| 00:07:17 | * | johnsoft quit (Ping timeout: 248 seconds) |
| 00:08:43 | * | hellonico joined #nim |
| 00:12:07 | * | johnsoft joined #nim |
| 00:15:38 | * | irrequietus quit () |
| 00:22:37 | reactormonk | yup, shared state doesn't seem to be a problem. I'm only running one thread and it still segfaults. |
| 00:24:28 | * | gsingh93 quit (Ping timeout: 244 seconds) |
| 00:27:01 | * | johnsoft quit (Ping timeout: 248 seconds) |
| 00:31:21 | * | a5i joined #nim |
| 00:32:14 | Araq | reactormonk: well a failing test is still a test you can submit |
| 00:32:18 | * | johnsoft joined #nim |
| 00:38:09 | reactormonk | Araq, I'll reduce it to a simpler testcase and submit as a separate issue |
| 00:38:18 | * | tstm quit (Ping timeout: 272 seconds) |
| 00:38:19 | Araq | sure |
| 00:38:35 | * | tstm joined #nim |
| 00:39:10 | * | gsingh93 joined #nim |
| 00:40:47 | flaviu | Any chance that the test suite will get faster? |
| 00:40:55 | flaviu | It isn |
| 00:41:12 | flaviu | 't useful if no one runs it, and people tend to avoid running it as it takes 1.5 hours |
| 00:41:45 | reactormonk | flaviu, go benchmark and make nice graphs :-) |
| 00:42:02 | def- | flaviu: it takes 1.5 hours? |
| 00:42:25 | flaviu | reactormonk: I'm about 5% of the way there on my add-test-metadata, but I've been working on other things lately. |
| 00:42:41 | flaviu | def-: http://buildbot.nim-lang.org/builders/linux-x32-builder/builds/322/steps/Run%20Testament |
| 00:42:53 | reactormonk | flaviu, just add it in the ugly way I'd say |
| 00:42:55 | flaviu | And something like 6 hours on the slower ARM boards. |
| 00:43:07 | def- | flaviu: Does that run with a release compiler or debug? |
| 00:43:40 | reactormonk | tests should run in release imo |
| 00:43:49 | Araq | flaviu: it takes 45 minutes for me |
| 00:43:50 | reactormonk | although debug would make kinda more sense... |
| 00:43:58 | Araq | and I do run it from time to time |
| 00:44:11 | reactormonk | Araq, do the tests run in debug or release? |
| 00:44:18 | flaviu | Araq: The correct approach would be to never make a commit without running the test suite. |
| 00:44:23 | Araq | the release version of the compiler |
| 00:44:44 | Araq | flaviu: that's not even worth replying to. |
| 00:44:56 | * | Menche quit (Quit: Leaving) |
| 00:45:13 | Araq | tests suites grow to the point where your suggestion is completely unrealistic |
| 00:45:44 | flaviu | At which time those test suites need to be optimized. |
| 00:45:50 | * | Menche joined #nim |
| 00:46:12 | * | hellonico quit (Quit: Leaving...) |
| 00:46:30 | Araq | you cannot optimize them to take as little time as test driven development demands |
| 00:46:30 | reactormonk | Araq, yes, the release version of the compiler - but are the tests which are run also compiled in release mode? |
| 00:46:41 | Araq | reactormonk: no. |
| 00:46:42 | flaviu | Even if once per commit is unrealistic, test suite run time should be less than 5minutes, max. |
| 00:47:14 | Araq | so name a production compiler which has a test suite that only runs for 5 minutes. |
| 00:47:23 | def- | I think go does |
| 00:47:25 | Araq | you have no idea what you're talking about. |
| 00:47:37 | * | johnsoft quit (Ping timeout: 250 seconds) |
| 00:48:11 | Araq | a proper full test suite can run for months before a release. |
| 00:49:03 | reactormonk | def-, lemme run |
| 00:49:54 | * | Menche quit (Client Quit) |
| 00:50:00 | Araq | flaviu: we have instead tests that are in groups. touched the GC? --> run the GC tests |
| 00:50:19 | Araq | touched template handling? --> run the templates tests |
| 00:50:30 | def- | Go even runs all tests when you build the compiler |
| 00:50:38 | cazov | there's a difference between unit tests and ftests. i'm not sure i've seen any sizeable project with ftests that are optimized for speed |
| 00:50:57 | flaviu | Araq: I'm not familiar with production compilers, but looks like rust and crystal have test suites shorter than 5 min. |
| 00:51:14 | Araq | def-: then Go either completely lacks long running tests or it doesn't run them. |
| 00:51:32 | def- | Araq: they don't seem to have many tests indeed, takes just 1 minute to run them |
| 00:51:43 | Araq | flaviu: rust doesn't even bootstrap in 5 minutes |
| 00:52:05 | * | johnsoft joined #nim |
| 00:53:42 | def- | wait, I'm wrong, go has some more tests |
| 00:54:13 | reactormonk | def-, can't bootstrap go from the go installed via package manager :-/ |
| 00:54:29 | flaviu | Araq: Good point, looks like they only run lint here: https://travis-ci.org/rust-lang/rust/builds/57285839 |
| 00:54:31 | def- | reactormonk: i just built it from source with ./all.bash |
| 00:54:49 | reactormonk | def-, yup, got Set $GOROOT_BOOTSTRAP to a working Go tree >= Go 1.4 |
| 00:55:31 | reactormonk | and if I set that one to /usr it starts complaining |
| 00:56:21 | * | HakanD joined #nim |
| 00:56:26 | def- | Maybe I'm doing something wrong, but for me the Nim tests run in 10 minutes |
| 00:56:33 | flaviu | But it should be possible to run compiler tests very rapidly if techniques like those used in AFL are used. |
| 00:57:06 | reactormonk | looks like GOROOT_BOOTSTRAP=/usr/lib/go is the correct one |
| 00:57:19 | reactormonk | flaviu, which techniques? |
| 00:58:08 | reactormonk | def-, looks reasonable |
| 00:58:11 | EXetoC | you could parallelize across machines |
| 00:58:23 | reactormonk | EXetoC, not very useful |
| 00:59:04 | reactormonk | how do I test for windows/not windows in the tester? |
| 01:00:05 | Araq | 5 minutes is no meaningful number anyway. A 5 minutes pause means I start to do something else. might as well take 10-15 minutes then. 5 minutes is a totally arbitrary number which has no psychological foundation. |
| 01:00:14 | flaviu | reactormonk: http://lcamtuf.blogspot.com/2014/10/fuzzing-binaries-without-execve.html |
| 01:00:42 | * | HakanD quit (Ping timeout: 252 seconds) |
| 01:01:09 | flaviu | Araq: Of course it's totally arbitrary, but it's also short enough that it's practical to run the test suite frequently. |
| 01:01:36 | EXetoC | reactormonk: why not? is it still not parallelized across cores though? |
| 01:01:52 | Araq | it's not short enough for development. |
| 01:01:53 | reactormonk | EXetoC, that would be more useful. the average developer doesn't have multiple machines. |
| 01:02:32 | reactormonk | flaviu, you could make a repl compiler with that idea and just feed the tests in |
| 01:03:32 | reactormonk | well more like fork - read - compile - print - die |
| 01:04:04 | def- | Maybe we could just speed up the Nim compiler in general? |
| 01:04:09 | flaviu | Well, it'd first be useful to instrument and see exactly how slow initialization is, but yeah. |
| 01:04:11 | reactormonk | def-, I don't think so. |
| 01:04:18 | reactormonk | flaviu, probably not very. |
| 01:04:24 | reactormonk | I mean the compiler startup is probably 0.5s |
| 01:04:41 | flaviu | That is quite slow. |
| 01:05:10 | flaviu | On AFL timescales, 500 invocations/second is pretty terrible. |
| 01:05:22 | reactormonk | they're doing fuzzying |
| 01:05:45 | flaviu | And what is a test suite except fuzzing with known input? :) |
| 01:05:58 | renesac | one approach: benchmark each test separatedly and record how much time it takes |
| 01:06:23 | def- | I guess the tests run in < 10 min for me because I'm using this: https://github.com/Araq/Nim/pull/2454 |
| 01:06:43 | renesac | then create a script that runs the tests for a given fixed amount of time given as a parameter |
| 01:06:46 | def- | But that's only if you had them compiled before |
| 01:06:51 | renesac | running it from the faster to the slower tests |
| 01:07:02 | * | johnsoft quit (Ping timeout: 246 seconds) |
| 01:07:12 | renesac | of course, some important test may be left because it is too slow |
| 01:07:33 | reactormonk | in the tester, how can I make a new tests that's basically "run this executable and go by the exit code" |
| 01:07:36 | renesac | or it is in a file with a lot other important tests |
| 01:07:44 | renesac | and the benchmark is per file |
| 01:10:40 | reactormonk | ... how do I tests a C file? |
| 01:10:48 | reactormonk | with testament |
| 01:12:01 | * | johnsoft joined #nim |
| 01:27:06 | * | johnsoft quit (Ping timeout: 252 seconds) |
| 01:30:13 | * | EXetoC quit (Ping timeout: 265 seconds) |
| 01:32:02 | * | johnsoft joined #nim |
| 01:34:43 | Varriount | reactormonk: Uh, can you give some info? |
| 01:38:19 | reactormonk | Varriount, currently implemeting a proc to run C files as test. |
| 01:46:56 | * | johnsoft quit (Ping timeout: 246 seconds) |
| 01:51:43 | * | brson joined #nim |
| 01:52:02 | * | BitPuffin quit (Ping timeout: 252 seconds) |
| 01:52:11 | * | johnsoft joined #nim |
| 02:02:30 | * | saml_ quit (Quit: Leaving) |
| 02:03:10 | * | untitaker quit (Ping timeout: 272 seconds) |
| 02:04:41 | * | brson quit (Quit: leaving) |
| 02:05:09 | * | brson joined #nim |
| 02:07:14 | * | johnsoft quit (Ping timeout: 246 seconds) |
| 02:09:19 | * | untitaker joined #nim |
| 02:12:11 | * | johnsoft joined #nim |
| 02:19:55 | * | Menche joined #nim |
| 02:22:58 | * | Senketsu quit (Read error: Connection reset by peer) |
| 02:27:01 | * | johnsoft quit (Ping timeout: 248 seconds) |
| 02:28:12 | * | darkf joined #nim |
| 02:32:02 | * | johnsoft joined #nim |
| 02:46:54 | * | johnsoft quit (Ping timeout: 245 seconds) |
| 02:52:07 | * | johnsoft joined #nim |
| 03:00:51 | * | endragor joined #nim |
| 03:00:53 | * | endragor quit (Remote host closed the connection) |
| 03:01:20 | * | endragor joined #nim |
| 03:07:50 | * | johnsoft quit (Ping timeout: 250 seconds) |
| 03:12:03 | * | johnsoft joined #nim |
| 03:13:14 | * | brson quit (Quit: leaving) |
| 03:13:28 | * | brson joined #nim |
| 03:13:36 | * | brson quit (Client Quit) |
| 03:13:54 | * | brson joined #nim |
| 03:14:13 | * | dashed joined #nim |
| 03:24:18 | * | brson quit (Quit: Lost terminal) |
| 03:24:37 | * | brson joined #nim |
| 03:25:06 | * | MagusOTB joined #nim |
| 03:27:25 | * | johnsoft quit (Ping timeout: 264 seconds) |
| 03:28:20 | MagusOTB | is openarray a thing that can actually be instantiated, or an abstraction? |
| 03:29:17 | MagusOTB | I have a list of things that I don't know the length of until runtime, but when I do, it'll be allocated once to the correct size, and I'm not sure what I should use. |
| 03:30:22 | * | brson quit (Quit: leaving) |
| 03:30:57 | * | brson joined #nim |
| 03:32:02 | * | johnsoft joined #nim |
| 03:47:01 | * | johnsoft quit (Ping timeout: 252 seconds) |
| 03:48:54 | flaviu | MagusOTB: openarray isn't a concrete type |
| 03:48:59 | flaviu | you're looking for seq. |
| 03:50:29 | flaviu | I think that either newSeq(seqLen) or mySeq.setLen(seqLen) can be used to preallocate space, but it's late and I need to sleep. |
| 03:52:06 | * | johnsoft joined #nim |
| 03:55:28 | * | Menche quit (Read error: Connection reset by peer) |
| 03:56:00 | * | Menche joined #nim |
| 04:07:47 | * | johnsoft quit (Ping timeout: 265 seconds) |
| 04:12:19 | * | johnsoft joined #nim |
| 04:27:36 | * | johnsoft quit (Ping timeout: 256 seconds) |
| 04:32:03 | * | johnsoft joined #nim |
| 04:48:01 | * | johnsoft quit (Ping timeout: 264 seconds) |
| 04:52:09 | * | johnsoft joined #nim |
| 05:07:37 | * | johnsoft quit (Ping timeout: 264 seconds) |
| 05:12:03 | * | johnsoft joined #nim |
| 05:27:16 | * | johnsoft quit (Ping timeout: 255 seconds) |
| 05:32:02 | * | johnsoft joined #nim |
| 05:36:42 | * | brson quit (Quit: leaving) |
| 05:39:44 | * | ChrisMAN quit (Ping timeout: 252 seconds) |
| 05:40:55 | reactormonk | MagusOTB, openarray is just the nim implementation of varargs |
| 05:42:19 | renesac | reactormonk: nim has varargs too |
| 05:42:28 | reactormonk | duh |
| 05:43:00 | renesac | open array is more like a type you use in function parameters when you expect either seq or array |
| 05:43:10 | renesac | *is a type |
| 05:43:16 | renesac | more like a concept |
| 05:43:46 | renesac | good night |
| 05:47:29 | * | johnsoft quit (Ping timeout: 250 seconds) |
| 05:52:16 | * | johnsoft joined #nim |
| 06:03:57 | * | gsingh93 quit (Ping timeout: 250 seconds) |
| 06:08:34 | * | johnsoft quit (Ping timeout: 244 seconds) |
| 06:12:10 | * | johnsoft joined #nim |
| 06:27:19 | * | johnsoft quit (Ping timeout: 245 seconds) |
| 06:32:16 | * | johnsoft joined #nim |
| 06:46:51 | * | HakanD joined #nim |
| 06:47:22 | * | johnsoft quit (Ping timeout: 255 seconds) |
| 06:48:11 | * | HakanD_ joined #nim |
| 06:50:48 | * | HakanD__ joined #nim |
| 06:52:04 | * | johnsoft joined #nim |
| 06:52:19 | * | HakanD quit (Ping timeout: 255 seconds) |
| 06:53:41 | * | HakanD_ quit (Ping timeout: 248 seconds) |
| 07:06:14 | * | endragor_ joined #nim |
| 07:07:02 | * | johnsoft quit (Ping timeout: 250 seconds) |
| 07:08:35 | * | endragor quit (Ping timeout: 246 seconds) |
| 07:12:08 | * | johnsoft joined #nim |
| 07:14:59 | * | endragor_ quit (Remote host closed the connection) |
| 07:22:26 | * | wb joined #nim |
| 07:24:18 | * | dashed quit (Quit: Connection closed for inactivity) |
| 07:27:24 | * | johnsoft quit (Ping timeout: 252 seconds) |
| 07:27:38 | * | EXetoC joined #nim |
| 07:32:11 | * | johnsoft joined #nim |
| 07:36:38 | * | repax joined #nim |
| 07:47:04 | * | johnsoft quit (Ping timeout: 256 seconds) |
| 07:47:38 | * | BlaXpirit joined #nim |
| 07:49:38 | * | Ven joined #nim |
| 07:52:01 | * | johnsoft joined #nim |
| 08:07:23 | * | johnsoft quit (Ping timeout: 246 seconds) |
| 08:12:03 | * | johnsoft joined #nim |
| 08:13:01 | * | endragor joined #nim |
| 08:13:03 | * | endragor quit (Remote host closed the connection) |
| 08:13:32 | * | endragor joined #nim |
| 08:15:36 | * | MyMind joined #nim |
| 08:27:31 | * | johnsoft quit (Ping timeout: 252 seconds) |
| 08:32:20 | * | johnsoft joined #nim |
| 08:32:47 | * | endragor quit () |
| 08:32:52 | * | Ven quit (Quit: My MacBook has gone to sleep. ZZZzzz…) |
| 08:32:58 | * | endragor joined #nim |
| 08:47:17 | * | johnsoft quit (Ping timeout: 248 seconds) |
| 08:52:05 | * | johnsoft joined #nim |
| 08:56:18 | * | Ven joined #nim |
| 08:57:04 | * | eigenlicht joined #nim |
| 09:04:02 | * | milosn quit (Ping timeout: 250 seconds) |
| 09:06:49 | * | johnsoft quit (Ping timeout: 244 seconds) |
| 09:12:06 | * | johnsoft joined #nim |
| 09:15:45 | * | HakanD__ quit (Read error: Connection reset by peer) |
| 09:19:18 | * | HakanD__ joined #nim |
| 09:27:29 | * | johnsoft quit (Ping timeout: 244 seconds) |
| 09:32:01 | * | johnsoft joined #nim |
| 09:44:30 | * | Ven quit (Quit: My MacBook has gone to sleep. ZZZzzz…) |
| 09:46:50 | * | dtscode is now known as fbi-bot |
| 09:47:22 | * | Ven joined #nim |
| 09:47:27 | * | johnsoft quit (Ping timeout: 252 seconds) |
| 09:48:37 | * | fbi-bot is now known as charmander |
| 09:52:04 | * | johnsoft joined #nim |
| 09:54:29 | * | filwit joined #nim |
| 09:54:43 | * | HakanD__ quit (Read error: Connection reset by peer) |
| 09:55:44 | * | Matthias247 joined #nim |
| 09:57:02 | * | Ven quit (Quit: My MacBook has gone to sleep. ZZZzzz…) |
| 10:00:00 | * | BitPuffin joined #nim |
| 10:01:34 | * | Ven joined #nim |
| 10:07:17 | * | johnsoft quit (Ping timeout: 248 seconds) |
| 10:10:42 | * | repax quit (Quit: repax) |
| 10:11:30 | * | HakanD joined #nim |
| 10:12:01 | * | johnsoft joined #nim |
| 10:15:37 | * | charmander is now known as mewtwo |
| 10:25:29 | * | milosn joined #nim |
| 10:28:09 | * | johnsoft quit (Ping timeout: 256 seconds) |
| 10:32:06 | * | milosn quit (Read error: Connection reset by peer) |
| 10:32:11 | * | johnsoft joined #nim |
| 10:32:47 | * | milosn joined #nim |
| 10:34:09 | * | milosn quit (Read error: Connection reset by peer) |
| 10:35:16 | * | milosn joined #nim |
| 10:36:35 | * | Trustable joined #nim |
| 10:42:45 | * | milosn quit (Read error: Connection reset by peer) |
| 10:43:40 | * | milosn joined #nim |
| 10:47:34 | * | johnsoft quit (Ping timeout: 244 seconds) |
| 10:52:04 | * | johnsoft joined #nim |
| 10:53:58 | * | milosn quit (Ping timeout: 252 seconds) |
| 10:58:19 | * | Senketsu joined #nim |
| 10:59:26 | * | milosn joined #nim |
| 11:00:04 | * | HakanD quit (Ping timeout: 265 seconds) |
| 11:01:19 | * | HakanD joined #nim |
| 11:07:38 | * | jefus__ joined #nim |
| 11:07:38 | * | milosn quit (Read error: Connection reset by peer) |
| 11:07:50 | * | johnsoft quit (Ping timeout: 272 seconds) |
| 11:07:57 | * | milosn joined #nim |
| 11:08:48 | * | milosn quit (Read error: Connection reset by peer) |
| 11:12:01 | * | jefus_ quit (Ping timeout: 264 seconds) |
| 11:12:09 | * | johnsoft joined #nim |
| 11:12:56 | * | milosn joined #nim |
| 11:13:37 | * | TEttinger quit (Ping timeout: 252 seconds) |
| 11:19:47 | * | milosn quit (Read error: Connection reset by peer) |
| 11:20:16 | * | milosn joined #nim |
| 11:22:32 | * | Varriount quit (Quit: Leaving) |
| 11:26:56 | * | Ven quit (Quit: My MacBook has gone to sleep. ZZZzzz…) |
| 11:27:37 | * | johnsoft quit (Ping timeout: 264 seconds) |
| 11:28:31 | * | Ven joined #nim |
| 11:28:31 | * | HakanD quit (Read error: Connection reset by peer) |
| 11:28:35 | * | HakanD_ joined #nim |
| 11:31:53 | * | Pisuke joined #nim |
| 11:32:04 | * | johnsoft joined #nim |
| 11:34:37 | * | MyMind quit (Ping timeout: 264 seconds) |
| 11:38:20 | * | milosn quit (Read error: Connection reset by peer) |
| 11:38:48 | * | milosn joined #nim |
| 11:44:32 | * | BitPuffin quit (Ping timeout: 265 seconds) |
| 11:46:59 | * | johnsoft quit (Ping timeout: 244 seconds) |
| 11:50:54 | * | milosn quit (Ping timeout: 272 seconds) |
| 11:52:05 | * | johnsoft joined #nim |
| 11:56:31 | * | mewtwo is now known as dtscode |
| 12:07:20 | * | johnsoft quit (Ping timeout: 250 seconds) |
| 12:10:52 | * | milosn joined #nim |
| 12:12:05 | * | johnsoft joined #nim |
| 12:13:03 | * | Menche quit (Ping timeout: 265 seconds) |
| 12:14:43 | * | BlaXpirit_ joined #nim |
| 12:16:58 | * | Menche joined #nim |
| 12:18:01 | * | BlaXpirit quit (Ping timeout: 264 seconds) |
| 12:27:19 | * | johnsoft quit (Ping timeout: 252 seconds) |
| 12:32:13 | * | johnsoft joined #nim |
| 12:35:05 | * | jefus__ is now known as jefus |
| 12:35:05 | * | milosn quit (Read error: Connection reset by peer) |
| 12:35:30 | * | milosn joined #nim |
| 12:40:43 | * | Kingsquee quit (Quit: Konversation terminated!) |
| 12:47:12 | * | johnsoft quit (Ping timeout: 250 seconds) |
| 12:52:09 | * | johnsoft joined #nim |
| 12:52:20 | * | milosn quit (Read error: Connection reset by peer) |
| 12:53:33 | * | milosn joined #nim |
| 12:56:30 | * | mpthrapp joined #nim |
| 12:59:57 | * | arnetheduck quit (Read error: Connection reset by peer) |
| 13:07:47 | * | irrequietus joined #nim |
| 13:08:31 | * | johnsoft quit (Ping timeout: 256 seconds) |
| 13:12:03 | * | johnsoft joined #nim |
| 13:12:21 | * | milosn quit (Ping timeout: 248 seconds) |
| 13:14:30 | * | endragor quit (Ping timeout: 272 seconds) |
| 13:15:41 | federico3 | http://nim-lang.org/talk01/slides.html |
| 13:15:52 | federico3 | very nice examples |
| 13:19:09 | * | Trustable quit (Remote host closed the connection) |
| 13:20:53 | * | milosn joined #nim |
| 13:25:22 | * | arnetheduck joined #nim |
| 13:26:06 | * | endragor joined #nim |
| 13:28:26 | * | johnsoft quit (Ping timeout: 272 seconds) |
| 13:30:31 | * | rational joined #nim |
| 13:32:02 | * | johnsoft joined #nim |
| 13:32:58 | * | endragor quit (Remote host closed the connection) |
| 13:38:00 | pigmej | when a proc returns an object, what's the initial value of result? |
| 13:45:35 | def- | pigmej: binary 0 |
| 13:47:25 | * | endragor joined #nim |
| 13:53:13 | * | johnsoft quit (*.net *.split) |
| 13:53:13 | * | Menche quit (*.net *.split) |
| 13:53:13 | * | pipeep quit (*.net *.split) |
| 13:53:13 | * | k1i quit (*.net *.split) |
| 13:53:13 | * | Matthias247 quit (*.net *.split) |
| 13:53:19 | * | betawaffle quit (*.net *.split) |
| 13:53:25 | * | ntjnsz quit (*.net *.split) |
| 13:53:25 | * | mahlon quit (*.net *.split) |
| 13:53:25 | * | MattAitchison quit (*.net *.split) |
| 13:53:25 | * | filwit quit (*.net *.split) |
| 13:53:25 | * | ekarlso quit (*.net *.split) |
| 13:53:25 | * | dhasenan quit (*.net *.split) |
| 13:53:25 | * | mnemonikk quit (*.net *.split) |
| 13:53:25 | * | jez0990 quit (*.net *.split) |
| 13:53:25 | * | gmpreussner|work quit (*.net *.split) |
| 13:53:25 | * | fowl quit (*.net *.split) |
| 13:53:25 | * | nulpunkt quit (*.net *.split) |
| 13:53:25 | * | jacksnipe quit (*.net *.split) |
| 13:53:25 | * | elbow quit (*.net *.split) |
| 13:53:25 | * | pigmej quit (*.net *.split) |
| 13:53:25 | * | onionhammer quit (*.net *.split) |
| 13:53:25 | * | mtj quit (*.net *.split) |
| 13:53:25 | * | mrkishi quit (*.net *.split) |
| 13:53:25 | * | rinukkusu quit (*.net *.split) |
| 13:53:25 | * | adam12 quit (*.net *.split) |
| 13:53:25 | * | arnetheduck quit (*.net *.split) |
| 13:53:25 | * | Ven quit (*.net *.split) |
| 13:53:25 | * | EXetoC quit (*.net *.split) |
| 13:53:25 | * | MagusOTB quit (*.net *.split) |
| 13:53:25 | * | walt quit (*.net *.split) |
| 13:53:25 | * | vinnie quit (*.net *.split) |
| 13:53:25 | * | dhasenan_ quit (*.net *.split) |
| 13:53:25 | * | skroll1 quit (*.net *.split) |
| 13:53:25 | * | qfire quit (*.net *.split) |
| 13:53:25 | * | jokra quit (*.net *.split) |
| 13:53:59 | * | johnsoft joined #nim |
| 13:55:22 | * | filwit joined #nim |
| 13:55:22 | * | ekarlso joined #nim |
| 13:55:22 | * | dhasenan joined #nim |
| 13:55:22 | * | mnemonikk joined #nim |
| 13:55:22 | * | jez0990 joined #nim |
| 13:56:41 | * | gmpreussner|work joined #nim |
| 13:56:41 | * | fowl joined #nim |
| 13:56:41 | * | nulpunkt joined #nim |
| 13:56:41 | * | pigmej joined #nim |
| 13:56:41 | * | jacksnipe joined #nim |
| 13:56:41 | * | elbow joined #nim |
| 13:56:41 | * | onionhammer joined #nim |
| 13:56:41 | * | mtj joined #nim |
| 13:56:41 | * | mrkishi joined #nim |
| 13:56:41 | * | rinukkusu joined #nim |
| 13:56:41 | * | adam12 joined #nim |
| 13:58:00 | * | saml joined #nim |
| 13:58:00 | * | Menche joined #nim |
| 13:58:00 | * | pipeep joined #nim |
| 13:58:00 | * | k1i joined #nim |
| 13:58:18 | * | Matthias247 joined #nim |
| 13:58:18 | * | betawaffle joined #nim |
| 13:58:18 | * | ntjnsz joined #nim |
| 13:58:18 | * | mahlon joined #nim |
| 13:58:18 | * | MattAitchison joined #nim |
| 13:58:44 | * | arnetheduck joined #nim |
| 13:58:44 | * | Ven joined #nim |
| 13:58:44 | * | EXetoC joined #nim |
| 13:58:44 | * | MagusOTB joined #nim |
| 13:58:44 | * | walt joined #nim |
| 13:58:44 | * | vinnie joined #nim |
| 13:58:44 | * | dhasenan_ joined #nim |
| 13:58:44 | * | skroll1 joined #nim |
| 13:58:44 | * | qfire joined #nim |
| 13:58:44 | * | jokra joined #nim |
| 14:00:22 | * | saml quit (*.net *.split) |
| 14:00:22 | * | Menche quit (*.net *.split) |
| 14:00:22 | * | pipeep quit (*.net *.split) |
| 14:00:22 | * | k1i quit (*.net *.split) |
| 14:01:06 | * | saml joined #nim |
| 14:01:06 | * | Menche joined #nim |
| 14:01:06 | * | pipeep joined #nim |
| 14:01:06 | * | k1i joined #nim |
| 14:01:10 | * | saml quit (Max SendQ exceeded) |
| 14:01:14 | * | johnsoft quit (Read error: Connection reset by peer) |
| 14:01:34 | * | saml joined #nim |
| 14:01:59 | * | johnsoft joined #nim |
| 14:07:07 | * | Ven quit (Quit: My MacBook has gone to sleep. ZZZzzz…) |
| 14:12:24 | * | Strikecarl joined #nim |
| 14:12:32 | Strikecarl | Can i make my own API? |
| 14:12:56 | Strikecarl | So i can just import lets say Steam |
| 14:13:05 | Strikecarl | and i can get all the procs without having to re define them. |
| 14:13:57 | def- | you can reexport procs from another module or the whole module with "export Steam" |
| 14:17:05 | Strikecarl | How would i do that? |
| 14:19:34 | def- | by writing "export Steam" |
| 14:20:30 | renesac | you already have that Steam module? |
| 14:20:35 | renesac | or you need to make bindings to it? |
| 14:21:53 | Strikecarl | I dont already have a direct Module. |
| 14:21:56 | Strikecarl | I just have stuff like |
| 14:21:59 | Strikecarl | "exitSteam" |
| 14:22:04 | Strikecarl | "openSteam" |
| 14:22:17 | renesac | in nim? |
| 14:22:20 | Strikecarl | ye |
| 14:22:39 | Strikecarl | def- so i do export steam at the bottom of my already made stuff |
| 14:22:39 | renesac | and why would you need to redefine them? |
| 14:22:54 | Strikecarl | and on the new one i do import steam? |
| 14:22:59 | Strikecarl | or export at the top? |
| 14:23:29 | renesac | to export procs in your file, you use "*" |
| 14:23:41 | renesac | the "export Steam" is to export something you imported |
| 14:23:52 | Strikecarl | *proc openSteam |
| 14:23:56 | Strikecarl | or proc *openSteam |
| 14:23:59 | Strikecarl | i guess the second one? |
| 14:24:00 | renesac | at the end |
| 14:24:12 | Strikecarl | proc openSteam*? |
| 14:24:30 | renesac | yeah |
| 14:24:33 | Strikecarl | kk |
| 14:24:39 | renesac | https://github.com/Araq/Nim/blob/master/lib/pure/collections/sets.nim |
| 14:24:43 | EXetoC | 'import Steam' might be what you want then, in addition to exporting with '*' (making available to modules that import it) |
| 14:24:45 | renesac | take a look at any module on nim stdlib |
| 14:25:21 | renesac | exported procs have that * |
| 14:25:30 | renesac | internal procs like "rawGet" don't |
| 14:25:38 | renesac | and then you import your module |
| 14:25:51 | * | darkf quit (Quit: Leaving) |
| 14:25:54 | Strikecarl | so i add * to the procs i want to export |
| 14:25:58 | Strikecarl | then in the new .nim |
| 14:26:07 | Strikecarl | i just do import nameoffilehere |
| 14:26:14 | Strikecarl | and i got them? |
| 14:26:14 | renesac | yeah |
| 14:26:17 | Strikecarl | Great! |
| 14:26:50 | renesac | don't forget to add * to the types/objects you want to export too |
| 14:27:01 | Strikecarl | :=) |
| 14:27:18 | EXetoC | d(:)|< |
| 14:27:40 | Strikecarl | that's one weird fish, EXetoC |
| 14:28:29 | EXetoC | you forgot to tilt your head |
| 14:28:54 | * | Ven joined #nim |
| 14:29:06 | Strikecarl | oh |
| 14:33:45 | Strikecarl | How do i read REG_SZ's? |
| 14:44:09 | * | milosn quit (Ping timeout: 250 seconds) |
| 14:49:00 | * | yymoto2 joined #nim |
| 14:49:49 | * | jholland joined #nim |
| 14:50:06 | * | Strikecarl quit (Quit: http://www.kiwiirc.com/ - A hand crafted IRC client) |
| 14:55:12 | * | pafmaf_ joined #nim |
| 15:00:39 | * | yymoto2 quit (Quit: leaving) |
| 15:08:42 | endragor | Do I understand it correctly that there is no point in making "immutable classes" in Nim? i.e. classes instances of which cannot be modified. Because you'll just use "let" if you don't want your instance to be modifiable |
| 15:09:49 | def- | endragor: sounds right |
| 15:12:52 | Araq | so ... I think I figured out how overloading of '=' should work in Nim. :-) |
| 15:13:55 | Araq | as a side-effect you can implement an efficient shared memory GC all in userland, I think. |
| 15:14:03 | * | Ven quit (Quit: My MacBook has gone to sleep. ZZZzzz…) |
| 15:15:18 | Araq | the basic idea is that a macro can be invoked for '=' which then can inspect the ASTs and generate highly efficient write barriers |
| 15:19:35 | EXetoC | :o |
| 15:20:04 | filwit | interesting |
| 15:23:44 | endragor | is there a way to override object constructor? or at least hide it? |
| 15:24:58 | EXetoC | neat |
| 15:26:59 | Araq | I read about similar in a children's book once. :-) |
| 15:27:38 | def- | endragor: don't export the fields? then you can only use the object constructor to make an empty object, same as with "var x: Foo" |
| 15:29:16 | endragor | not exactly. Basically I want to ensure that objects of my class always satisfy certain properties. I won't be able to do that, if I allow clients to call constructor setting arbitrary fields |
| 15:29:43 | def- | the clients could always just overwrite the raw memory |
| 15:29:45 | endragor | so I want to have my own constructor, and not allow the default one |
| 15:29:59 | endragor | well, yes, but then they are on their own |
| 15:32:55 | * | Ven joined #nim |
| 15:33:46 | endragor | a pragma like {.noConstructor.} would be nice |
| 15:35:43 | filwit | endragor: there's been much discussion about this. You're not alone. I've started an official proposal and patch, but I've another project I spend Nim time on too, so no strong ETA just yet. |
| 15:36:17 | Araq | endragor: also .noConstructor doesn't work well with newSeq[Foo]() |
| 15:36:57 | Araq | the language assume the axiom of choice is available, so a Foo can be constructed without passing any parameters around |
| 15:37:23 | Araq | it keeps surprising me how often this little detail is overlooked |
| 15:37:51 | Araq | as a result objects without default constructor don't work well in C++ either. |
| 15:38:01 | Araq | bbl |
| 15:38:07 | filwit | it's not overlooked, Araq.. it's just often you want sensible defaults for a type, and often that's not it's member's defaults |
| 15:38:10 | endragor | well, ok, what if I add *my own* implementation of empty constructor? |
| 15:38:26 | endragor | and if I don't, throw compile-time error |
| 15:40:23 | endragor | empty constructor is not necessarily equal to "fields have default values". There can be more logic involved. Although in my case empty constructor = fields with default. And I only want to control non-empty constructor |
| 15:41:49 | * | Ven quit (Quit: My MacBook has gone to sleep. ZZZzzz…) |
| 15:50:54 | * | HakanD__ joined #nim |
| 15:55:01 | * | HakanD_ quit (Ping timeout: 264 seconds) |
| 15:55:36 | * | ChrisMAN joined #nim |
| 15:55:58 | endragor | can you declare multidimensional array in Nim? |
| 15:56:05 | endragor | e.g. 2x2 matrix |
| 15:56:47 | filwit | array[2, array[2, float]] |
| 15:57:00 | endragor | thanks |
| 15:59:12 | * | Ven joined #nim |
| 16:00:59 | * | endragor quit (Remote host closed the connection) |
| 16:01:06 | renesac | https://www.openhub.net/p/Nimrod <-- why does it think nimrod is mostly written in C? |
| 16:01:18 | renesac | it must be the C sources directorty |
| 16:02:27 | renesac | and the name should be changed |
| 16:02:58 | renesac | https://www.openhub.net/sessions/new <--- does anyone have an account there? |
| 16:03:24 | * | filwit quit (Quit: Leaving) |
| 16:04:04 | renesac | and even more strange is the 6% of python |
| 16:05:14 | renesac | https://www.openhub.net/p/Nimrod/analyses/latest/languages_summary <-- WTF |
| 16:05:26 | federico3 | besides, is openhub still useful/relevant? |
| 16:05:42 | renesac | I don't know, but it has pretty graphics |
| 16:05:57 | federico3 | heh :) |
| 16:06:43 | * | milosn joined #nim |
| 16:06:44 | renesac | federico3: can you fill the "TODO" parts on the nim for python programmers? |
| 16:07:03 | federico3 | renesac: I just added few data structures a minute ago |
| 16:07:04 | renesac | with the problem you had an your solution |
| 16:07:11 | renesac | before you forget it |
| 16:07:28 | renesac | oh, cool |
| 16:16:15 | * | endragor joined #nim |
| 16:30:13 | * | brson joined #nim |
| 16:37:45 | Araq | endragor: "there can be more logic involved" yes but very often you can also simply make the default value have proper semantics |
| 16:41:23 | endragor | that's true. As I said, my concern is mostly about non-default constructor. |
| 16:43:25 | BlaXpirit_ | honestly, i wouldn't mind constructors disappearing from the language altogether |
| 16:44:17 | BlaXpirit_ | even their syntax feels out of place |
| 16:47:10 | Araq | BlaXpirit_: tuple construction: (field: 2, fieldB: 4) |
| 16:47:28 | Araq | object construction: Obj(field: 2, fieldB: 4) |
| 16:47:39 | Araq | "out of place" my ass |
| 16:47:41 | BlaXpirit_ | that's cool |
| 16:54:12 | Araq | endragor: type Foo {.requiresInit.} = object |
| 16:54:14 | Araq | a, b: int |
| 16:54:15 | Araq | var f: Foo # cannot prove 'f' is initialized |
| 16:55:57 | endragor | Araq: thanks. but I'm looking to override behaviour of Obj(field: 2, fieldB: 4) (or to hide it altogether, and provide my own newObj() proc) |
| 16:57:54 | endragor | to be more concrete, I have Fraction class, and I want to make sure objects of the class are always normalized, which means gcd(numerator, denominator) == 1 and denominator is positive. Without this, almost in every procedure I implement for my fractions, I have to normalize the input fraction first |
| 16:59:10 | Araq | you can always map denominator to denominator+1 internally :P |
| 16:59:27 | Araq | and then (0,0) is a valid fraction |
| 16:59:54 | Araq | btw he have fractions in the stdlib now. |
| 17:00:17 | endragor | Where? |
| 17:00:27 | Araq | in devel |
| 17:00:37 | endragor | having (0,0) as valid fraction is odd |
| 17:00:56 | endragor | and that doesn't solve the problem of keeping the fraction normalized |
| 17:01:50 | Araq | you only need to normalize the result anyway, not the input |
| 17:02:32 | endragor | if I want to compare fractions, I have to normalize them first |
| 17:02:33 | federico3 | renesac: I still don't have an answer |
| 17:05:02 | endragor | Araq: how's new fractions module called? |
| 17:05:16 | Araq | lib/pure/rationals.nim |
| 17:06:06 | * | Trustable joined #nim |
| 17:16:03 | * | yeye123 joined #nim |
| 17:27:50 | * | endragor quit (Remote host closed the connection) |
| 17:37:44 | * | gever-gever quit (K-Lined) |
| 17:41:44 | * | BitPuffin joined #nim |
| 17:44:55 | * | HakanD__ quit (Read error: Connection reset by peer) |
| 17:48:52 | * | Ven quit (Quit: My MacBook has gone to sleep. ZZZzzz…) |
| 17:49:31 | * | Ven joined #nim |
| 17:52:25 | * | repax joined #nim |
| 17:53:32 | * | Ven quit (Client Quit) |
| 18:03:07 | * | BitPuffin quit (Ping timeout: 244 seconds) |
| 18:15:40 | * | Strikecarl joined #nim |
| 18:16:00 | * | elbow is now known as elbow_jason |
| 18:26:36 | Strikecarl | Little problem regarding ftpclient |
| 18:26:47 | Strikecarl | When i excecute "connect()" |
| 18:27:14 | Strikecarl | i get the error code 220, but it expects 230 or 331, i am running a private FTP system that gives me 220 when i login. |
| 18:27:25 | Strikecarl | any way i can add the code 220 to be acceptable? |
| 18:31:27 | * | gsingh93 joined #nim |
| 18:33:33 | * | Trustable quit (Quit: Leaving) |
| 18:35:57 | Araq | Strikecarl: fix the ftpclient module then |
| 18:37:17 | Strikecarl | Also, Araq, what lib allows me to read registry? |
| 18:37:18 | Strikecarl | Winlean? |
| 18:40:59 | * | Trustable joined #nim |
| 18:44:37 | federico3 | the documentation pages are missing a nice search function :) |
| 18:44:53 | Strikecarl | c: |
| 18:46:06 | flaviu | http://nim-lang.org/theindex.html |
| 18:46:55 | federico3 | thanks flaviu |
| 18:50:41 | * | HakanD joined #nim |
| 18:54:34 | Araq | Strikecarl: we don't have a registry lib, only the raw winapi stuff in windows.nim |
| 18:54:46 | Strikecarl | Yeah, trying to use it atm. |
| 18:57:28 | * | Menche quit (Ping timeout: 245 seconds) |
| 19:01:12 | federico3 | a non-blocking readline from stdin? |
| 19:02:06 | dom96 | federico3: doesn't exist |
| 19:02:23 | dom96 | I was able to get it working with asyncfile on Linux |
| 19:02:29 | dom96 | but it doesn't work on Windows IIRC |
| 19:03:39 | federico3 | dom96: any hope if I check the size of the stdin file maybe? |
| 19:04:09 | dom96 | You'd be better off getting async working with it I think |
| 19:08:17 | federico3 | ok, I can do a readline inside an .async. - but then where do I put the output? |
| 19:08:35 | dom96 | not sure what you mean |
| 19:08:43 | dom96 | you can return data from async procs |
| 19:09:49 | reactormonk | federico3, for dictionaries, use tables, not hashset |
| 19:10:26 | * | Pisuke quit (Ping timeout: 272 seconds) |
| 19:11:30 | federico3 | yet, I don't want to block on await |
| 19:15:42 | * | Menche joined #nim |
| 19:16:49 | federico3 | *sigh* |
| 19:17:45 | reactormonk | federico3, so yu basically want a future and continue on that? |
| 19:24:43 | Strikecarl | Can anyone give me an example of RegQueryValueExA? |
| 19:26:38 | federico3 | I'm not sure: one option is have a "thread" wait on input and execute procs that will change the status of some objects while another "thread" mostly sleep and prints the objects status. |
| 19:27:03 | federico3 | reactormonk: and I suspect it won't work because both "threads" are reading/writing attributes on the same objects |
| 19:28:41 | Araq | federico3: can you elaborate on what you're trying to do? |
| 19:29:24 | federico3 | another option is to simply use async to implement a stdin.readline() with a timeout |
| 19:34:10 | federico3 | Araq: generally speaking I want to react to two type of events: a new line in stdin or 1 second has elapsed |
| 19:37:53 | * | Menche quit (Quit: Leaving) |
| 19:38:45 | * | Strikecarl quit (Quit: http://www.kiwiirc.com/ - A hand crafted IRC client) |
| 19:39:38 | * | pafmaf_ quit (Quit: This computer has gone to sleep) |
| 19:43:17 | Araq | federico3: sounds like you want 3 threads. 1 waiting on stdin, 1 using a timer and the main thread waiting on a channel. the other 2 threads writing to the channel |
| 19:46:42 | dom96 | async await supports timers |
| 19:46:43 | * | ChrisMAN quit (Read error: Connection reset by peer) |
| 19:47:03 | dom96 | You can wait 2 futures. |
| 19:47:05 | dom96 | *await |
| 19:47:31 | dom96 | So you can combine sleepAsync() with stdin.readLine (once you implement it for stdin) |
| 19:48:18 | * | yeye123 quit (Quit: Leaving) |
| 19:49:37 | * | Etheco quit (Quit: Leaving) |
| 19:52:59 | * | UberLambda joined #nim |
| 20:29:24 | * | ChrisMAN joined #nim |
| 20:31:11 | * | fizzbooze joined #nim |
| 20:37:35 | * | pafmaf joined #nim |
| 20:39:09 | federico3 | dom96: by starting an async readline and an async sleepasync together and then having the latter go and complete the future from the readline? |
| 20:39:19 | * | Trustable quit (Remote host closed the connection) |
| 20:40:48 | dom96 | federico3: no. You can start them both and then: await readLineFut or sleepAsync(1000) |
| 20:40:59 | dom96 | execution will then resume once either is finished |
| 20:41:10 | dom96 | then you can check which finished |
| 20:41:32 | dom96 | with .finished |
| 20:41:37 | federico3 | oh, awaits to both together, I see |
| 20:48:03 | * | repax quit (Quit: repax) |
| 20:52:42 | * | HakanD quit (Quit: Be back later ...) |
| 21:02:15 | * | bat_R joined #nim |
| 21:02:27 | * | jefus_ joined #nim |
| 21:02:27 | * | user7181 joined #nim |
| 21:03:04 | * | bat_R quit (Client Quit) |
| 21:03:55 | * | mwbrown joined #nim |
| 21:05:05 | * | jefus quit (Ping timeout: 256 seconds) |
| 21:07:03 | * | walt is now known as ggVGc |
| 21:12:02 | * | mpthrapp quit (Ping timeout: 272 seconds) |
| 21:23:20 | Araq | arnetheduck: well now the compiler outputs a shiton of deprecation warnings |
| 21:24:36 | * | user7181 quit (Quit: leaving) |
| 21:27:09 | dtscode | didnt it before |
| 21:27:10 | dtscode | ? |
| 21:28:10 | Araq | no. |
| 21:28:24 | dtscode | hmmm |
| 21:32:06 | * | Joe-T joined #nim |
| 21:33:52 | * | user7181 joined #nim |
| 21:38:20 | * | mpthrapp joined #nim |
| 21:43:42 | * | mpthrapp quit (Ping timeout: 272 seconds) |
| 21:44:48 | * | mwbrown quit (Ping timeout: 252 seconds) |
| 21:48:09 | * | zama quit (Ping timeout: 245 seconds) |
| 21:49:01 | * | zama joined #nim |
| 21:50:07 | * | a5i quit () |
| 21:55:46 | * | BitPuffin joined #nim |
| 21:56:16 | * | nortiero joined #nim |
| 21:58:26 | * | nortiero left #nim (#nim) |
| 21:58:37 | * | nande joined #nim |
| 22:00:16 | * | fizzbooze quit (Quit: WeeChat 1.1.1) |
| 22:13:18 | * | UberLambda quit (Remote host closed the connection) |
| 22:14:15 | * | a5i joined #nim |
| 22:17:54 | * | Joe-T quit (Quit: ZNC - http://znc.in) |
| 22:18:06 | * | Matthias247 quit (Read error: Connection reset by peer) |
| 22:22:41 | Araq | so ... today I implemented the last missing feature for version 1.0. now "only" bugs and "minor" things are left ... |
| 22:23:42 | fowl | what feature |
| 22:23:58 | Araq | overloading of the assignment operator |
| 22:24:11 | ldlework | was that hard |
| 22:24:16 | ldlework | for you, I mean |
| 22:24:40 | Araq | it was tedious |
| 22:24:45 | * | ldlework nods. |
| 22:25:18 | Araq | and afterwards I had a couple of brilliant ideas ... |
| 22:25:35 | BlaXpirit_ | i'm just scared that the standard library would be frozen |
| 22:26:08 | BlaXpirit_ | you can't deny that parts of it were built chaotically |
| 22:26:28 | Araq | *shrug* |
| 22:26:49 | Araq | you didn't even manage to push the regex patches upstream |
| 22:27:28 | BlaXpirit_ | i didn't like the changes because they can't work well based on the "re" library |
| 22:27:36 | flaviu | Araq: I've been playing with rust recently. You were right, Result<> everwhere is a PITA. |
| 22:27:41 | flaviu | Exceptions do work better. |
| 22:27:58 | Araq | hrm ... i need to save this somewhere |
| 22:28:01 | ldlework | lol |
| 22:28:17 | Araq | but oh well the channel is logged |
| 22:28:27 | flaviu | haha :) |
| 22:29:01 | ldlework | flaviu: how's modern rust treating you in other respects? |
| 22:32:30 | Araq | BlaXpirit_: IMO all we need to do is to integrate Nimble with the installer and ensure it works really well. I don't want to maintain the stdlib really, I want to trim it. |
| 22:32:50 | BlaXpirit_ | trimming it would be great |
| 22:32:55 | BlaXpirit_ | Araq, i said earlier that there is just too much to fix about the current "re" library |
| 22:33:16 | BlaXpirit_ | we have a new beautiful lib |
| 22:33:21 | Araq | BlaXpirit_: works for my projects though |
| 22:33:35 | flaviu | ldlework: It's pretty nice, with problems, of course. I need to have some generics in non-generic code, which I assume is because of borrow checker. Documentation tends to be incomplete, although I do like their "stable", "deprecated", "experimental" categories. |
| 22:34:23 | BlaXpirit_ | and the changes to "random" - i'm at a loss |
| 22:34:34 | BlaXpirit_ | so many difficulties about it |
| 22:35:11 | BlaXpirit_ | RAND_MAX can't be a constant and isn't even guaranteed to be a power of two |
| 22:35:46 | BlaXpirit_ | i used to advocate removing it and just leaving rand() but then i remembered about JS backend |
| 22:39:11 | BlaXpirit_ | i don't do pull requests about the standard library because it often feels foreign to me. |
| 22:39:39 | BlaXpirit_ | we have functions operating on arrays when we could have them work on iterators too, for free |
| 22:40:12 | BlaXpirit_ | but there is no infrastructure for that |
| 22:41:07 | BlaXpirit_ | there is 0 support for custom container types |
| 22:41:56 | BlaXpirit_ | if i make a super awesome sparse array, i can't use sequtils on it |
| 22:42:39 | Araq | why not? mapIt just requires an 'items' iirc |
| 22:43:13 | BlaXpirit_ | hm let me see |
| 22:43:45 | BlaXpirit_ | ok the templates would work |
| 22:45:19 | Araq | and 'concept' just recently got a nice name |
| 22:46:01 | BlaXpirit_ | concepts for collections/containers are sorely needed |
| 22:46:34 | def- | BlaXpirit_: they don't work well enough yet in my experience |
| 22:47:06 | BlaXpirit_ | openarray is harmful |
| 22:47:49 | BlaXpirit_ | it's literally better to just use a generic than openarray |
| 22:48:37 | BlaXpirit_ | (let alone seq) |
| 22:50:27 | Araq | def-: any bug number I should give priority? (for concepts) |
| 22:50:35 | * | brson quit (Quit: leaving) |
| 22:50:49 | BlaXpirit_ | I tried to make a concept for "random access container" and failed |
| 22:51:17 | Araq | len and [] ? |
| 22:51:30 | Araq | what is your definition of it? |
| 22:51:49 | BlaXpirit_ | https://github.com/BlaXpirit/nim-random/blob/a18cb/src/random/private/util.nim#L32 |
| 22:51:55 | BlaXpirit_ | low,high,len,[] |
| 22:52:45 | BlaXpirit_ | i tried different things, there used to be a [T] of course |
| 22:54:56 | BlaXpirit_ | hm it actually seems to work now. c[0] is T |
| 22:55:16 | def- | Araq: I haven't tried much but this didn't work: https://github.com/Araq/Nim/pull/2449#commitcomment-10564458 |
| 22:55:37 | * | wb quit (Ping timeout: 264 seconds) |
| 22:59:40 | BlaXpirit_ | yeah, definitely seems like concept was improved recently |
| 23:00:32 | Araq | def-'s stuff is spot on though |
| 23:02:10 | fowl | nooo |
| 23:02:17 | fowl | container[t] isnt useful |
| 23:02:47 | BlaXpirit_ | k |
| 23:02:55 | Araq | fowl: why not? |
| 23:03:18 | fowl | Araq, this is actually Iterable[T] |
| 23:03:39 | BlaXpirit_ | no need for low,high then |
| 23:04:03 | Araq | fowl: it's only a name |
| 23:04:13 | Araq | we wonder why it doesn't work |
| 23:04:35 | BlaXpirit_ | anything's better than just making the whole standard libreary work only for seq |
| 23:04:35 | fowl | it does work if you remove the ordinal stuff |
| 23:04:41 | fowl | and the "value.type is T" which is weird |
| 23:05:10 | def- | well, that example is mostly from the manual iirc |
| 23:05:21 | Araq | yeah |
| 23:05:22 | * | pafmaf quit (Quit: This computer has gone to sleep) |
| 23:05:25 | BlaXpirit_ | ordinal actually doesn't work |
| 23:05:28 | BlaXpirit_ | SomeInteger does |
| 23:05:47 | Araq | interesting |
| 23:05:53 | fowl | Araq, whats the point of being able to pass multiple args to concept |
| 23:06:09 | Araq | hrm? |
| 23:06:09 | fowl | ie you could have "concept x, var y" does that mean y is the var version of x? |
| 23:06:31 | Araq | good question |
| 23:06:40 | def- | oh right, ordinal doesn't even exist |
| 23:06:45 | def- | it's SomeOrdinal |
| 23:06:58 | Araq | yeah that's actually the biggest problem |
| 23:07:02 | Araq | make a typo |
| 23:07:13 | Araq | and nothing whatsoever is reported! |
| 23:07:36 | BlaXpirit_ | i don't know what happened, but i tried this not so long ago and nothing worked |
| 23:07:39 | BlaXpirit_ | now everything is great |
| 23:07:50 | * | jefus__ joined #nim |
| 23:07:58 | Araq | BlaXpirit_: well I *did* fix some things wrt concept |
| 23:08:04 | BlaXpirit_ | great |
| 23:08:37 | BlaXpirit_ | is it time to replace openarray with concepts? |
| 23:09:17 | fowl | def-, theres also Ordinal |
| 23:09:45 | Araq | wtf if I replace ordinal with SomeOrdinal everything works? :O |
| 23:09:54 | def- | Araq: yes, seems so |
| 23:10:43 | fowl | pairs() doesnt |
| 23:10:57 | * | vendethiel quit (Ping timeout: 250 seconds) |
| 23:11:04 | BlaXpirit_ | pairs is weird btw |
| 23:11:44 | * | jefus_ quit (Ping timeout: 272 seconds) |
| 23:12:29 | BlaXpirit_ | nvm i should go sleep |
| 23:13:08 | * | BlaXpirit_ quit (Quit: Quit Konversation) |
| 23:13:12 | * | pafmaf joined #nim |
| 23:14:12 | * | TEttinger joined #nim |
| 23:18:05 | * | pafmaf quit (Client Quit) |
| 23:19:37 | * | Joe-T joined #nim |
| 23:20:17 | onionhammer | Araq we need to trim more out of the std lib |
| 23:20:23 | onionhammer | and work on making nimble a lot better |
| 23:20:41 | federico3 | can I "map" an integer to an eval value? |
| 23:26:10 | Araq | onionhammer: well? you just like to repeat what I said? |
| 23:26:23 | fowl | federico3, do you mean fold? |
| 23:26:34 | onionhammer | araq dont see a lot of movement |
| 23:26:39 | onionhammer | on that front |
| 23:27:02 | Araq | we made several things Nimble packages |
| 23:27:07 | onionhammer | months ago |
| 23:27:08 | federico3 | fowl: yep |
| 23:27:10 | onionhammer | but theres a lot more |
| 23:27:24 | fowl | federico3, fold would be for example going from @[1,2,3] apply operation `+` and end up with 6 |
| 23:27:30 | onionhammer | unless you did more recently |
| 23:27:34 | fowl | federico3, there is foldl/foldr in sequtils |
| 23:27:56 | Araq | well I can only work every day and hope my wife doesn't file for divorce. |
| 23:28:14 | federico3 | hm, I mean, given an integer, extract the enum value where ord(value) == int |
| 23:28:38 | fowl | federico3, that is easy |
| 23:28:51 | fowl | federico3, just use type conversion on the int |
| 23:29:04 | onionhammer | Araq maybe just mark them, you dont have to do the work of moving them |
| 23:29:09 | fowl | .eval type tx = enum aa,bb,cc; echo tx(1) |
| 23:29:13 | Mimbus | fowl: bb |
| 23:29:20 | onionhammer | Araq nimble needs to improve, and the lib list needs to include nimble packages |
| 23:29:22 | federico3 | thanks! |
| 23:29:55 | Araq | onionhammer: make the db_* modules a Nimble package |
| 23:29:55 | * | user7181 quit (Quit: leaving) |
| 23:30:09 | Araq | graphics.nim -- Nimble package |
| 23:30:13 | onionhammer | do they even still work? |
| 23:30:35 | Araq | quite sure they do |
| 23:30:42 | Araq | I regularly get PRs about them |
| 23:34:37 | onionhammer | well, i'm not really interested in maintaining nimble projects that i didnt write; it'd be better if they were moved into nim-code or something |
| 23:34:55 | Araq | isn't that nim-lang by now? |
| 23:35:13 | onionhammer | right, thats the one i mean |
| 23:35:31 | Araq | also the distribution should bundle these somehow |
| 23:35:40 | onionhammer | that would be nice |
| 23:36:28 | Araq | heck with 100+ packages we could even bundle everything, I think |
| 23:36:31 | onionhammer | the distribution of nim could be built on nimble |
| 23:36:55 | Araq | well make it happen |
| 23:36:56 | * | irrequietus quit () |
| 23:37:02 | onionhammer | nimble install oldstdlib :P |
| 23:38:06 | onionhammer | i'm going to have a ton of free time... after july :P |
| 23:38:47 | Araq | want to watch me on OSCON. a wise move :P |
| 23:39:54 | onionhammer | hmm? |
| 23:40:03 | onionhammer | im out of the loops :) |
| 23:40:28 | * | Menche joined #nim |
| 23:41:51 | def- | http://www.oscon.com/open-source-2015/public/schedule/detail/42497 |
| 23:42:18 | Araq | oh so it's finally online |
| 23:46:52 | reactormonk | onionhammer, how would you make nimble better? |
| 23:48:22 | Araq | good night |
| 23:53:27 | * | HakanD joined #nim |
| 23:57:29 | * | netroby joined #nim |
| 23:57:36 | * | HakanD quit (Ping timeout: 240 seconds) |
| 23:59:50 | * | Kingsquee joined #nim |