| 00:17:23 | disruptek | cps needs more testes. |
| 00:17:26 | disruptek | !repo cps |
| 00:17:26 | disbot | https://github.com/disruptek/cps -- 9cps: 11Continuation-Passing Style for Nim 🔗 15 17⭐ 1🍴 |
| 00:17:36 | disruptek | new version that kinda works more than it did. |
| 00:18:40 | disruptek | you can see how fast zevv's web-server is, for example. |
| 00:20:05 | disruptek | oh, nevermind, it doesn't work in arc. |
| 00:28:24 | FromDiscord | <flywind> cps seems very nice. I will try it. |
| 00:29:20 | disruptek | cool, it's very early but let's see what you can make with it. 😄 |
| 00:30:29 | FromDiscord | <flywind> I think async ecosystem in Nim is very early too. 😜 |
| 00:30:54 | disruptek | there is a lot of room for improvement everywhere. |
| 00:52:29 | * | maier joined #nim |
| 00:57:49 | * | maier quit (Ping timeout: 264 seconds) |
| 01:06:44 | FromDiscord | <flywind> `select` doesn't scale in windows. I happen to work on `ioselctors plus`(based on `wepoll`) for `httpbeast`. Though it is only for test `httpbeast`(not robust) now, it maybe useful for `cps` too. |
| 01:06:45 | FromDiscord | <flywind> https://github.com/xflywind/ioselectors |
| 01:06:54 | FromDiscord | <flywind> https://github.com/xflywind/wepoll |
| 01:07:19 | FromDiscord | <flywind> based on: https://github.com/piscisaureus/wepoll |
| 01:08:57 | FromDiscord | <flywind> Of course, when the project becomes larger, we should have a pure nim version of `wepoll` just like `mio` in `Rust`. |
| 01:09:55 | * | apahl quit (Ping timeout: 272 seconds) |
| 01:11:21 | * | apahl joined #nim |
| 01:13:10 | * | unihernandez22 joined #nim |
| 01:20:57 | * | Jjp137_ quit (Read error: Connection reset by peer) |
| 01:21:14 | * | Jjp137 joined #nim |
| 01:22:17 | * | unihernandez22 quit (Quit: Leaving) |
| 01:22:30 | disruptek | nice. |
| 01:23:27 | disruptek | does the api have timers? |
| 01:24:54 | FromDiscord | <flywind> Current not, because `httpbeast` doesn't need timer. When `ioselectors` is stable, I'm happy to contribute to `cps`. |
| 01:25:23 | disruptek | okay, i don't have a windows machine to test it. maybe i can build one, though. |
| 01:25:58 | disruptek | i was just curious because iirc it's one of the main issues for portability (without regard to performance). |
| 01:26:17 | * | unihernandez22 joined #nim |
| 01:27:38 | FromDiscord | <flywind> I am a newbie in system programming. But I will try to learn that. Thanks for tips. 😄 |
| 01:27:59 | disruptek | i don't really know anything. |
| 01:28:35 | disruptek | wepoll looks pretty decent, though. |
| 01:28:47 | disruptek | it should work okay. |
| 01:29:18 | disruptek | i am hoping to use iouring in linux in any event, so to speak, so rest assured that cps will not depend on any particular event system. |
| 01:30:02 | disruptek | cps has a broader mandate towards providing a paradigm that just happens to be conducive to async design... |
| 01:32:16 | FromDiscord | <flywind> I am also working on `liburing`(rewrite in pure Nim), but doesn't have enough time to finish it. |
| 01:32:46 | disruptek | i only did a small amount, myself. |
| 01:32:49 | disruptek | !repo yourang |
| 01:32:50 | disbot | https://github.com/disruptek/yourang -- 9yourang: 11Thread-safe performant async I/O for Linux 15 6⭐ 0🍴 |
| 01:34:07 | * | unihernandez22 quit (Quit: Leaving) |
| 01:36:40 | FromDiscord | <flywind> BTW, there is also one person is working on a new async runtime. 😜 |
| 01:36:41 | FromDiscord | <flywind> https://github.com/iocrate/netkit |
| 01:37:38 | disruptek | ah, but that's async/await-based. very different. |
| 01:37:54 | * | unai joined #nim |
| 01:37:55 | disruptek | how is your iouring interface going to work? |
| 01:38:00 | * | unai is now known as unihernandez22 |
| 01:38:34 | * | unihernandez22 quit (Client Quit) |
| 01:38:43 | * | unihernandez22 joined #nim |
| 01:39:34 | FromDiscord | <flywind> I just rewrite c code to Nim based on `liburing`. |
| 01:39:54 | FromDiscord | <flywind> https://github.com/axboe/liburing |
| 01:40:00 | disruptek | but will you provide a high-level interface? |
| 01:43:17 | FromDiscord | <flywind> Now I want to only supply similar interface like `liburing`. |
| 01:43:26 | FromDiscord | <flywind> ```nim↵proc io_uring_prepare_readv*(ring: var io_uring, index: int, fd: int, buf: IOVec, off: uint64) =↵ io_uring_prepare_base(ring, index, IOURING_OP_READV, fd, buf.iov_base, uint32(buf.iov_len), off)↵``` |
| 01:43:44 | disruptek | gotcha. |
| 02:03:17 | FromDiscord | <Varriount> Hm, but uring doesn't work on Windows |
| 02:04:11 | FromDiscord | <Varriount> What kernel versions is it available in for Linux? |
| 02:04:26 | disruptek | i wanna say low 4+. 4.14 or something. |
| 02:07:22 | disruptek | i think i want testes to just echo each line of your code and highlight any assertions or exceptions. |
| 02:07:48 | disruptek | less like a test suite and more like a debugging aid. |
| 02:21:39 | disruptek | Running 10s test @ http://localhost:8000 |
| 02:21:39 | disruptek | 10 threads and 10 connections |
| 02:21:39 | disruptek | Thread Stats Avg Stdev Max +/- Stdev |
| 02:21:40 | disruptek | Latency 66.00us 9.94us 437.00us 92.19% |
| 02:21:41 | disruptek | Req/Sec 14.97k 329.88 15.29k 85.15% |
| 02:21:42 | disruptek | Latency Distribution |
| 02:21:44 | disruptek | 50% 69.00us |
| 02:21:46 | disruptek | 75% 71.00us |
| 02:21:49 | disruptek | 90% 74.00us |
| 02:21:51 | disruptek | 99% 103.00us |
| 02:21:54 | disruptek | 1504409 requests in 10.10s, 139.17MB read |
| 02:21:56 | disruptek | Requests/sec: 148963.27 |
| 02:21:59 | disruptek | Transfer/sec: 13.78MB |
| 02:22:01 | disruptek | this is zevv's lil server under cps. single-threaded. |
| 02:27:46 | * | hyiltiz quit (Quit: hyiltiz) |
| 02:31:53 | FromDiscord | <flywind> Wonderful! |
| 02:46:20 | * | audiofile quit (Quit: Default Quit Message) |
| 02:53:40 | * | maier joined #nim |
| 02:56:51 | * | waleee-cl quit (Quit: Connection closed for inactivity) |
| 02:57:07 | * | muffindrake quit (Ping timeout: 260 seconds) |
| 02:58:32 | * | maier quit (Ping timeout: 256 seconds) |
| 02:59:00 | * | muffindrake joined #nim |
| 03:01:17 | FromGitter | <iffy> What's the equivalent of Python's struct.{pack,unpack}? I'm trying to pack an int into a binary string and then read it out again |
| 03:02:27 | FromDiscord | <flywind> !repo struct |
| 03:02:27 | disbot | https://github.com/OpenSystemsLab/struct.nim -- 9struct.nim: 11Python-like 'struct' for Nim 15 25⭐ 5🍴 7& 8 more... |
| 03:03:51 | FromGitter | <iffy> perfect, thank you! |
| 03:04:01 | FromDiscord | <flywind> np |
| 03:10:53 | * | apahl quit (Ping timeout: 272 seconds) |
| 03:11:41 | * | apahl joined #nim |
| 03:17:02 | * | rockcavera quit (Remote host closed the connection) |
| 03:20:42 | * | endragor joined #nim |
| 03:29:25 | * | vicfred quit (Quit: Leaving) |
| 04:03:00 | * | krux02_ quit (Remote host closed the connection) |
| 04:03:19 | * | unihernandez22 quit (Quit: Leaving) |
| 04:03:33 | * | unai joined #nim |
| 04:03:58 | * | unai is now known as Guest15123 |
| 04:04:11 | * | unai_ joined #nim |
| 04:04:39 | * | unihernandez22 joined #nim |
| 04:04:41 | * | unai_ quit (Client Quit) |
| 04:04:41 | * | unihernandez22 quit (Remote host closed the connection) |
| 04:04:41 | * | Guest15123 quit (Client Quit) |
| 04:04:52 | * | unihernandez22 joined #nim |
| 04:06:02 | * | supakeen quit (Quit: WeeChat 2.8) |
| 04:06:05 | * | Kaivo quit (Ping timeout: 240 seconds) |
| 04:06:43 | * | supakeen joined #nim |
| 04:08:17 | * | unihernandez22 quit (Client Quit) |
| 04:11:20 | FromDiscord | <Varriount> Hm, I wonder if using generics to compose API traits would work. |
| 04:12:51 | * | unihernandez22 joined #nim |
| 04:13:11 | FromDiscord | <Varriount> `Buffered[Readable[Writable[FileStream]()` |
| 04:13:18 | * | hyiltiz joined #nim |
| 04:13:18 | * | hyiltiz quit (Changing host) |
| 04:13:18 | * | hyiltiz joined #nim |
| 04:13:18 | * | hyiltiz quit (Client Quit) |
| 04:13:43 | FromDiscord | <Varriount> Though, order would matter in that case |
| 04:15:42 | * | unihernandez22 quit (Client Quit) |
| 04:16:31 | FromDiscord | <Varriount> Anyone know of a language with an easy but efficient data stream model? |
| 04:17:38 | * | hyiltiz joined #nim |
| 04:17:39 | * | hyiltiz quit (Changing host) |
| 04:17:39 | * | hyiltiz joined #nim |
| 04:41:31 | * | Kaivo joined #nim |
| 04:54:50 | * | maier joined #nim |
| 04:59:24 | * | maier quit (Ping timeout: 265 seconds) |
| 05:03:40 | * | jhgalino joined #nim |
| 05:05:02 | * | jhgalino quit (Client Quit) |
| 05:05:14 | * | lritter joined #nim |
| 05:08:28 | * | vicfred joined #nim |
| 05:09:43 | * | vicfred quit (Client Quit) |
| 05:23:19 | * | maier joined #nim |
| 05:25:52 | * | Tlanger joined #nim |
| 05:28:25 | * | Tongir quit (Ping timeout: 264 seconds) |
| 05:34:14 | * | Kaivo quit (Ping timeout: 260 seconds) |
| 05:52:39 | Zevv | haskell? |
| 06:10:04 | * | solitudesf joined #nim |
| 06:18:45 | * | narimiran joined #nim |
| 06:34:53 | voltist | Is there a nim library for downloading from FTP servers where one doesn't have to deal with async stuff? |
| 06:35:27 | voltist | Or is the async features in the standard library not too much hassle for this? I'm not particularly experienced with that kinda stuff in Nim |
| 06:57:43 | Zevv | You can just take the async example and do the waitfor() |
| 06:57:45 | * | PMunch joined #nim |
| 06:57:52 | Zevv | under the hood it does async, but in practice it will just block until done |
| 06:57:56 | Zevv | would that work? |
| 06:58:21 | Zevv | it doesn't force you to leak async into the rest of your program |
| 07:03:47 | voltist | Zevv: Yeah I'm trying that now |
| 07:09:31 | voltist | Hmm, why would I be getting `Exception message: Expected reply '227' got: 530 Please login with USER and PASS` for a public ftp server? |
| 07:11:53 | Oddmonger | hello |
| 07:12:10 | Oddmonger | are "set" defined only as type ? |
| 07:12:30 | Oddmonger | can i do var a = set[int16] (obviously no) |
| 07:13:00 | narimiran | change `=` to `:` |
| 07:13:10 | Oddmonger | ah… oh… sorry |
| 07:13:12 | Oddmonger | thank you |
| 07:24:38 | PMunch | That would define `a` as an empty set of type int16 |
| 07:24:50 | PMunch | So you could do a.incl(5.int16) |
| 07:25:08 | euantor | Looks like `devel` has a change to `asyncdispatch` that broke one of my libraries :( |
| 07:25:46 | euantor | https://github.com/euantorano/serial.nim/issues/44#issuecomment-669027454 - seems like an object was renamed that was exposed publicly. Probably going to break other stuff too |
| 07:25:49 | disbot | ➥ undeclared identifier: 'PCustomOverlapped' ; snippet at 12https://play.nim-lang.org/#ix=2tid |
| 07:26:53 | euantor | Will have to re-think the approach a bit today and see what I can do to resolve it |
| 07:27:42 | narimiran | euantor: would `when (NimMajor, NimMinor) < (1, 3): oneThing; else: otherThing` help? |
| 07:28:54 | PMunch | Was it just a rename? |
| 07:29:00 | euantor | narimiran: possibly, but might be easier if I can just alter the library to use the file types from the stdlib, which should be possible so long as I can get the underlying handle/fd |
| 07:29:20 | euantor | Yeah, a whole bunch of objects were renamed in a commit that doesn’t mention the rename :( |
| 07:29:55 | euantor | Was impossible to even find through GitHub search due to it not being mentioned in the commit message. Had to go through the history on the file to find it |
| 07:30:10 | * | Kaivo joined #nim |
| 07:46:40 | voltist | Is there a standard library proc that can parse scientific notation like "1.3e+04" into a float? |
| 07:49:04 | PMunch | parseFloat from strutils |
| 07:49:09 | PMunch | https://play.nim-lang.org/#ix=2tie |
| 07:51:29 | voltist | Hmmm `Error: unhandled exception: invalid float: 1.3e+05 [ValueError]` |
| 07:51:51 | voltist | Oh too big maybe? |
| 07:52:43 | voltist | Yeah |
| 07:54:06 | narimiran | there is `parseBiggestFloat` in parseutils |
| 07:54:44 | narimiran | hmmm, but `parseFloat` already calls it, so it shouldn't make a difference!? |
| 07:56:57 | narimiran | are you sure you're not having a whitespace after the last character? |
| 07:57:12 | narimiran | because `1.3e+05 [ValueError]` has two spaces there |
| 07:57:44 | narimiran | an i can trigger the same thing if i try to parse `1.3e+05 ` |
| 07:57:49 | narimiran | voltist: ^ |
| 08:05:25 | * | Vladar joined #nim |
| 08:09:03 | voltist | Sorry AFK |
| 08:09:44 | voltist | Yeah twas a whitespace issue |
| 08:26:29 | * | oculuxe quit (Quit: blah) |
| 08:26:47 | * | oculux joined #nim |
| 08:26:56 | * | andinus quit (Remote host closed the connection) |
| 08:32:32 | * | ehmry joined #nim |
| 08:37:05 | * | NimBot joined #nim |
| 08:48:21 | voltist | That feeling in your gut when you realize "I'm gonna have to use python for this" |
| 08:48:43 | voltist | I get it less now that I'm an avid Nim user, but it still happens |
| 08:50:10 | FromDiscord | <Varriount> voltist: You should see the Python DSL I wrote for Nimterop |
| 08:50:41 | PMunch | What do you have to resort to Python for now? |
| 08:51:02 | PMunch | Only thing I've used Python for in the past couple of years has been as a scripting language for something that didn't accept anything else.. |
| 08:51:21 | PMunch | (Until I wrote a PR that made it accept dynamic libraries so I could rewrite my code in Nim) |
| 08:51:53 | PMunch | Anyone know why this wasn't included in the latest release by the way? https://github.com/nim-lang/Nim/pull/14137 |
| 08:51:54 | disbot | ➥ Add RSA key reading and encrypt/decrypt to openssl |
| 08:52:08 | voltist | It's not that there is any one thing I can't do, just that I'm doing some quick data processing that doesn't need to be fast (and Python has an excellent data science ecosystem as well :)) |
| 08:52:22 | voltist | So its just a major convenience |
| 08:52:29 | PMunch | Ah |
| 08:55:15 | FromDiscord | <Varriount> PMunch: Primarily ease of subprocess launching |
| 08:55:53 | FromDiscord | <Varriount> https://github.com/Varriount/Nimterop-Deprocessor |
| 08:56:23 | FromDiscord | <Varriount> See my regex-powered monstrosity |
| 08:57:33 | PMunch | Hmm, what am I looking at here? |
| 08:58:13 | FromDiscord | <Varriount> It's mainly a tool to help wrap the Windows SDK |
| 08:58:14 | voltist | Although, it would be more fun if I did it in Nim... |
| 08:58:37 | FromDiscord | <Varriount> PMunch: Which file? |
| 09:09:11 | * | waleee-cl joined #nim |
| 09:12:21 | PMunch | No just in general :P I'm not sure why you sent that to me :P |
| 09:12:49 | PMunch | Or did you just reply to me with why you ended up using Python instead of Nim and then send the github link to voltist? |
| 09:13:32 | FromDiscord | <Varriount> I meant to show both of you |
| 09:13:46 | FromDiscord | <Varriount> *should have shown |
| 09:23:33 | PMunch | Hmm, it would be cool if the new-tab page of browsers could be set to a local HTML file that had access to some special JS API to get information about your browser |
| 09:23:47 | PMunch | So that you could create a custom start page with recently closed tabs, bookmarks, etc. |
| 09:34:00 | PMunch | Hmm, it seems to be doable in FF at least with a pretty minimal addon |
| 09:40:58 | shashlick | @varriount you need to explain this at some point |
| 09:46:56 | FromDiscord | <Varriount> @shashlick Why I'm using it? |
| 09:47:03 | FromDiscord | <Varriount> Or how it works? |
| 09:51:13 | FromDiscord | <Varriount> shashlick: My biggest reason for using it is that I need to perform regex replacements on header files before they are processed, and want to run Nimterop in parallel |
| 10:10:22 | * | andinus joined #nim |
| 10:10:37 | * | Trustable joined #nim |
| 10:31:39 | * | Trustable quit (Remote host closed the connection) |
| 10:32:39 | FromDiscord | <Ricky Spanish> is there a difference between ```{.push,exportc,dynlib.}``` and ```{.stdcall,exportc,dynlib.}``` because nim seems to have a seizure with the latter even though iv found multiple examples doing it |
| 10:34:00 | FromDiscord | <Ricky Spanish> or lemme ask in a different way, what does .push actually do |
| 10:34:27 | FromDiscord | <Rika> one applies exportc and dynlib to everything below it before a pop↵and the other only works when its beside a proc signature |
| 10:34:47 | FromDiscord | <Rika> push pushes the pragmas onto a "stack" which then applies the pragmas to everything below it |
| 10:34:53 | FromDiscord | <Rika> should be on the manual actually |
| 10:35:16 | FromDiscord | <Ricky Spanish> aaah ok thanks so much for explaining @Rika |
| 10:36:24 | PMunch | https://nim-lang.org/docs/manual.html#pragmas-push-and-pop-pragmas |
| 10:36:41 | voltist | Does anyone know if it's possible to embed a web page in NiGui? Like some sort of embedded browser? |
| 10:37:02 | FromDiscord | <Rika> that is complex |
| 10:37:17 | FromDiscord | <Rika> you'd need to actually embed a browser as you say |
| 10:38:01 | FromDiscord | <Rika> then somehow render the output onto your window (i dont know how it would exactly work) |
| 10:39:11 | voltist | Yeah |
| 10:44:20 | PMunch | It shouldn't be impossible I think.. |
| 10:44:53 | PMunch | Not quite sure how you'd do the browser <-> nigui bridging though.. |
| 10:49:13 | voltist | Problem might be making it cross-platform |
| 10:55:26 | PMunch | Hmm, I think embedded Chromium should be cross platform |
| 10:55:49 | PMunch | Or Gecko for that matter, but I'm not sure if that has bindings for Nim |
| 10:55:54 | PMunch | You could also try Sciter |
| 10:56:01 | PMunch | Depends on what you actually want to show |
| 11:09:53 | FromDiscord | <Clyybber> servo would probably be a good choice |
| 11:19:02 | FromDiscord | <Yardanico> Sciter is not really about real web pages though |
| 11:19:22 | FromDiscord | <Yardanico> It's their own HTML + CSS + TiScript engine |
| 11:20:30 | PMunch | Which is why I said "depends on what you actually want to show" ;) |
| 11:20:47 | PMunch | If he just wanted to create a more complex UI component using HTML then it could work |
| 11:25:22 | * | narimiran quit (Ping timeout: 246 seconds) |
| 11:29:57 | * | narimiran joined #nim |
| 11:31:06 | Oddmonger | come on, no Nim on https://benchmarksgame-team.pages.debian.net/benchmarksgame/index.html |
| 11:38:14 | * | Vladar quit (Quit: Leaving) |
| 11:38:38 | FromDiscord | <kodkuce> @treeform your ws has no exposed ws headers right, there is that TODO, https://github.com/treeform/ws/blob/master/src/ws.nim 142.. i want to add authorization token to reqest, so only way to do is to hack this right |
| 11:56:19 | * | waleee-cl quit (Quit: Connection closed for inactivity) |
| 11:59:33 | * | audiofile joined #nim |
| 12:00:18 | FromDiscord | <kodkuce> hmm and 1 wierd thing, if no req hader i return 404 code and dont even build ws but your client for some reason says its open xD |
| 12:04:15 | * | rockcavera joined #nim |
| 12:05:17 | * | krux02 joined #nim |
| 12:06:02 | * | supakeen quit (Quit: WeeChat 2.8) |
| 12:06:39 | * | supakeen joined #nim |
| 12:06:46 | * | abm joined #nim |
| 12:10:05 | * | vicfred joined #nim |
| 12:19:22 | FromDiscord | <kodkuce> @treeform https://pastebin.com/xZWrrGuq |
| 12:26:01 | dom96 | Oddmonger, I don't know if the author of that changed their mind, but they were pretty against Crystal/Nim being included not too long ago: https://web.archive.org/web/20170316004217/http://benchmarksgame.alioth.debian.org/play.html |
| 12:27:15 | FromDiscord | <kodkuce> am i retarded or what if ws connection fails should it default to http response or what? |
| 12:27:28 | * | mbuchel quit (Ping timeout: 256 seconds) |
| 12:29:13 | * | mbuchel joined #nim |
| 12:40:54 | Oddmonger | dom96: ok |
| 12:41:00 | * | FromDiscord quit (Remote host closed the connection) |
| 12:41:17 | * | FromDiscord joined #nim |
| 12:44:24 | FromDiscord | <Ricky Spanish> @kodkuce doesent the ws implementation provide defaults for connection failures during the initiali http handshake with the server? |
| 12:49:26 | FromDiscord | <kodkuce> it rasises soem exeptions form what i see it does not default to read normal HTTP resonse "raise newException(WebSocketError, ...." |
| 13:22:32 | * | dulsi_ joined #nim |
| 13:25:04 | * | dulsi quit (Ping timeout: 246 seconds) |
| 13:28:42 | FromDiscord | <Zachary Carter> Voltist maybe checkout webview |
| 13:30:31 | FromDiscord | <Zachary Carter> https://github.com/webview/webview |
| 13:39:51 | * | dulsi_ is now known as dulsi |
| 13:50:32 | * | rockcavera quit (Remote host closed the connection) |
| 13:51:09 | FromDiscord | <Zachary Carter> dom96: It looks like Julia was mentioned in that blurb but now is on the benchmark game |
| 13:52:05 | FromGitter | <iffy> What would be a good way to read from one File (at an offset) and write to another (at an offset)? File.readX procs all expect non-file things. Am I stuck reading into memory first? |
| 13:52:32 | FromDiscord | <Zachary Carter> streams? |
| 13:52:55 | FromDiscord | <Zachary Carter> you can `setPosition` and `getPosition` |
| 13:53:07 | FromDiscord | <Zachary Carter> https://nim-lang.org/docs/streams.html#getPosition%2CStream |
| 13:54:15 | FromGitter | <iffy> @zacharycarter I'm not seeing a proc that will read from one stream and write directly to another stream |
| 13:54:51 | FromGitter | <iffy> I guess just `dst.write(src.read())`? |
| 13:55:56 | FromGitter | <iffy> err... there is no `proc read()` which returns a string (that isn't the whole file) |
| 13:56:55 | FromDiscord | <arnetheduck> so what's with the new stale bot? I spend time on creating minimized test cases for issues that I run across - they're still not fixed, and now a bot closes them? |
| 13:58:09 | FromDiscord | <mratsim> https://github.com/nim-lang/Nim/issues/15159 |
| 13:58:09 | disbot | ➥ [RFC] Stale bot |
| 13:58:11 | dom96 | The bot pings issues to make sure they are still relevant |
| 13:58:28 | dom96 | it doesn't close them unless you (or someone else) doesn't respond |
| 13:59:21 | * | Vladar joined #nim |
| 14:01:52 | FromDiscord | <Clyybber> Almost everybody is against it, for ovious reasons |
| 14:07:01 | * | PMunch quit (Quit: Leaving) |
| 14:13:13 | * | schurig joined #nim |
| 14:14:36 | schurig | in a proc, can I dynamically allocate an array (e.g. for a height map) process and return it? All I saw was to make use seq.add(), but I want to access arbitrary elements |
| 14:19:54 | shashlick | What's the fastest sha512 implementation |
| 14:23:33 | FromDiscord | <Rika> schurig: newSeq[T](size) |
| 14:26:52 | FromDiscord | <Zachary Carter> iffy: well you have to know how much data you want to read |
| 14:26:56 | FromDiscord | <Zachary Carter> and write |
| 14:27:11 | FromDiscord | <Zachary Carter> no it won't be as simple as `dst.write(src.read())` |
| 14:35:04 | schurig | Rika: thanks, worked |
| 14:35:04 | * | Guest39052 quit (Ping timeout: 246 seconds) |
| 14:35:46 | * | vicfred quit (Quit: Leaving) |
| 14:39:42 | FromDiscord | <Zachary Carter> Who developed the stale bot and let it loose? |
| 14:40:01 | FromDiscord | <Rika> you mean who let it loose |
| 14:40:54 | FromDiscord | <Zachary Carter> well I'm curious to know answers to both questions |
| 14:41:06 | FromDiscord | <kodkuce> i just now figured out that var mystuff is same as my_stuff |
| 14:41:20 | FromDiscord | <kodkuce> dident know you can put _ in var name |
| 14:41:22 | FromDiscord | <Zachary Carter> and var myStuff |
| 14:41:56 | FromDiscord | <Rika> but not Mystuff |
| 14:42:13 | FromDiscord | <kodkuce> and no _mystuff |
| 14:43:16 | FromDiscord | <Zachary Carter> ah narimiran did it |
| 14:43:42 | FromDiscord | <Rika> _mystuff is impossible; leading or trailing underscores are not allowed |
| 14:43:53 | FromDiscord | <Rika> while Mystuff is still allowed, just not equal to mystuff |
| 14:47:06 | * | dadada joined #nim |
| 14:47:31 | * | dadada is now known as Guest10225 |
| 14:51:25 | FromDiscord | <Zachary Carter> Someone should write a bot to iterate through all the issues that are marked stale and leave a comment |
| 14:51:27 | FromDiscord | <Zachary Carter> robot wars |
| 14:53:14 | disruptek | no point. |
| 14:54:05 | disruptek | varriount: see concepts. |
| 14:54:24 | FromDiscord | <Zachary Carter> I mean I would counter with there's no point of having a stale bot |
| 14:54:25 | FromDiscord | <Zachary Carter> but... |
| 14:54:26 | disruptek | oh i'm way, way far back in scroll. 🤣 |
| 14:55:08 | disruptek | i don't get enough email these days, so it's nice to hear from stale[bot]. |
| 14:57:01 | FromDiscord | <Zachary Carter> I'll email you |
| 14:57:04 | FromDiscord | <Zachary Carter> dick picks |
| 14:57:08 | FromDiscord | <Zachary Carter> (edit) 'picks' => 'pics' |
| 14:57:21 | FromDiscord | <Zachary Carter> but that's it |
| 14:57:58 | disruptek | beggars can't be choosers. |
| 14:59:32 | dom96 | I'm honestly amazed by the push back. |
| 15:00:41 | Yardanico | because many people (including me) don't see advantages of the stalebot in the context of nim repo :) |
| 15:00:42 | narimiran | dom96: i guess it is only three of us for it, against the whole world against it |
| 15:02:31 | FromDiscord | <kodkuce> what is stalebot? |
| 15:02:43 | narimiran | it is modern day hitler |
| 15:02:49 | Yardanico | a bot which marks old issues (which didn't have activity for a long time) as "stale" |
| 15:02:52 | dom96 | I think this is a classic case of being naturally against something that you are not familiar with. |
| 15:02:58 | Yardanico | and then if still no one replies it's closed |
| 15:03:05 | Yardanico | dom96: yeah, sure, let's just forget about old issues, who needs them |
| 15:03:12 | Yardanico | as we discussed yesterday, we just need to close all issues older than 1 day |
| 15:03:46 | dom96 | Not every "issue" is a relevant issue, some are not even issues! |
| 15:03:47 | Yardanico | most of old nim issues are still not solved |
| 15:03:52 | Yardanico | dom96: most of them are |
| 15:03:54 | FromDiscord | <Zachary Carter> what's there to get familiar with? |
| 15:04:05 | FromDiscord | <Zachary Carter> well just arbitrarily closing old issues doesn't solve that problem |
| 15:04:11 | dom96 | It's not arbitrary |
| 15:04:20 | FromDiscord | <Zachary Carter> it is if people don't leave a comment |
| 15:04:25 | dom96 | so leave a comment |
| 15:04:30 | dom96 | if you care about an issue leave a comment |
| 15:04:32 | FromDiscord | <Zachary Carter> not everyone is going to though |
| 15:04:33 | dom96 | I don't see the problem here |
| 15:04:35 | Yardanico | it's not about "caring" |
| 15:04:44 | dom96 | so if nobody does then why should we keep these issues open? |
| 15:04:45 | Yardanico | we shouldn't just "forget" about issues even if the original poster doesn't use nim anymore |
| 15:04:48 | dom96 | obviously nobody cares about them |
| 15:04:51 | Yardanico | dom96: because they're still valid issues |
| 15:05:01 | FromDiscord | <Zachary Carter> there are people that opened issues that are relevant that aren't active community members anymore |
| 15:05:26 | Yardanico | I don't understand why do you not like old issues so much |
| 15:05:35 | Yardanico | they don't make the repo worse |
| 15:05:57 | dom96 | I don't dislike old issues |
| 15:06:00 | FromDiscord | <Zachary Carter> I don't think lemonboy is going to suddenly reappear and be like - oh I care about these issues now because of stale bot |
| 15:06:01 | dom96 | In fact the very opposite |
| 15:06:07 | dom96 | I want to be reminded of all these old issues |
| 15:06:14 | dom96 | I want others to be reminded too |
| 15:06:17 | disruptek | then visit the repo some time. |
| 15:06:38 | Yardanico | yeah, so we just mark them as stale and close because we want to be reminded of them |
| 15:06:52 | Yardanico | even if it's already solved, you still need a test-case so it doesn't break in the future |
| 15:06:53 | disruptek | many, many people put effort in to enumerate nim's shortcomings. |
| 15:07:57 | Yardanico | I guess you all already saw https://github.com/nim-lang/Nim/issues/15159 :) |
| 15:07:58 | disbot | ➥ [RFC] Stale bot |
| 15:08:01 | disruptek | i think it would make more sense to close issues that don't have a test case we can monitor with software. |
| 15:08:30 | disruptek | if it's a discussion, it belongs in the rfc repo. |
| 15:08:38 | Yardanico | this is directly related to the nim _repo_ though |
| 15:09:01 | narimiran | disruptek: agree about don't having a test case, or having one which you can only replicate if you use special custom made switches |
| 15:09:11 | Yardanico | hehe |
| 15:09:30 | Yardanico | well in my issues the first test case I make is usually the one which fails :) |
| 15:09:34 | disruptek | i still want to finish my tool that builds a test suite from issues. |
| 15:09:35 | Yardanico | but I don't have many issues outside of ARC |
| 15:09:36 | narimiran | i had so many false negatives, until i've gone through the code and noticed `when declared(someShit)` |
| 15:09:52 | narimiran | or `when defined`, i always confuse the two |
| 15:10:08 | Yardanico | yeah timo's issues are not easy for me to understand quite often |
| 15:10:22 | narimiran | ...to put it mildtly |
| 15:10:26 | narimiran | *mildly |
| 15:11:00 | narimiran | if we cannot run `tissue` on your issue to reproduce what you've reported — adios |
| 15:11:21 | dom96 | sounds fair |
| 15:12:19 | FromDiscord | <treeform> @kodkuce Yeah it should not say open when it 404s, ill take a look. |
| 15:13:33 | * | vicfred joined #nim |
| 15:15:43 | disruptek | i'm getting to where i have too many arc blockers to make progress on projects. |
| 15:17:28 | narimiran | so, how do we proceed with some issue-closing bot for issues that don't have a code block with a reproducible bug? because there are some valid issues without it too |
| 15:18:20 | disruptek | for one thing, it would make much more sense to have the template indicate this behavior so that new users have the right expectation set. |
| 15:18:22 | narimiran | what about stale PRs? do we want to ask the author to rebase it on the latest devel? |
| 15:18:39 | disruptek | then, we only enact the bot on issues created since, say, august 1st. |
| 15:18:57 | disruptek | it's hard to consider PRs that aren't rebased. |
| 15:18:58 | Yardanico | well yeah, I agree about PRs a bit |
| 15:19:02 | Yardanico | we have quite a lot of staled once |
| 15:19:04 | Yardanico | ones* |
| 15:19:54 | Yardanico | also btw about https://github.com/nim-lang/Nim/pull/10781 |
| 15:19:55 | disbot | ➥ default to threads:on (#9878) |
| 15:20:00 | Yardanico | there apparently was an accepted RFC about this |
| 15:20:12 | Yardanico | https://github.com/nim-lang/Nim/issues/9878 |
| 15:20:13 | disbot | ➥ [RFC] switch threads to on by default |
| 15:20:24 | Yardanico | it's accepted for 1.5 years without progress :) |
| 15:22:28 | dom96 | It's okay, let me reject it now :P |
| 15:22:36 | Yardanico | why |
| 15:22:38 | Yardanico | it's a good idea :) |
| 15:22:59 | * | Vladar quit (Quit: Leaving) |
| 15:23:00 | Yardanico | but it should be moved to RFCs repo |
| 15:23:59 | FromDiscord | <treeform> @kodkuce I have added a 404 test to ws, can you show me how your case differs? https://github.com/treeform/ws/blob/master/tests/test_404.nim |
| 15:26:04 | FromDiscord | <kodkuce> yo sec |
| 15:30:14 | * | ForumUpdaterBot quit (Remote host closed the connection) |
| 15:30:21 | * | ForumUpdaterBot joined #nim |
| 15:31:49 | FromDiscord | <kodkuce> ok am b |
| 15:34:20 | FromDiscord | <treeform> Also sorry I don't think I will be merging your HTTP headers PR because well Browser HTML5 API does not support it. |
| 15:36:24 | FromDiscord | <treeform> But I am interested in a reproducible case for your other bug. |
| 15:36:44 | * | lritter quit (Ping timeout: 256 seconds) |
| 15:38:59 | FromDiscord | <kodkuce> sent a long message, see http://ix.io/2tjY |
| 15:39:46 | * | arecacea1 quit (Read error: Connection reset by peer) |
| 15:40:42 | FromDiscord | <kodkuce> about my PR hmm, i am no network expert or whatewer but is it not the point of using jwt tokens for autorization, so i wanted to prevent ws connection even establishing if no token, ofc plan to use https ? |
| 15:41:37 | * | arecacea1 joined #nim |
| 15:42:07 | FromDiscord | <kodkuce> @treeform what cookies has to do with anything? |
| 15:42:26 | shashlick | disruptek: tissue can create tests out of issues |
| 15:43:39 | FromDiscord | <treeform> @kodkuce setting cookis is a known way to set a single websocket header. There are no other ways it looks like. |
| 15:45:29 | FromDiscord | <treeform> if you want to set headers from HTML5 API that's the only one you can. |
| 15:45:31 | FromDiscord | <kodkuce> hmm duno if i am too dumb or what but isent header sent with HTTP request itself |
| 15:46:04 | FromDiscord | <treeform> yeah but you are not allowed to add your own HTTP headers ... its not in the spec. |
| 15:46:22 | FromDiscord | <treeform> the only one browsers do is cookies |
| 15:46:41 | FromDiscord | <kodkuce> hmm then how you add jwt tokens to http requests if not in header? |
| 15:47:35 | FromDiscord | <treeform> you don't use http headers to pass the jwt token to a websocket... |
| 15:47:47 | FromDiscord | <treeform> you pass it in like a jwt packet after you connected |
| 15:47:53 | FromDiscord | <treeform> if you want to use jwt |
| 15:48:27 | FromDiscord | <dom96> Isn't the typical flow that it gets passed via a Set-Cookie header? |
| 15:48:41 | FromDiscord | <treeform> To a websocket? I don't think so. |
| 15:49:24 | FromDiscord | <kodkuce> hmm isent a websocket just a http request that gets upgraded, so i am doing http request and pass jwt in header, if jwt is not good i newer upgrade socket so its not an websocket connection ? |
| 15:49:56 | FromDiscord | <treeform> yes but passing jwt header is not allowed, there isn't an API for it. |
| 15:50:17 | FromDiscord | <treeform> you had to PR an API to set header for my library `ws` |
| 15:50:22 | FromDiscord | <treeform> but you can't PR every browser to add that? |
| 15:50:28 | FromDiscord | <dom96> You can pass headers when the ws connection is established |
| 15:50:34 | FromDiscord | <treeform> No you can't |
| 15:50:37 | FromDiscord | <dom96> Why not? |
| 15:50:38 | FromDiscord | <treeform> how me |
| 15:50:40 | FromDiscord | <treeform> (edit) 'how' => 'show' |
| 15:51:09 | FromDiscord | <treeform> Show me in Chrome or Firefox how you would do that. |
| 15:51:53 | FromDiscord | <dom96> Doesn't the browser pass cookies implicitly in the HTTP request it makes when connecting via WS? |
| 15:52:04 | FromDiscord | <kodkuce> hmm w8 , so if i am not doing an websockt connection, emample am doing just an normal Get request i am allowed to put jwt in header , but if i do it for WS i am not, is that what you saying? |
| 15:52:06 | FromDiscord | <treeform> yes the only way to pass it is via cookies that is correct |
| 15:52:07 | * | dddddd quit (Remote host closed the connection) |
| 15:52:36 | FromDiscord | <treeform> @kodkuce yeah that's what I am saying. I don't like it. But that's how it is. |
| 15:53:31 | FromDiscord | <kodkuce> ok q number 2, what bad stuff will heppend to me in i do 🙂 |
| 15:53:32 | FromDiscord | <treeform> This is how normal people to the JWT with websocket: https://gist.github.com/jfromaniello/8418116#gistcomment-2245753 |
| 15:53:58 | FromDiscord | <treeform> The just create their own "authenticate" packet and go on from there. |
| 15:54:11 | FromDiscord | <treeform> You can call it "login", "auth" etc... up to you |
| 15:54:31 | disruptek | normal people sure are creepy. |
| 15:54:54 | FromDiscord | <kodkuce> hmm cuz atm am passing it in header and it all works |
| 15:55:20 | FromDiscord | <treeform> Can you show me how you would do that from the browser? |
| 15:56:00 | FromDiscord | <treeform> I am making a WebSocket API, not some WebSocket+Header API. |
| 15:56:45 | FromDiscord | <dom96> I don't know how ws' API looks like, but I would assume that you have access to the raw HTTP request before you call ws' code |
| 15:57:01 | FromDiscord | <dom96> so ws doesn't need to care about any extra headers at all |
| 15:57:03 | FromDiscord | <treeform> What I don't want people todo, is make rely on headers, then say "hey" my server supports websockets, but it doesn not, because no browser can use it because it can't set headers. |
| 15:57:52 | FromDiscord | <kodkuce> oh meybe you right |
| 15:57:53 | FromDiscord | <treeform> @dom96 only on the server side, on the client side I don't expose access to headers because browser don't expose access to headers. |
| 15:58:20 | FromDiscord | <treeform> Well except the cookie header... that you can set. |
| 15:58:23 | FromDiscord | <kodkuce> sec i was testing whole time in godot editor did not build for browser |
| 15:58:34 | FromDiscord | <dom96> oh, I'm not thinking at all about the client |
| 15:58:48 | disruptek | situation normal, then. |
| 15:58:50 | FromDiscord | <treeform> this conversion was all about the client |
| 15:59:08 | FromDiscord | <treeform> on the server side you can do whatever you want. |
| 16:06:05 | * | dddddd joined #nim |
| 16:07:22 | * | maier quit (Ping timeout: 265 seconds) |
| 16:09:22 | FromDiscord | <kodkuce> yep i think you are right, i was whole time testing vie godot but not in browser and it worked, but when build for browser from what i see my header gets striped away |
| 16:09:40 | FromDiscord | <kodkuce> https://media.discordapp.net/attachments/371759389889003532/740602431565725816/2020-08-05_180648.png |
| 16:09:43 | FromDiscord | <kodkuce> 😦 |
| 16:09:56 | FromDiscord | <kodkuce> no tocker from what isee |
| 16:10:32 | disruptek | my great grandfather got his tocker shot off in the war. |
| 16:11:51 | FromDiscord | <kodkuce> wtf |
| 16:12:37 | disruptek | i hear that a lot. |
| 16:19:22 | Zevv | disruptek: Arak told me to keep an eye on you during his absence. I'm watching you. |
| 16:19:33 | disruptek | wut |
| 16:19:34 | Zevv | No scaring the visitors |
| 16:20:03 | disruptek | i'm not scary. |
| 16:20:12 | Zevv | Be nice. Be polite. Be well-manared |
| 16:20:13 | Yardanico | you can be |
| 16:20:15 | Yardanico | to new people :) |
| 16:20:16 | disruptek | do you wanna see a picture of my great-grandfather's tocker? |
| 16:20:25 | Zevv | before or after the incident? |
| 16:20:52 | disruptek | how could i take a picture before? |
| 16:20:59 | disruptek | i was holding a rifle. |
| 16:21:02 | FromDiscord | <treeform> Trying to debug this odd --gc:arc and {.threadvar.} but... It just segfaults trying to add to a table for the first time. Any one know which commit could have caused it? |
| 16:21:10 | Yardanico | give me a snippet |
| 16:21:14 | Yardanico | I'll find it for you |
| 16:21:36 | FromDiscord | <treeform> I can't distill it to a snippet. It does not happen in a small code base. |
| 16:21:47 | FromDiscord | <treeform> Well I can try. |
| 16:21:52 | disruptek | treeform: try manual init the table, i guess. |
| 16:22:06 | disruptek | i don't have this problem with eventqueue in cps. |
| 16:22:19 | disruptek | i have other problems... |
| 16:22:36 | disruptek | Zevv: we really need this arc thing fixed. |
| 16:22:37 | FromDiscord | <treeform> I know that Jun2's commit 4301a3da9d333de88cf01cd952760716e8117bfa, after that its broken. |
| 16:23:00 | FromDiscord | <treeform> its a recent breakage in the last 2 months. |
| 16:23:11 | Yardanico | try before/after the optimizer PR |
| 16:23:18 | Zevv | disruptek: you hold clyybber |
| 16:23:29 | Yardanico | @treeform try before https://github.com/nim-lang/Nim/pull/14962 |
| 16:23:31 | disbot | ➥ An optimizer for ARC |
| 16:23:57 | Zevv | while I talk to him, persuasively |
| 16:24:04 | disruptek | i'm minimizing an inheritance bug with exceptions in macros. |
| 16:24:12 | disruptek | works in cpp, not in c. |
| 16:25:09 | FromDiscord | <treeform> @Yardanico I'll try it. |
| 16:25:28 | Yardanico | also you can patch 1 line in the compiler |
| 16:25:34 | Yardanico | to see if it's a cursor inference fault |
| 16:25:39 | FromDiscord | <kodkuce> @treeform https://pastebin.com/hmN04x2R |
| 16:25:54 | FromDiscord | <kodkuce> here this confused me, duno if this is bug or i am traded |
| 16:25:56 | disruptek | Zevv: how does 0.0.13 look? |
| 16:26:12 | Yardanico | @treeform line 1005 in injectdestructors.nim |
| 16:26:16 | Yardanico | computeCursors |
| 16:26:25 | Yardanico | if you comment that one and your code works - it's a cursor inference bug |
| 16:26:39 | Yardanico | I'm talking about latest devel of course |
| 16:26:53 | Zevv | disruptek: works for my as far as I tested |
| 16:27:05 | Zevv | If I found the courage I will continue growing tzevv |
| 16:27:19 | disruptek | AIIEIEEE |
| 16:27:35 | disruptek | i need to make my testes bigger. |
| 16:27:44 | disruptek | i mean, better. |
| 16:28:15 | Zevv | what's with the mutant thingy, is that something you want to keep? |
| 16:28:57 | disruptek | i want to be able to operate on values. |
| 16:29:11 | disruptek | maybe it's a stupid idea. it doesn't work yet, anyway. |
| 16:29:16 | Zevv | ah ok |
| 16:30:53 | Zevv | so, one optimization we need to make |
| 16:30:56 | Zevv | is not cps's non-cps blocks |
| 16:31:07 | disruptek | yeah, i know. |
| 16:31:18 | FromDiscord | <treeform> is there a way to choosenim my location (the git checkout of nim)? |
| 16:31:27 | disruptek | it's probably one line of code, if you wanna tackle it. |
| 16:31:28 | Yardanico | i think there is but I don't use it |
| 16:32:01 | Zevv | disruptek: I don't dare to admit it. But I really didn't spend enought time grokking your code yet. |
| 16:32:05 | disruptek | fork gitnim. |
| 16:32:24 | disruptek | the code is crap. but it's useful because it passes tests. |
| 16:33:11 | Zevv | right. anyway, in the original xfrms were only done if `isCpsBlock` agreed. Does that still hold? |
| 16:33:20 | FromDiscord | <treeform> compiling nim takes soo long |
| 16:33:27 | Yardanico | not really :P |
| 16:33:32 | Yardanico | ./koch temp is pretty fast |
| 16:33:37 | disruptek | probably not, because it's a little more complex than that. |
| 16:33:37 | Yardanico | because it uses -d:leanCompiler |
| 16:33:38 | Zevv | treeform: not for me. Are you compilking nim with a debug compiler maybe? |
| 16:33:42 | FromDiscord | <treeform> I guess I am doing build_all |
| 16:33:47 | Yardanico | that's unneccessary |
| 16:34:03 | disruptek | ~gitnim |
| 16:34:04 | disbot | gitnim: 11https://gitnim.com/ -- choosenim for choosey nimions -- disruptek |
| 16:34:04 | Zevv | that builds from c sources and builds again and again to make sure the 2nd build matches the 3d build |
| 16:34:06 | Yardanico | it rebuilds nim from csources and then builds new nim with that and builds new nim with new nim |
| 16:34:08 | Zevv | it's doing all the bootstrap |
| 16:34:13 | Yardanico | same as ./koch danger |
| 16:34:19 | Yardanico | ./koch boot I mean |
| 16:34:30 | Yardanico | if you just want to test a change in the compiler, ./koch temp is what you should use usually |
| 16:34:39 | FromDiscord | <treeform> what should I run instead? |
| 16:34:42 | Yardanico | ./koch temp |
| 16:34:44 | Zevv | ./koch temp |
| 16:34:49 | Yardanico | e.g. |
| 16:34:53 | Zevv | ok, I'll shut up, there's echo on the line |
| 16:34:55 | FromDiscord | <treeform> looks like its working |
| 16:34:57 | disruptek | lol |
| 16:34:59 | Yardanico | ./koch temp c -r file.nim |
| 16:35:22 | Yardanico | Zevv: you sent yours two seconds after me :P |
| 16:35:42 | Zevv | time is relative on IRC. It's a distributed network |
| 16:35:48 | Yardanico | i know |
| 16:35:51 | Yardanico | but two seconds |
| 16:35:53 | Zevv | so messages don't necesserliy come in order |
| 16:36:11 | Zevv | and https://en.wikipedia.org/wiki/Pre-echo |
| 16:36:17 | Yardanico | lol |
| 16:36:39 | Zevv | it's due to quantization noise being spread over the entire transform-window of the IRC codec |
| 16:36:44 | Yardanico | mhm |
| 16:36:51 | disruptek | Zevv: inside a non-cps block you have to handle returns, named breaks, and other calls which might change control flow. so "non-cps block" is a misnomer much of the time. |
| 16:36:53 | FromDiscord | <kodkuce> @treeform did you check pastebin? |
| 16:36:56 | Zevv | I like good explanations. Did I ever show http://zevv.nl/bug-a-day/ ? |
| 16:37:25 | Zevv | disruptek: no, but if I have a while that just whiles over some locals while not doing CPS things, I don't want that while transformed |
| 16:37:27 | FromDiscord | <treeform> @kodkuce yes, but focused on this arc issue now |
| 16:37:29 | dom96 | > is there a way to choosenim my location (the git checkout of nim)? |
| 16:37:31 | dom96 | yes |
| 16:37:38 | dom96 | choosenim ~/path/to/nim |
| 16:37:40 | disruptek | Zevv: good, yes, yes, good. |
| 16:38:13 | disruptek | so many bugs. |
| 16:38:19 | Zevv | *so* *many* *bugs* |
| 16:38:25 | Zevv | and I love each and every one of them |
| 16:38:38 | Zevv | " |
| 16:38:43 | Zevv | Serious Syntax Problem Bug" |
| 16:38:48 | Zevv | who makes up such a thing |
| 16:38:55 | FromDiscord | <treeform> @dom96 thanks! |
| 16:39:07 | Zevv | "Temporary Topology Condition" |
| 16:39:19 | disruptek | i like this termtosvg but i dislike that it removes all my emojis. |
| 16:39:31 | Yardanico | that's good |
| 16:39:35 | Yardanico | you use too much emojis :P |
| 16:39:39 | Yardanico | in source code |
| 16:39:46 | disruptek | emojis are great. |
| 16:40:25 | FromDiscord | <treeform> @Yardanico commenting out `#computeCursors(owner, n, g.config)` did not work 😦 |
| 16:40:34 | Yardanico | well then it's not so simple :) |
| 16:40:51 | Yardanico | try before/after the optimizer PR, but really try to run git bisect |
| 16:40:56 | Yardanico | ./koch temp supports git bisect run btw |
| 16:41:34 | Yardanico | you have to make sure that when your stuff works it completes with 0 exit code though |
| 16:41:51 | Yardanico | git bisect start; git bisect good goodcommit; git bisect bad devel; git bisect run ./koch temp c --gc:arc blablabla file.nim |
| 16:42:06 | FromDiscord | <treeform> yeah that does not work for me |
| 16:42:09 | Yardanico | why? |
| 16:42:14 | FromDiscord | <treeform> dome DLL issue |
| 16:42:20 | Yardanico | ? |
| 16:44:02 | FromDiscord | <Varriount> I don't see how a DLL issue would happen, unless there's some commit that changes DLL loading behavior |
| 16:44:08 | FromDiscord | <treeform> https://gist.github.com/treeform/0c9c57acd2322a7003093d5c2332dec2 |
| 16:44:22 | FromDiscord | <treeform> ok "lib" issue |
| 16:44:41 | FromDiscord | <treeform> I think both koch and my thing wants to be in a spesific path |
| 16:44:44 | Yardanico | yeah |
| 16:44:51 | Yardanico | koch expects to run in nim repo as cwd I think :) |
| 16:45:05 | FromDiscord | <treeform> my thing expects to run in its own repo as cwd |
| 16:45:21 | FromDiscord | <treeform> I could fix my thing |
| 16:45:23 | FromDiscord | <treeform> fix on a fix |
| 16:45:30 | FromDiscord | <Varriount> Merge the directories? |
| 16:45:41 | FromDiscord | <kodkuce> @treeform ok np 🙂 |
| 16:45:55 | FromDiscord | <Varriount> If they don't share any file/subdirectory names |
| 16:46:02 | Yardanico | btw, anyone has experience with wikipedia table editing? |
| 16:46:21 | Yardanico | I want to fix the version table in https://en.wikipedia.org/wiki/Nim_(programming_language), add last 1.0.8 release and mark it as LTS |
| 16:46:33 | FromDiscord | <Varriount> Yardanico: A tiny amount. It's reminiscent of markdown tables |
| 16:47:02 | FromDiscord | <Varriount> There should be a reference on Wikipedia |
| 16:49:31 | FromDiscord | <treeform> ok I fixed my thing |
| 16:49:38 | FromDiscord | <treeform> it cd to my folder after starting |
| 16:50:03 | FromDiscord | <treeform> I should make my thing quit with quit(0) after N seconds? |
| 16:50:10 | FromDiscord | <treeform> to use git bisect |
| 16:50:18 | Yardanico | well yes |
| 16:50:26 | Yardanico | "Note that the script (my_script in the above example) should exit with code 0 if the current source code is good/old, and exit with a code between 1 and 127 (inclusive), except 125, if the current source code is bad/new." |
| 16:52:53 | Yardanico | https://git-scm.com/docs/git-bisect#_bisect_run |
| 16:53:10 | FromDiscord | <treeform> I think with the koch tmp trick I can manually bisect and get to ti faster |
| 16:53:23 | Yardanico | ?? |
| 16:53:24 | FromDiscord | <treeform> now that it does not take 10 minutes to switch to a commit |
| 16:53:34 | Yardanico | git bisect run is the same as manual |
| 16:53:36 | Yardanico | it's just more automated |
| 16:53:46 | Yardanico | btw I usually build old nim first |
| 16:53:47 | Yardanico | before bisecting |
| 16:53:50 | FromDiscord | <treeform> but I don't know when to quit |
| 16:54:00 | Yardanico | because otherwise you can get all sorts of weird bootstrapping issues |
| 16:54:07 | Yardanico | so I build csources and then start bisecting |
| 16:54:29 | FromDiscord | <treeform> Ok puzzle time, |
| 16:54:33 | FromDiscord | <treeform> 4301a3da9d333de88cf01cd952760716e8117bfa from choosnim works |
| 16:54:36 | FromDiscord | <treeform> 4301a3da9d333de88cf01cd952760716e8117bfa from github |
| 16:54:37 | FromDiscord | <treeform> does not |
| 16:54:47 | Yardanico | as in? |
| 16:55:00 | FromDiscord | <treeform> I get `C:\Users\me\nimcache\earthsea_r\stdlib_io.nim.cpp(237): error C2664: 'BOOL SetHandleInformation(HANDLE,DWORD,DWORD)': cannot convert argument 1 from 'int' to 'HANDLE'` now |
| 16:55:11 | Yardanico | yeah as I said |
| 16:55:20 | Yardanico | well first of all it might be a vcc bug from an older nim version lol |
| 16:55:26 | Yardanico | I only tested on linux :) |
| 16:55:38 | FromDiscord | <treeform> I think I need to take a break |
| 16:55:49 | FromDiscord | <treeform> Maybe I should just not use --gc arc |
| 16:55:53 | FromDiscord | <treeform> just stick with refc it works well |
| 16:56:02 | Yardanico | just wait until 1.4 :) |
| 16:56:10 | Yardanico | but it would be nice if you would help with bug reports |
| 16:56:23 | FromDiscord | <treeform> Simple example works |
| 16:56:24 | Yardanico | do you have the code open? |
| 16:56:27 | FromDiscord | <treeform> my complex thing does not |
| 16:56:28 | Yardanico | which triggers that bug |
| 16:56:32 | Yardanico | or is it closed |
| 16:56:37 | FromDiscord | <treeform> nope its for company |
| 16:56:40 | Yardanico | oh |
| 16:57:58 | FromDiscord | <treeform> be back later |
| 16:58:07 | FromDiscord | <treeform> with more confusion! |
| 16:59:22 | * | superbia2 joined #nim |
| 17:02:01 | * | superbia1 quit (Ping timeout: 264 seconds) |
| 17:02:57 | * | abm quit (Read error: Connection reset by peer) |
| 17:06:46 | ForumUpdaterBot | New thread by Adilh: How to get access to chunks of data downloaded from http server using httpclient, see https://forum.nim-lang.org/t/6641 |
| 17:21:18 | FromDiscord | <Varriount> @treeform Handle should be the size of a pointer. I think there was an incorrect wrapping in system that treated it as 32 bit for a while though |
| 17:21:58 | FromDiscord | <Varriount> I recall approving and merging a PR that fixed the issue in system.nim |
| 17:22:08 | Yardanico | yes, but he used git bisect |
| 17:22:11 | Yardanico | older commits |
| 17:22:50 | FromDiscord | <Recruit_main707> can you pass a for/while loop to a macro? (possibly as a pragma) |
| 17:23:02 | Yardanico | there are for loop macros as an experimental feature |
| 17:23:12 | Yardanico | allows a macro to fully rewrite the full loopo |
| 17:23:18 | Yardanico | I think they're in 1.2.x too, but not documented |
| 17:23:20 | Yardanico | documented in devel though |
| 17:23:29 | Yardanico | but if you want to pass the whole loop, you of course can |
| 17:23:33 | Yardanico | mymacro: for x in ...: stuff |
| 17:23:52 | FromDiscord | <Recruit_main707> not as pragma righ? |
| 17:23:55 | Yardanico | no |
| 17:24:39 | Yardanico | https://nim-lang.github.io/Nim/manual_experimental.html#case-statement-macros-for-loop-macros |
| 17:29:11 | FromDiscord | <Recruit_main707> i would have liked that, i guess it will do |
| 17:29:19 | Yardanico | it should work in latest 1.2.x too |
| 17:29:20 | Yardanico | I think |
| 17:29:26 | Yardanico | it just wasn't documented there |
| 17:35:26 | Zevv | hey disruptek |
| 17:36:40 | disruptek | sup |
| 17:36:53 | Zevv | was generics something that works yet? |
| 17:37:01 | Zevv | or was that just philosiphically speaking |
| 17:37:23 | disruptek | works somewhere, not sure it's in master. |
| 17:37:39 | Zevv | ah right |
| 17:37:53 | Zevv | was trying http://ix.io/2tkp |
| 17:37:53 | disruptek | what are you trying to do, generically speaking? |
| 17:38:10 | Zevv | that's the stash/iterator.nim, but then genericalized |
| 17:38:19 | disruptek | what happens? |
| 17:38:52 | Zevv | most of the stuff is ok, but in the end a compiler error about undeclared T |
| 17:39:00 | Zevv | at a line where it does not make sense to me yet |
| 17:39:10 | Zevv | at the `proc counter[T]` |
| 17:39:24 | Zevv | xform looks pretty much ok |
| 17:39:29 | disruptek | i guess this isn't in head? |
| 17:40:09 | Zevv | not yet, want me to check it in? |
| 17:40:30 | disruptek | i dl'd it. hmm, i don't think this could ever work. |
| 17:40:35 | Zevv | pushed, iteratorT.nim |
| 17:40:49 | Zevv | I think it's almost there |
| 17:40:54 | disruptek | use : Iterator[T] syntax. |
| 17:41:07 | disruptek | it cannot "solve" the sigmatch without it. |
| 17:41:25 | disruptek | ie. pragmas don't know about generics afaik. |
| 17:41:35 | Zevv | you'd be surprised |
| 17:41:40 | * | Vladar joined #nim |
| 17:41:43 | disruptek | i guess so. |
| 17:41:46 | Zevv | xfrm repr looks just fine, anyway |
| 17:42:53 | disruptek | it might /look/ fine. |
| 17:43:24 | disruptek | maybe this is a bug in cps that we can fix. |
| 17:45:31 | disruptek | are generics simple `bracketExpr`s? i don't even know. |
| 17:45:47 | disruptek | what does the chain demo produce? |
| 17:45:58 | Zevv | i don't know what that is |
| 17:46:06 | disruptek | it's in experiments. |
| 17:46:28 | Zevv | it says hello HELLO |
| 17:46:33 | disruptek | generics + concepts + experimental operators = chain cps |
| 17:46:46 | Zevv | oow |
| 17:47:02 | Zevv | right, that chain.nim |
| 17:47:08 | disruptek | it lets you mutate values with arbitrary fns. |
| 17:47:46 | Zevv | anyway, just trying stuff, nevermind me |
| 17:48:16 | disruptek | chain is something i hope you'll look at. i would ignore the rest of the codebase, honestly. |
| 17:48:45 | disruptek | var c: Foo; var upper = "hello" -> c -> toUpperAscii; assert upper() == "HELLO" |
| 17:48:58 | disruptek | it's kind of impressive when you think about it. |
| 17:56:59 | FromDiscord | <kodkuce> nim has no webrtc lib? |
| 17:57:28 | * | endragor quit (Remote host closed the connection) |
| 17:58:01 | * | endragor joined #nim |
| 17:59:34 | Zevv | I do like the expressiveness, but its a total different use case for cps |
| 18:00:25 | Zevv | its very lamda |
| 18:03:30 | * | endragor quit (Ping timeout: 256 seconds) |
| 18:04:56 | * | audiofile quit (Quit: Default Quit Message) |
| 18:07:14 | FromDiscord | <kodkuce> does nim have some kind of events, like in c# i can tell one obj to listen to some even that some other obj trigers, can subscribe and unscribe |
| 18:09:02 | FromDiscord | <exelotl> Not in the stdlib afaik, but its easy enough to roll your own |
| 18:09:23 | FromDiscord | <exelotl> just make a seq[proc()] |
| 18:09:53 | FromDiscord | <kodkuce> hmm will try |
| 18:10:29 | Zevv | easy to implement and works nicely with closures |
| 18:16:30 | euantor | There used to be an events library in the stdlib - you could take some inspiration from there: https://nim-lang.org/0.13.0/events.html |
| 18:16:58 | Yardanico | you can provide a newer "older" doc version :) |
| 18:16:58 | Yardanico | https://nim-lang.org/0.18.0/events.html |
| 18:17:18 | Yardanico | that module is still available |
| 18:17:27 | Yardanico | ah no nvm |
| 18:17:32 | Yardanico | I thought it's in https://github.com/nim-lang/graveyard but it's not |
| 18:23:27 | * | kungtotte quit (Read error: Connection reset by peer) |
| 18:24:27 | * | kungtotte joined #nim |
| 18:41:20 | disruptek | Zevv: did you figure anything out with generics? |
| 18:41:33 | * | sschwarzer joined #nim |
| 18:42:07 | Zevv | nope was out for a stroll in the park |
| 18:43:30 | sschwarzer | I have a question about Nim coding style. I have some `$` procs, as in https://hg.sr.ht/~sschwarzer/vppdiff/browse/src/vppdiff/umlmodel.nim?rev=tip#L41 , that meanwhile have additional parameters beyond the object to convert to a string. Is it common to have such "extended" `$` procs or should I rather rename these to something like `asString`? |
| 18:43:51 | Zevv | I do that, but haven't seen anyone else use it |
| 18:44:15 | sschwarzer | Zevv: do extended `$` or use `asString`? :) |
| 18:44:18 | Zevv | $ |
| 18:44:27 | Yardanico | sschwarzer: I think it's pretty rare |
| 18:44:38 | Yardanico | and $ is a conventional op used by procs like echo to convert arguments to stringsf |
| 18:44:43 | Yardanico | so I think it would be safer to rename it to something else |
| 18:45:13 | Zevv | yeah, but $ with additional args will not match for that |
| 18:45:22 | Yardanico | yes |
| 18:45:23 | Zevv | so you're safe to define those and use them, if you like the style |
| 18:45:45 | FromDiscord | <Rika> unless all additional args have default |
| 18:45:45 | sschwarzer | Yardanico: right, one consequence of the extra parameters is that I can no longer say `echo object`, but instead `echo `$`(object, extra_arg)` |
| 18:46:34 | Zevv | https://github.com/zevv/npeg/blob/master/src/npeg/common.nim#L265 |
| 18:46:37 | supakeen | I think that's enough reason to use `asString`. |
| 18:46:51 | Zevv | optional additional args is also handy sometimes |
| 18:47:00 | supakeen | Perhaps if they're optional. Yea, that. |
| 18:47:07 | Yardanico | Zevv: optional is okay |
| 18:47:12 | sschwarzer | My reasoning _for_ `$` is that it's the usual name of a proc that converts an object to a string. My reasoning _against_ `$` is that having additional parameters in it may be confusing. |
| 18:47:16 | Zevv | it's all ok. It's *my* code |
| 18:47:22 | Zevv | I call my procs whatever I like |
| 18:47:58 | supakeen | Yea that's valid, the question was more about how common it is and I wouldn't expect it and be confused initially until I look up the signatures. |
| 18:48:01 | Zevv | sschwarzer: well, the stropping makes it tedious also. `$`(thing, param) |
| 18:48:20 | sschwarzer | Zevv: yes, that's also one reason against :) |
| 18:48:35 | Zevv | I find none of those in the nim stdlib anyway |
| 18:48:45 | sschwarzer | Yardanico: I have one optional/default argument, but I think I need the other. |
| 18:49:27 | sschwarzer | Zevv: maybe because the objects in the stdlib are more "elementary" (half-serious) |
| 18:49:33 | disruptek | i wonder how useful an ast fuzzer could be. |
| 18:50:17 | Zevv | disruptek: not sure how you would do that. You mix it up, if it compiles you don't know what behavior to check for. If it doesn't compile it's probably because the AST is invalid |
| 18:50:33 | Zevv | so the only thing you could test for is ICEs |
| 18:50:56 | sschwarzer | The extra argument is a table that's used for lookups. I _could_ attach the table to the object to print, but this would require attaching the same table to every of these objects because the printing is a recursive algorithm (the object to print is a tree structure). |
| 18:51:06 | Zevv | I tried afl'ing nim source code once, but the amount of poop coming out quickly made me stop |
| 18:51:24 | disruptek | well, i would fuzz removals of ast. the idea is to reduce code. part of timmy's required for dustmite. |
| 18:51:26 | Yardanico | ask @varriount |
| 18:51:32 | Yardanico | I think he's trying to fuzz the nim compiler |
| 18:51:33 | sschwarzer | Zevv: What's "afl'ing"? |
| 18:51:41 | disruptek | i don't care about fuzzing code itself. |
| 18:51:47 | Yardanico | sschwarzer: https://github.com/google/AFL |
| 18:51:51 | Yardanico | https://en.wikipedia.org/wiki/American_fuzzy_lop_(fuzzer) |
| 18:52:09 | disruptek | i mean, i think fuzzy the compiler the way he is doing it is silly. |
| 18:52:20 | Zevv | disruptek: https://blog.regehr.org/archives/1678 |
| 18:53:07 | Zevv | there's also a part 2 there |
| 18:53:37 | sschwarzer | Yardanico: intriguing |
| 18:55:01 | sschwarzer | Regarding my original question, since there doesn't seem to be a "common" way to approach this, I'll probably go with `asString` (unless I get ideas to get rid of the arguments). |
| 18:56:05 | sschwarzer | Originally the table was module-global, but potentially I want to run the same code from multiple threads where each thread needs its own lookup table (since different trees will be processed). |
| 18:56:14 | sschwarzer | Thanks for your input. |
| 18:56:17 | sschwarzer | :-) |
| 18:56:29 | Yardanico | variables can be threadvar |
| 18:56:38 | Yardanico | which means that for each thread the variable will be "unique" |
| 18:56:49 | sschwarzer | Yardanico: right ... I'll think of it. |
| 18:57:07 | sschwarzer | Yardanico: actually I use `threadvar`s in another place in the program :) |
| 18:57:31 | sschwarzer | or one threadvar, depending on how you count them. ;-) |
| 18:58:06 | disruptek | Zevv: 50k loc to do ast transforms. |
| 18:58:08 | disruptek | holy shit. |
| 18:58:29 | sschwarzer | disruptek: uh, oh |
| 18:59:33 | Yardanico | yeah something like dustmite will be very useful |
| 19:01:21 | Zevv | disruptek: well, it's also half a c compiler right |
| 19:01:43 | disruptek | ofc. i'm just glad i don't have to do that. |
| 19:01:49 | * | tane joined #nim |
| 19:03:05 | disruptek | i'd be happy with reducing all c-codegen bugs, since those are relatively expensive. |
| 19:03:20 | disruptek | so basically, typed tests. |
| 19:05:27 | disruptek | what's interesting about typed input is that you can have an algorithm write equivalencies, optimizations, and tests. |
| 19:07:00 | disruptek | i mean, look at this issue i made. |
| 19:07:04 | disruptek | !issue author:disruptek |
| 19:07:05 | disbot | https://github.com/nim-lang/Nim/issues/15160 -- 3exceptions codegen broken inside macros 7& 27 more... |
| 19:07:32 | disruptek | seems like something a program could not just find, but add lots of test cases for. |
| 19:14:49 | * | sschwarzer quit (Quit: leaving) |
| 19:22:34 | * | endragor joined #nim |
| 19:27:05 | * | endragor quit (Ping timeout: 240 seconds) |
| 19:32:40 | FromDiscord | <kodkuce> hmm, i am totaly lost i dont even know anymore whats point of events if evry proc is basicly a static function and objects are just data storages |
| 19:32:51 | disruptek | there's no point. |
| 19:34:06 | FromDiscord | <kodkuce> i wanted to implement some tough it would help me clean up the code but now am bonkers i guess that heppends when you come form c# |
| 19:37:23 | FromDiscord | <kodkuce> hmm |
| 19:41:58 | * | narimiran quit (Ping timeout: 265 seconds) |
| 19:46:14 | * | vicfred_ joined #nim |
| 19:48:42 | * | vicfred quit (Ping timeout: 260 seconds) |
| 19:51:35 | FromDiscord | <kodkuce> i cant put in table key:string, value:seq[int] right? |
| 19:51:39 | Yardanico | yes |
| 19:58:19 | * | waleee-cl joined #nim |
| 19:58:29 | FromDiscord | <Varriount> disruptek: You mentioned fuzzing? |
| 19:58:36 | Yardanico | he did |
| 20:00:49 | * | rockcavera joined #nim |
| 20:04:23 | * | maier joined #nim |
| 20:09:27 | * | maier quit (Ping timeout: 256 seconds) |
| 20:10:42 | * | hyiltiz quit (Quit: hyiltiz) |
| 20:14:31 | * | Vladar quit (Quit: Leaving) |
| 20:16:03 | * | superbia2 is now known as superbia |
| 20:25:37 | Yardanico | AFL is a fun thing lol |
| 20:25:51 | Yardanico | It found over 33 examples of math expressions which hang my mathexpr lib |
| 20:25:56 | Yardanico | (I have some while true loops) |
| 20:26:00 | Yardanico | and it's easy to use it with nim |
| 20:26:13 | Yardanico | nim c --cc:clang --clang.exe="afl-clang" --clang.linkerexe="afl-clang" --opt:speed -o:here examples/fuzz.nim |
| 20:26:35 | Yardanico | I gave it 2+2*2 as the only input sample |
| 20:26:38 | Yardanico | https://i.imgur.com/QZkxvdM.png |
| 20:27:40 | FromDiscord | <Varriount> Yardanico: I used it on my AWS signing library, to try finding inputs that would break the signing functions (since they perform a lot of string manipulation) |
| 20:27:48 | Yardanico | nice |
| 20:27:54 | Yardanico | well hanging for my lib is really bad, I've gotta to fix these |
| 20:27:58 | * | hyiltiz joined #nim |
| 20:27:58 | * | hyiltiz quit (Changing host) |
| 20:27:58 | * | hyiltiz joined #nim |
| 20:28:33 | FromDiscord | <Varriount> Yardanico: What does your library do when given `1/0`? |
| 20:28:43 | Yardanico | inf |
| 20:28:45 | Yardanico | it's really basic |
| 20:28:52 | Yardanico | just some recursive-descent with in-tree evaluation |
| 20:28:53 | Yardanico | no AST |
| 20:30:54 | Zevv | Yardanico: show me your cases |
| 20:31:04 | Yardanico | "r(..E." |
| 20:31:06 | Yardanico | lol |
| 20:31:07 | Zevv | haha |
| 20:31:11 | Zevv | dude |
| 20:31:17 | Yardanico | the first one it found was "r(2E." |
| 20:31:29 | Yardanico | seems like some float parsing thing |
| 20:31:36 | Yardanico | oh yeah I think I know why LOL |
| 20:32:40 | Yardanico | I wasn't handling the case where parseFloat from parseutils returns 0 |
| 20:32:48 | Yardanico | fixed that |
| 20:32:49 | Yardanico | time to retest |
| 20:34:17 | Yardanico | I also expanded my samples, use most of the samples from my test suite |
| 20:34:23 | Yardanico | oh boy I guess I'll get tons of errors now |
| 20:34:58 | Zevv | lemme see what npeg makes of it :) |
| 20:35:03 | Yardanico | hohoho |
| 20:37:25 | Yardanico | the reason you compile with afl-gcc or afl-clang is in order to give AFL info about coverage |
| 20:37:32 | Yardanico | so it can fuzz to try to cover as much code as possible |
| 20:39:14 | * | NimBot joined #nim |
| 20:44:29 | Zevv | 0 crashes, 0 hangs |
| 20:44:39 | Yardanico | yeah after my fix I don't get any either |
| 20:44:42 | Zevv | :) |
| 20:45:00 | Yardanico | need to learn how to parallelize AFL though |
| 20:45:00 | Yardanico | I have 16 coures |
| 20:45:04 | Yardanico | cores |
| 20:45:14 | Zevv | yeah but that's kind of tricky iirc |
| 20:45:24 | Yardanico | apparently not *that* hard |
| 20:45:41 | Yardanico | https://afl-1.readthedocs.io/en/latest/user_guide.html#single-system-parallelization |
| 20:45:45 | Yardanico | you just start one master process |
| 20:45:47 | Yardanico | and secondary instances |
| 20:47:11 | Zevv | ooh I got a hang |
| 20:47:14 | Yardanico | hohoho |
| 20:47:21 | Yardanico | bet it's some random nonsense |
| 20:47:23 | Yardanico | care to share? |
| 20:47:28 | * | vicfred_ quit (Quit: Leaving) |
| 20:47:33 | Yardanico | and what exactly are you testing? |
| 20:47:38 | Yardanico | one of the parsers with npeg? |
| 20:47:52 | Zevv | arithmatic expression evaluator |
| 20:47:56 | Yardanico | oh |
| 20:49:01 | Zevv | 4^7^26 |
| 20:49:18 | Yardanico | 3.757668132438133e+109 |
| 20:49:24 | Zevv | you bet :) |
| 20:49:25 | Yardanico | tested with my mathexpr :) |
| 20:49:31 | Yardanico | maybe you don't support multiple ^s ? |
| 20:49:49 | Zevv | no i get an overflow but compiled nim with -d:danger so isn't handled |
| 20:49:52 | Zevv | just stupid ints |
| 20:49:59 | Yardanico | oh you're doing integers |
| 20:50:01 | Zevv | python doesn't chew it |
| 20:50:05 | Yardanico | my lib is more dumb |
| 20:50:07 | Yardanico | it's simple floats everywhere |
| 20:50:15 | Zevv | try 4**7**26 in python |
| 20:50:49 | Zevv | your answer is very wrong by the way |
| 20:50:54 | Zevv | do you have proper precedence for ^? |
| 20:51:06 | Yardanico | why wrong? |
| 20:51:32 | Yardanico | 2^3^4 is (2^3)^4 no ?\ |
| 20:51:32 | Zevv | it's not (4^7)^26 |
| 20:51:35 | Yardanico | I do it the same as https://github.com/codeplea/tinyexpr#compile-time-options |
| 20:51:37 | Zevv | but 4^(7^26) |
| 20:51:39 | * | snowolf quit (Remote host closed the connection) |
| 20:51:39 | * | snowolf joined #nim |
| 20:51:40 | * | snowolf quit (Changing host) |
| 20:51:40 | * | snowolf joined #nim |
| 20:51:41 | Zevv | ^ is wrong associative |
| 20:51:41 | Yardanico | "By default, TinyExpr does exponentiation from left to right. F" |
| 20:51:57 | Zevv | that's just wrong :) |
| 20:52:00 | Yardanico | I can add a option to switch it line tinyexpr does |
| 20:52:06 | Yardanico | Zevv: "This is by design. It's the way that spreadsheets do it (e.g. Excel, Google Sheets)." |
| 20:52:12 | Zevv | that's just wrong |
| 20:52:21 | Yardanico | whatever :P |
| 20:52:24 | Zevv | hehe |
| 20:52:37 | Zevv | it doesn't math |
| 20:55:55 | Yardanico | https://i.imgur.com/8ThVjIO.png with one master and 3 instances |
| 20:56:04 | Yardanico | they sync from time to time |
| 20:56:54 | Yardanico | wait I wonder what'll happen if I use --debugger:native |
| 21:01:16 | * | vicfred joined #nim |
| 21:06:57 | Yardanico | wait |
| 21:06:59 | Yardanico | it still found one hang |
| 21:07:18 | Yardanico | 3 already |
| 21:12:54 | * | Zectbumo joined #nim |
| 21:21:49 | * | solitudesf quit (Ping timeout: 264 seconds) |
| 21:48:11 | FromDiscord | <dom96> what you fuzzing |
| 21:48:39 | Yardanico | just my smol mathexpr lib |
| 21:48:44 | Yardanico | found one stupid bug :) |
| 21:48:47 | Yardanico | thanks to fuzzing |
| 21:57:12 | * | tane quit (Quit: Leaving) |
| 22:05:20 | * | maier joined #nim |
| 22:10:05 | * | maier quit (Ping timeout: 240 seconds) |
| 22:16:04 | disruptek | here's a great one... when false: if not n.isNil: t.renderTrace(n) -- i have to leave this as `when true:` or i get a codegen error. |
| 22:16:12 | disruptek | n is an exception symbol. |
| 22:16:21 | disruptek | bananas. |
| 22:18:28 | * | vicfred quit (Quit: Leaving) |
| 22:37:32 | FromDiscord | <Varriount> Is news.ycombinator.com down for anyone? Or have I been classified as a robot. |
| 22:37:40 | disruptek | i knew it. |
| 22:37:51 | federico3 | it's up |
| 22:38:24 | FromDiscord | <Rika> varriount is a robot! |
| 22:38:59 | FromDiscord | <Varriount> Oh dear, I've been found out. |
| 22:57:35 | disruptek | is someone else going to build dustmite? |
| 23:00:14 | * | arecacea1 quit (Remote host closed the connection) |
| 23:05:49 | * | arecacea1 joined #nim |
| 23:10:15 | Yardanico | hm, I forgot, is there a good way to importc a C macro? |
| 23:10:40 | Yardanico | e.g. if in C it's used as while(MACROCALL(ARG)){} |
| 23:11:03 | Yardanico | for __AFL_LOOP |
| 23:12:15 | * | suchasurge quit (Quit: Ping timeout (120 seconds)) |
| 23:12:35 | * | suchasurge joined #nim |
| 23:15:23 | disruptek | what's the current idiom to find out if i'm compiling to c++? |
| 23:16:31 | disruptek | i guess compilesettings is the new hotness? |
| 23:22:49 | federico3 | a better c2nim would be welcome |
| 23:23:27 | disruptek | nimterop is that beast. |
| 23:46:03 | * | D_ quit (Ping timeout: 260 seconds) |
| 23:48:07 | * | D_ joined #nim |
| 23:52:08 | * | sekao joined #nim |
| 23:52:10 | FromGitter | <bung87> ```code paste, see link``` |
| 23:52:15 | Yardanico | ? |
| 23:52:23 | FromDiscord | <Elegant Beef> Is bung trolling? |
| 23:52:28 | Yardanico | no |
| 23:52:33 | Yardanico | seems like the FromGitter bot is broken |
| 23:52:44 | FromGitter | <bung87> what ? |
| 23:52:52 | FromDiscord | <Elegant Beef> We didnt get a link from that |
| 23:53:02 | Yardanico | @bung87 https://i.imgur.com/FO1MUU0.png |
| 23:53:06 | Yardanico | but he sent https://i.imgur.com/wcYgk2b.png |
| 23:53:11 | FromGitter | <bung87> oh, let me log in irc |
| 23:53:17 | sekao | is the compiler api in stable nim yet? when i do `import compiler/nimeval` i get `cannot open file: compiler/nimeval` |
| 23:53:18 | Yardanico | @bung87 that means it's already downloaded |
| 23:53:23 | Yardanico | so I don't think you need to handle it |
| 23:53:29 | * | krux02_ joined #nim |
| 23:53:48 | FromDiscord | <Elegant Beef> did you `nimble install compiler` sekao? |
| 23:53:58 | FromGitter | <bung87> but it stop with ```Error: Unable to extract. Error was 'Cannot create a file when that file already exists.``` |
| 23:54:18 | FromDiscord | <flywind> https://github.com/dom96/choosenim/pull/201 |
| 23:54:19 | disbot | ➥ Fix #199 - missing dlls |
| 23:54:20 | * | bung joined #nim |
| 23:54:29 | sekao | oh no i didn't, so it's a dependency in nimble! ok ill try that |
| 23:54:40 | Yardanico | sekao: well it's not really a dependency in nimble |
| 23:54:54 | Yardanico | it's just that if you install nim from e.g. your distro, you won't have the source code for the compiler |
| 23:55:35 | * | blackbeard420 quit (Quit: ZNC 1.8.0 - https://znc.in) |
| 23:55:53 | * | blackbeard420 joined #nim |
| 23:56:00 | sekao | just installed it but the error still happens. i'm on windows btw. |
| 23:56:08 | Yardanico | how are you trying to run? |
| 23:56:09 | Yardanico | your app |
| 23:56:13 | * | krux02 quit (Ping timeout: 244 seconds) |
| 23:56:28 | sekao | i use nimble. so nimble run appname |
| 23:56:37 | Yardanico | then you need to specify compiler in nimble requirements as well |
| 23:56:51 | Yardanico | because nimble will only expose libraries which are specified in "requirements" |
| 23:56:55 | FromGitter | <bung87> ok let me run on devel |
| 23:56:55 | Yardanico | requires "compiler" |
| 23:57:31 | sekao | ah that makes sense. thank you, it works now |
