get the case insensitivity |
| 18:42:18 | stefantalpalaru | https://en.wikipedia.org/wiki/Race_condition#Software - anybody has an example of 2 threads accessing the same memory location in parallel, without locks, and the compiler catching it? |
| 18:42:40 | ldlework | stefantalpalaru: there is an example on the front page!! |
| 18:42:45 | ldlework | the /front page!/ |
| 18:42:48 | ldlework | :) |
| 18:42:50 | Araq_ | ldlework: not anymore ;-) |
| 18:42:53 | ldlework | oop |
| 18:42:56 | ldlework | lol |
| 18:43:05 | Araq_ | it's not that it didn't work. |
| 18:43:37 | Araq_ | it's just that the proof engine was really stupid and would reject valid code |
| 18:43:49 | Araq_ | for no obvious reason. |
| 18:44:03 | ldlework | Ah |
| 18:44:08 | ldlework | Did we deprecate the feature for now? |
| 18:44:18 | Araq_ | we moved it to .experimental |
| 18:44:44 | Araq_ | it's pretty cool nevertheless, it's just a lot of work left |
| 18:45:01 | ldlework | neat |
| 18:45:05 | * | Matthias247 joined #nim |
| 18:45:28 | ldlework | Does it use complicated math and stuff, or could someone like me just add in tons of heuristics and stuff to get it to work better and better over time? |
| 18:45:56 | Araq_ | it uses complicated math and adhoc control flow algorithms |
| 18:46:02 | Araq_ | sorry. |
| 18:46:41 | * | Trustable_2 joined #nim |
| 18:46:49 | * | Trustable quit (Read error: Connection reset by peer) |
| 18:47:11 | ldlework | hehe no problem just curious :) |
| 18:47:13 | * | desophos joined #nim |
| 18:47:41 | Araq_ | but in general these "safety over everything" programming languages (like Pony and Rust) have problems with expressiveness. |
| 18:47:53 | stefantalpalaru | true |
| 18:48:01 | Araq_ | creating a safe language is not hard, it's in fact trivial. |
| 18:48:26 | Araq_ | "index out of bounds? no problem, we make indexing use modulo array-length" |
| 18:48:35 | ldlework | lol |
| 18:48:52 | Xe | that's one way lol |
| 18:48:55 | ldlework | That pcwalton guy does a lot to hurt Nim's image in a time when memory-safety is king/popularized though. |
| 18:49:11 | Xe | oh, where's the string type defined in the nim compiler? |
| 18:49:21 | ldlework | Maybe hold off 1.0 for a couple of years until the fad wears off. |
| 18:49:24 | ldlework | :P |
| 18:49:58 | ldlework | Everyone will eventually realize that noone's life depends on their shitty web software written in Rust and flock away |
| 18:50:39 | Araq_ | lol, I like Rust but I don't see the need to shit on under-staffed competitors |
| 18:50:43 | ldlework | (I don't mean to say that software written in Rust is shitty, just that the mainstream is looking at rust right now, so there is bound to be tons of really poorly written software in Rust in the next few years) |
| 18:50:53 | ldlework | I'm not shitting on Rust at all |
| 18:51:05 | ldlework | Rust is glorious progression. |
| 18:51:28 | Araq_ | well with "under-staffed competitors" I mean Nim... |
| 18:51:31 | ldlework | Oh you were remarking .. yeah |
| 18:51:41 | Xe | i have been considering trying to do haskell FFI from nim before |
| 18:51:59 | Xe | i mean, you can call haskell from C |
| 18:52:01 | Xe | why not nim? |
| 18:52:34 | Araq_ | well you can. |
| 18:52:50 | Xe | the FFI barrier would be a bit insane i'd think |
| 18:53:50 | Araq_ | I have written Nim code consisting only of .emit with C code in it just to get Nim's awesome build environment |
| 18:54:04 | ldlework | Araq_: don't tell anyone that. |
| 18:54:16 | Araq_ | ooops. too late. |
| 18:54:38 | Xe | hmm |
| 18:55:41 | * | lagoon_monster quit (Quit: Page closed) |
| 18:56:02 | Araq_ | xe: calling Nim code from C is documented btw: http://nim-lang.org/docs/backends.html |
| 18:56:04 | Xe | how fleshed out are effects? |
| 18:56:10 | Xe | Araq_: yeah i've done it before |
| 18:56:15 | Xe | very nice actually |
| 18:56:35 | dom96 | "lol, I like Rust but I don't see the need to shit on under-staffed competitors" couldn't have said it better myself |
| 19:01:24 | * | darkf_ joined #nim |
| 19:01:57 | * | darkf quit (Disconnected by services) |
| 19:02:01 | * | darkf_ is now known as darkf |
| 19:04:20 | * | irrequietus quit () |
| 19:06:59 | * | yglukhov joined #nim |
| 19:07:37 | * | pregressive quit (Read error: Connection reset by peer) |
| 19:08:10 | * | pregressive joined #nim |
| 19:08:58 | * | buMPnet quit (Remote host closed the connection) |
| 19:09:35 | * | buMPnet joined #nim |
| 19:11:46 | * | buMPnet quit (Remote host closed the connection) |
| 19:14:59 | * | askatasuna joined #nim |
| 19:16:16 | * | libman joined #nim |
| 19:16:54 | * | libman bows guiltily. |
| 19:17:08 | * | libman left #nim (#nim) |
| 19:17:52 | Araq_ | stefantalpalaru: to answer your question: It's really easy to create races in Nim, but the much bigger problem is that the non-shared-heaps create expressiveness problems |
| 19:18:15 | stefantalpalaru | OK |
| 19:18:32 | Araq_ | we have lots and lots of ways to fight races though. |
| 19:19:27 | Araq_ | but as long as "sorry, the stdlib has no shared dict" is the real problem I am not concerned with the races |
| 19:20:57 | Araq_ | especially since I know how to fight them. I'm pretty confident in our '.locks' annotation. |
| 19:21:04 | pleiosaur | Araq_: take the Go approach and just have the default map type be thread-unsafe :P |
| 19:21:19 | ldlework | Please take the anti-go approach in every case, thanks. |
| 19:21:34 | * | Salewski quit () |
| 19:23:08 | * | bozaloshtsh joined #nim |
| 19:26:53 | * | biscarch joined #nim |
| 19:33:29 | Araq_ | VM: Pegs do not work at compile-time. Do they work now? |
| 19:48:54 | * | Matthias247 quit (Read error: Connection reset by peer) |
| 19:48:58 | * | bozaloshtsh quit (Ping timeout: 250 seconds) |
| 19:52:14 | * | gokr quit (Quit: Leaving.) |
| 20:00:43 | * | Ven joined #nim |
| 20:10:35 | * | Demon_Fox joined #nim |
| 20:19:46 | * | Arrrr quit (Quit: WeeChat 1.2) |
| 20:24:39 | * | jsudlow quit (Ping timeout: 260 seconds) |
| 20:25:26 | * | jsudlow joined #nim |
| 20:26:07 | * | Senketsu quit (Remote host closed the connection) |
| 20:28:04 | * | wuehlmaus quit (Ping timeout: 245 seconds) |
| 20:28:21 | * | Senketsu joined #nim |
| 20:53:27 | * | bozaloshtsh joined #nim |
| 20:55:13 | * | wuehlmaus joined #nim |
| 21:03:50 | * | satbyy_ joined #nim |
| 21:04:48 | * | stefantalpalaru left #nim ("Leaving") |
| 21:08:48 | * | gokr joined #nim |
| 21:12:28 | * | xificurC joined #nim |
| 21:22:10 | * | askatasuna quit (Quit: WeeChat 1.3) |
| 21:24:09 | yglukhov | Araq: does pragma rtl have some special meaning? |
| 21:26:45 | yglukhov | damn, i have to learn to wait for 2 minutes before asking a question... |
| 21:27:56 | ldlework | what? |
| 21:32:53 | ldlework | Araq_: here are the types I came up with to be able to express just type declarations: https://gist.github.com/dustinlacewell/957646128a24cd6c8141 |
| 21:32:57 | ldlework | Araq_: do you see any glaring holes? |
| 21:33:51 | ldlework | these types will render to the raw AST nodes, so we're mostly interested in whether or not I've covered type declarations in terms of the various semantically distinct terms that make them up |
| 21:50:26 | * | brson quit (Ping timeout: 272 seconds) |
| 21:52:37 | * | BitPuffin quit (Ping timeout: 276 seconds) |
| 22:01:04 | * | regtools joined #nim |
| 22:01:46 | Araq_ | TypeKind = enum tkNamed, tkTuple |
| 22:02:04 | Araq_ | pretty sure you left out lots and lots of things ;-) |
| 22:02:25 | Araq_ | distinct types, proc types, iterator types, concepts |
| 22:03:01 | Araq_ | but you're overengineering things |
| 22:03:29 | Araq_ | just model what you're confortable with and use NimNode as the fallback covering the rest |
| 22:03:34 | Araq_ | *comfortable |
| 22:03:34 | * | Ven quit (Quit: My MacBook has gone to sleep. ZZZzzz…) |
| 22:04:46 | Araq_ | you're layering type safety and convenience on top of a system that was designed to cover Nim completely. it's perfectly fine to stop after a useful subset. |
| 22:11:25 | * | brson joined #nim |
| 22:12:31 | * | [1]jsudlow joined #nim |
| 22:13:12 | * | blob_ joined #nim |
| 22:13:51 | blob_ | is it supposed to be possible to build nim 0.13.0 on windows? |
| 22:14:40 | Araq_ | depends on how you build it |
| 22:15:55 | blob_ | when trying to install it from github, I get "Error: unknown operating system: mingw64_nt-6.1" |
| 22:16:02 | * | jsudlow quit (Ping timeout: 276 seconds) |
| 22:16:35 | Araq_ | well don't run .sh scripts on windows. windows has its own scripts |
| 22:18:25 | * | [1]jsudlow is now known as jsudlow |
| 22:21:59 | blob_ | yeah, I'm an idiot. .bat works |
| 22:23:16 | * | desophos quit (Read error: Connection reset by peer) |
| 22:25:35 | * | blob_ quit (Quit: Page closed) |
| 22:25:55 | * | nchambers quit (Ping timeout: 240 seconds) |
| 22:31:48 | * | [1]jsudlow joined #nim |
| 22:32:04 | * | jsudlow quit (Ping timeout: 265 seconds) |
| 22:34:01 | * | [1]jsudlow quit (Read error: Connection reset by peer) |
| 22:34:51 | * | jsudlow joined #nim |
| 22:34:55 | * | darkf quit (Quit: Leaving) |
| 22:35:20 | * | Jesin joined #nim |
| 22:37:15 | * | [1]jsudlow joined #nim |
| 22:39:12 | * | kniteli quit (Ping timeout: 272 seconds) |
| 22:39:19 | * | jsudlow quit (Ping timeout: 245 seconds) |
| 22:39:20 | * | [1]jsudlow is now known as jsudlow |
| 22:40:56 | * | ephja quit (Ping timeout: 255 seconds) |
| 22:42:30 | * | Trustable_2 quit (Remote host closed the connection) |
| 22:44:04 | ldlework | Araq_: yeah I was just encouraged by your words yesterday. |
| 22:44:13 | ldlework | But I guess you too are seeing the scope of the idea now too. |
| 22:44:18 | ldlework | Its huge! |
| 22:44:30 | ldlework | But maybe not impossible if made iteratively... |
| 22:50:23 | * | derka_ joined #nim |
| 22:52:56 | * | derka quit (Ping timeout: 256 seconds) |
| 22:52:56 | * | derka_ is now known as derka |
| 22:55:41 | * | derka quit (Quit: derka) |
| 23:02:56 | * | pregressive quit (Remote host closed the connection) |
| 23:03:29 | * | pregressive joined #nim |
| 23:05:26 | * | jsudlow quit (Ping timeout: 240 seconds) |
| 23:08:41 | * | pregressive quit (Ping timeout: 276 seconds) |
| 23:09:42 | * | Matthias247 joined #nim |
| 23:14:19 | * | gokr quit (Ping timeout: 245 seconds) |
| 23:14:55 | * | jsudlow joined #nim |
| 23:24:26 | * | xificurC quit (Quit: WeeChat 1.3) |
| 23:26:49 | * | vendethiel quit (Ping timeout: 245 seconds) |
| 23:30:38 | * | derka joined #nim |
| 23:35:52 | * | jsudlow quit (Read error: Connection reset by peer) |
| 23:36:34 | * | jsudlow joined #nim |
| 23:48:30 | * | Matthias247 quit (Read error: Connection reset by peer) |
| 23:51:08 | * | jsudlow quit (Ping timeout: 255 seconds) |
| 23:53:15 | derka | Guys |
| 23:53:25 | derka | there is a typo in http://nim-lang.org/news.html#Z2016-01-18-version-0-13-0-released |
| 23:54:02 | derka | in the closure changes first example there is an additional ) at the end, so it should be s.add(proc(): int = return ii*ii) and not s.add(proc(): int = return ii*ii))) |
| 23:56:15 | * | desophos joined #nim |
| 23:56:28 | * | yglukhov quit (Remote host closed the connection) |