<< 28-09-2016 >>
00:23:19*fredrik92 quit (Quit: Client disconnecting)
00:27:29*chemist69 quit (Ping timeout: 272 seconds)
00:44:24*brechtm quit (Remote host closed the connection)
00:46:08*ftsf joined #nim
00:50:15*libman quit (Read error: Connection reset by peer)
00:53:49*chemist69 joined #nim
01:02:19*gangstacat quit (Ping timeout: 272 seconds)
01:08:57*ARCADIVS joined #nim
01:39:41*chemist69 quit (Ping timeout: 272 seconds)
01:42:49*zielmicha[m] joined #nim
01:48:26*ehmry joined #nim
01:52:00*brson quit (Ping timeout: 276 seconds)
01:52:29*chemist69 joined #nim
02:12:27*hohlerde joined #nim
02:15:51*Snircle_ quit (Quit: Textual IRC Client: www.textualapp.com)
02:54:15*brson joined #nim
02:58:41*brson quit (Ping timeout: 240 seconds)
03:03:23*Jesin quit (Quit: Leaving)
03:53:35*NimBot joined #nim
03:58:23*chemist69 quit (Ping timeout: 272 seconds)
04:00:09*chemist69 joined #nim
04:23:10*yglukhov joined #nim
04:29:21*yglukhov quit (Ping timeout: 240 seconds)
05:05:59*jeffc quit (Ping timeout: 252 seconds)
05:34:09FromGitter<MiniDude22> how do you initialize an array that is the var result in a proc?
05:37:08FromGitter<MiniDude22> and is there a way to turn a sequence of a type into an array of a type?
05:38:04ftsfmake an var array then copyMem from seq to array?
05:38:16ftsfor cast it as an array
05:41:15*wgf_ joined #nim
05:50:45*desophos quit (Read error: Connection reset by peer)
05:54:32FromGitter<Araq> you cast it to a ptr array, perhaps
05:54:58FromGitter<Araq> in general you don't do it, seqs have dynamic size, arrays don't so there is no way to convert it
05:55:11FromGitter<MiniDude22> Right. Makes sense.
05:55:33FromGitter<Araq> post a snippet here please. it smells like you have no idea what you're doing ;-)
05:56:00FromGitter<MiniDude22> This is how I ended up doing it.
05:56:26FromGitter<MiniDude22> ```
05:56:51FromGitter<MiniDude22> ```code paste, see link``` [https://gitter.im/nim-lang/Nim?at=57eb5ba335e0f31c6c4b7600]
05:57:17*jeffc joined #nim
05:57:21FromGitter<MiniDude22> the toRunes returned a seq, I was trying to have it simpler, but this works fine
05:58:56FromGitter<Araq> ok, you can remove the s.runeLen call, but it looks good
05:59:24FromGitter<Araq> if index > 137: raise ... in the loop body is possible too
05:59:48FromGitter<MiniDude22> Ah okay that might work better
06:00:45FromGitter<MiniDude22> is it possible to return an array that's size is the strings length? this returns blanks after the characters
06:01:11FromGitter<MiniDude22> whose size is*
06:14:39FromGitter<Araq> you can return a tuple len, data or have a 'len: var int' parameter but I think your 'chatString' should return a seq
06:22:24*bjz joined #nim
06:22:53*bjz quit (Client Quit)
06:40:01*bjz joined #nim
06:48:52*gokr joined #nim
06:50:27*fredrik92 joined #nim
06:52:59FromGitter<MiniDude22> I'm passing it over a named pipe so I need it in binary form so that i can come out as a wchar_t on the other side
06:53:10*Arrrr joined #nim
06:53:38FromGitter<MiniDude22> and array of things passes and is read properly, but a sequence doesn't go through when i tried it.
06:54:54FromGitter<Araq> for the seq you need to do addr(s[0])
06:55:02FromGitter<Araq> that's common knowledge :P
06:55:20FromGitter<Araq> or maybe even unsafeAddr(s[0])
06:55:25FromGitter<MiniDude22> Maybe for pro's like you :P
06:56:16*yglukhov joined #nim
06:57:08FromGitter<MiniDude22> So I'm writing to the file like ⏎ ```code paste, see link``` [https://gitter.im/nim-lang/Nim?at=57eb69c454302caa6f262e3e]
06:57:49FromGitter<MiniDude22> I'll play with the sequences again in a bit xD thanks!
06:59:21FromGitter<Araq> and please tell me what you read: tutorial, manual, nim-in-action?
06:59:49FromGitter<Araq> what section needs to be expanded and explain about unsafeAddr/addr s[0] ?
07:00:09ftsfwhat's the difference between addr(myseq) and addr(myseq[0]) ?
07:00:18FromGitter<Araq> it surely is frustrating to always answer the same question
07:01:02Araq_ftsf: addr(myseq) returns the address of the seq
07:01:32Araq_a seq starts with a size, capacity header so this is where the resulting pointer points to
07:01:37ftsfahh I see
07:02:16ftsfthanks
07:04:39*FromGitter * girvo waves
07:04:41FromGitter<MiniDude22> I've been through the tutorials, read a lot of the manual, though I guess I missed the part about seq's and their addr's
07:05:18FromGitter<girvo> Quick question: anyone implemented/used object pools or something similar in Nim?
07:05:25ArrrrIt is irc-tier knowledge
07:05:58ArrrrI have done something like that, but not too complicated
07:06:46FromGitter<Araq> @girvo are you the guy who promised me an email about comp-sci papers years ago?
07:07:06*bjz quit (Quit: My MacBook Pro has gone to sleep. ZZZzzz…)
07:07:10FromGitter<girvo> @Araq I am indeed, which I don't think I ever sent through
07:08:08FromGitter<Araq> exactly. :P
07:08:57FromGitter<girvo> Type system discussion, yeah? Trying to see if I can remember hah
07:10:18FromGitter<MiniDude22> @Araq sorry for having you explain things so many times. xD We all appreciate the halp!
07:15:01*gokr quit (Ping timeout: 240 seconds)
07:18:49*gokr joined #nim
07:23:30*Andris_zbx joined #nim
07:24:49*Trustable joined #nim
07:25:06FromGitter<Araq> @MiniDude22 you're welcome
07:25:23FromGitter<MiniDude22> <3
07:41:21*PMunch joined #nim
08:05:48*zielmicha[m] quit (Remote host closed the connection)
08:05:49*M-Quora quit (Remote host closed the connection)
08:05:50*TheManiac quit (Remote host closed the connection)
08:05:50*ehmry quit (Remote host closed the connection)
08:05:51*hohlerde quit (Write error: Connection reset by peer)
08:12:02*zielmicha[m] joined #nim
08:13:17*MyMind joined #nim
08:20:10*ftsf quit (Quit: :q!)
08:31:15*enthus1ast1 quit (Ping timeout: 265 seconds)
08:39:22*brechtm joined #nim
08:46:21*bjz joined #nim
08:59:52*zielmicha[m] quit (K-Lined)
09:01:45*chemist69 quit (Ping timeout: 272 seconds)
09:07:27*chemist69 joined #nim
09:17:36*Demon_Fox quit (Quit: Leaving)
09:17:38*aedigix joined #nim
09:21:30*zielmicha[m] joined #nim
09:36:11*Pisuke joined #nim
09:37:59*bjz quit (Quit: My MacBook Pro has gone to sleep. ZZZzzz…)
09:39:21*Sembei quit (Ping timeout: 276 seconds)
09:39:31*gangstacat joined #nim
09:40:04*Sembei_ joined #nim
09:42:41*Pisuke quit (Ping timeout: 240 seconds)
09:46:27*gangstacat quit (Quit: Leaving)
09:47:11*gangstacat joined #nim
09:49:30*Sembei joined #nim
09:49:50*bjz joined #nim
09:53:00*Sembei_ quit (Ping timeout: 276 seconds)
10:29:35*ftsf__ quit (Remote host closed the connection)
10:35:31*elrood joined #nim
10:51:23*Guest26614 joined #nim
10:52:37*enthus1ast joined #nim
11:01:03*arnetheduck joined #nim
11:01:50*Snircle joined #nim
11:11:37*brechtm_ joined #nim
11:13:30*brechtm_ quit (Remote host closed the connection)
11:15:23*brechtm quit (Ping timeout: 272 seconds)
11:29:04*brechtm joined #nim
11:48:05*filcuc joined #nim
11:59:18*brechtm quit (Remote host closed the connection)
12:06:02*brechtm joined #nim
12:07:17*corecode joined #nim
12:07:19corecodehi!
12:08:43*brechtm_ joined #nim
12:10:15Araq_corecode: welcome
12:11:07corecodei just found out about nim, and i'm intrigued. mostly interested for systems (embedded) development
12:11:07*brechtm quit (Ping timeout: 272 seconds)
12:13:06*Arrrr quit (Quit: WeeChat 1.5)
12:19:35*fredrik92 quit (Read error: Connection reset by peer)
12:19:36*Guest26614 quit (Quit: Leaving)
12:19:50*PMunch quit (Quit: leaving)
12:30:46federico3welcome corecode
12:38:27*cheatfate quit (Read error: Connection reset by peer)
12:38:56*cheatfate joined #nim
12:51:02*hohlerde joined #nim
12:51:03hohlerdecorecode: welcome
12:51:22hohlerdearaq: just found NimEdit. looks awesome, especially to use it in presentations
13:03:27Araq_ah yeah I need to open source it and fix its remaining bugs
13:03:46Araq_which admittedly make it hardly workable even for me :P
13:04:15Araq_but hey, I use it all the time for Nim development so at least I'm eating my dogfood
13:04:31*bjz quit (Read error: Connection reset by peer)
13:05:23*bjz joined #nim
13:06:44*arnetheduck quit (Ping timeout: 244 seconds)
13:07:22*arnetheduck joined #nim
13:13:23*ARCADIVS quit (Quit: ARCADIVS)
13:34:59*M-Quora joined #nim
13:35:07*TheManiac joined #nim
13:35:08*ehmry joined #nim
13:59:21*MyMind quit (Ping timeout: 240 seconds)
14:16:05hohlerdearaq: hehe, I realized that nimsuggest has some problems with NimEdit, but it may be a problem with my nim installation
14:16:22*pregressive joined #nim
14:17:58*MyMind joined #nim
14:22:33*krux02 joined #nim
14:23:07krux02is there a dumpTree alternative that runs after template and macro expansions?
14:23:48krux02I would like to see the generated code of my macros
14:29:32*alexvl_zbx joined #nim
14:34:51*alexvl_zbx quit (Quit: leaving)
14:37:27FromGitter<endragor> @krux02: put `echo treeRepr result` at the end of the macro
14:38:37krux02endragor: dosn't work for two reasons, treeRepr result does not run macro expansion on the result, and in my case I have a template
14:43:50krux02endragor: I now use this macro debugAst(ast: typed): untyped = echo ast.repr; ast
14:44:47krux02the typed argument expands macros for me
14:53:04hohlerdeare there any game vids showing games made in nim? so far I have found the urhonimo preview on youtube
14:55:41*arnetheduck quit (Ping timeout: 240 seconds)
15:03:27*mitai joined #nim
15:08:51FromGitter<coffeepots> krux02 thanks for that macro, that's pretty handy. Should that be in the stdlib?
15:19:51*chemist69 quit (Ping timeout: 272 seconds)
15:24:37*yglukhov_ joined #nim
15:26:48*yglukho__ joined #nim
15:27:47*yglukhov quit (Ping timeout: 244 seconds)
15:28:41*yglukhov_ quit (Ping timeout: 240 seconds)
15:30:09*FreezerburnV joined #nim
15:30:09krux02I would appreciate it
15:30:56krux02hohlerde: making a good game is really a lot of work
15:31:24Araq_krux02: doesn't have to be good, only needs nice screenshots for our website :P
15:31:25krux02so I am working on some graphcis library that should eventually be used to make games
15:31:39*yglukho__ quit (Ping timeout: 276 seconds)
15:31:58*kulelu88 joined #nim
15:32:03krux02but a finished game like a new Diablo, does not only take a while to make, it also requires the people who really want to make such a game to believe in Nim
15:33:04Araq_I know of 2 commercial game projects, one being released anytime soon
15:35:30*pregressive quit (Remote host closed the connection)
15:37:08krux02I hate this magic pragma when browsing the standard library
15:37:16FromGitter<dom96> This game is now made in Nim: https://twitter.com/impbox/status/765102630280409088
15:37:35*corecode is still working through the nim language spec
15:37:58FromGitter<coffeepots> i chose nim to make games. Game dev is certainly loads of work, but it's fun. Would love to see what others have done though.
15:39:31FromGitter<coffeepots> i can see why nim is (or seems to be) popular for gamedev
15:40:52Araq_krux02: what's the alternative?
15:41:21Araq_direct implementation with whens and .emits? :-)
15:41:33Araq_hmmm could have worked for most magics
15:41:37Araq_oh well
15:41:38federico3nice came
15:41:44federico3*game
15:42:24Araq_Nim's way of doing .magic is something I'm proud of
15:43:20krux02I don't think that emit should be in the standard library, but I think a lot of stuff might work without magic
15:43:55Araq_I have heard this before.
15:44:07Araq_it's correct for about 5 out of 100 magics.
15:44:21krux02I feel that a lot that has been done in an early state of the language and now it can be done by using other features of the language
15:44:58krux02the problem is just that it makes reading the code much harder, because it feels like a dead end, and I don't know where I should continue
15:45:32Araq_how do you implement system.`+` for ints so that it produces decent code, acknowledges the --overflowChecks switch and works at compile time?
15:45:57krux02I would not want to make any statements, because I really don't know the dependencies, but you have to admit, there are really a lot of magics
15:46:14Araq_how would hiding the prototype of this `+` help readability?
15:46:25*chemist69 joined #nim
15:46:27Araq_in C# it's simply not in a library at all, hurray.
15:47:26krux02I don't have the problem with `+` or that there are some magics, some things just need to be done that way, but I have the feeling, that the barrier to do something as magic is quite low
15:47:35Araq_I think this is actually a big problem of Nim. People look at system.nim and see a mess.
15:48:30Araq_and they compare it to Go which doesn't have a system.nim. Not having system.nim is not a benefit, it's a design flaw.
15:48:44Araq_because then it's even moreso built into the compiler
15:48:52krux02it would be nice, if there would be some instructions that tell me where I would need to look for the implementation when in code it is implemented as magic
15:48:59Araq_with even more possibilities to special case things
15:49:34Araq_just think about it. Nim uses one set of overloading resolution rules for *everything*.
15:49:59Araq_it's super honest and the rules scale down to its integer primitives.
15:50:47Araq_krux02: just grep for mMagic in the compiler then and see its implementation
15:50:56krux02languages like Go or C have harder distinction between language features and library features. In rust I thought I was using a library feature when I was calling it's version of printf, but when I looked up how it worked it ended up being a language feature
15:53:53*brson joined #nim
15:55:02krux02Araq_: thanks for the hint. I think this comment should be somewhere in the documentation, because when reading the system.nim it is essential to know that
15:55:38Araq_why?
15:56:02Araq_do we want people to read the compiler implementation?
15:56:28cheatfateAraq_, is it possible to make text constant multiline?
15:56:49cheatfatei have very big text constant and i want to store it inside of 80 columns
15:57:27cheatfateahh i forgot.`"""`
15:58:29krux02yes we do
15:58:35*krux02 quit (Quit: Verlassend)
16:04:58*Ven_ joined #nim
16:06:05*gokr quit (Ping timeout: 272 seconds)
16:06:17*Andris_zbx quit (Remote host closed the connection)
16:09:15*brechtm_ quit (Remote host closed the connection)
16:09:54*pregressive joined #nim
16:10:06*Ven_ quit (Ping timeout: 264 seconds)
16:10:59*Ven_ joined #nim
16:13:29dom96hello
16:13:41*filcuc quit (Ping timeout: 252 seconds)
16:14:01Araq_hi
16:16:40dom96Developers Developers Developers. Release Release Release.
16:18:06flyxthanks for reminding me of that video.
16:18:56*kulelu88 quit (Ping timeout: 244 seconds)
16:27:49*yglukhov joined #nim
16:27:54*FreezerburnV quit (Quit: Leaving)
16:28:04*aziz joined #nim
16:32:17*yglukhov quit (Ping timeout: 264 seconds)
16:32:52*xet7_ joined #nim
16:33:21*enthus1ast quit (Ping timeout: 240 seconds)
16:47:44dom96I seriously need destructors
16:47:55dom96My site is suffering from constant crashes because of fd leaks :/
16:48:17Araq_a destructor hardly does more than a withFile() template would give you
16:48:29Araq_tried a withFile?
16:48:43dom96Pretty sure this is to do with AsyncHttpClient
16:49:14Araq_also you should attach a finalizer that does the close() for you
16:49:16dom96I would hope a destructor would do what withFile does but implicitly
16:49:31Araq_implicitly yes, but hardly with any more power
16:50:10Araq_and krux02 is working on their design, so it's better you fix the problem now rather than wait for the perfect solution and be disappointed by it
16:51:03*xet7_ quit (Ping timeout: 272 seconds)
16:52:41*wgf_ quit (Quit: Leaving)
16:53:32dom96Of course, I'm trying to fix it now
16:55:13dom96Unfortunately finding the cause is tricky...
16:59:14*MyMind quit (Ping timeout: 244 seconds)
17:03:51*MyMind joined #nim
17:06:07*brechtm joined #nim
17:07:24*libman joined #nim
17:09:18*cheatfate quit (Read error: Connection reset by peer)
17:09:25*cheatfate_ joined #nim
17:13:51*MyMind quit (Ping timeout: 272 seconds)
17:22:41*nsf quit (Quit: WeeChat 1.5)
17:25:23*krux02 joined #nim
17:26:54*cheatfate_ is now known as cheatfate
17:28:45krux02Araq_: I had to leave sorry for disappearing. We do not necessaryly want people to read the compiler implementation, but with the current documentation state for certain things the best documentation is the code. And for those people the code should be readable
17:29:14Araq_we could also hide the .magic pragma :P
17:29:26Araq_actually we already do in devel
17:29:39Araq_pragmas have to expanded per mouse click
17:29:53krux02are you joking?
17:31:55Araq_no, that was a feature request
17:32:24Araq_either way, add this remark to the docs, I will accept it
17:32:41krux02ok I will do it
17:33:46krux02I mean when someone want's to have pragmas to be folded in the editor and expand them per mousclick
17:33:48krux02that's ok
17:33:53krux02that's an editor config
17:34:16krux02But generally I am against every feature that tries the hide how things work internally
17:34:17*libman quit (Read error: Connection reset by peer)
17:34:19*Trustable quit (Remote host closed the connection)
17:34:42krux02things should be as transparent as possible so people can understand them as easy as possible
17:34:46krux02no implementation hiding
17:35:14Araq_yes and no. too much documentation can be overwhelming.
17:35:30krux02yes documentation should simplify things a bit
17:35:50Araq_90% of all the questions that I answer here (number made up) can be read somewhere in Nim's docs
17:36:12Araq_and I don't blame people for not reading things, nobody can read everything.
17:37:08krux02ah you mean you hide the pragmas in the documentation generation?
17:38:37Araq_in the generated docs, yes
17:39:23krux02I mean some pragmas are really optional to see in the documentation, and for magics that is definitively the case, but other pragmas are really important to see, for example a union praga or something like that that really changes the behavior of the underlying structure
17:40:11Araq_behaviour changing pragmas are really rare and should be spelt out explicitly in the text
17:40:24Araq_and link to the manual section or something
17:42:40krux02ok I start to relax now again
17:42:55krux02was a bit afraid that in source pragmas would disappear
17:45:02*brechtm quit (Remote host closed the connection)
17:45:39euantorPart of the problem regarding the docs thing in the past has been that searching on Google turns up very little. Since the new docs have search built in, hopefully that will be improved a little :)
17:46:05*brechtm joined #nim
17:47:29*brechtm quit (Remote host closed the connection)
17:52:57*libman joined #nim
17:54:44*brechtm joined #nim
17:54:54*Demon_Fox joined #nim
17:56:16*Ven_ quit (Read error: Connection reset by peer)
17:56:36baabelfishAraq_: on the topic of pragmas, what is the status of func keyword? :P
17:57:22*aedigix quit (Remote host closed the connection)
17:57:42*aedigix joined #nim
17:59:31Araq_as long as the exciting new effects 'writes' and 'escapes' are not finished but might end up to be the default for 'func', I don't want to rush what 'func' should mean
17:59:54*brechtm quit (Ping timeout: 264 seconds)
18:00:48Araq_a shortcut for .noSideEffect is sexy but we have many more effects
18:00:55Araq_and more interesting effects
18:01:17Araq_so it's not obvious what 'func' should mean
18:01:18*Ven_ joined #nim
18:05:30flyxAraq_: can you give some feedback on https://github.com/nim-lang/Nim/issues/4832 ? whether it is actually a bug or I am just holding it wrong?
18:07:09Araq_it's a bug
18:08:37baabelfishAraq_: is there a talk or a blog post somewhere around the internet what the effect system can/will/could do?
18:08:59*Ven_ quit (Ping timeout: 244 seconds)
18:09:13Araq_hmm no, the manual mentions what's available
18:09:38*xet7_ joined #nim
18:09:43baabelfishAraq_: yeah I have read it but still my imagination is limited
18:10:07baabelfishI mean like "what can I use these for in real world applications"
18:14:09Araq_proving at compile-time that code has no dependencies on a database surely would have been helpful for me in the real world
18:21:13*Ven_ joined #nim
18:21:53*Ven_ quit (Read error: Connection reset by peer)
18:22:57*kulelu88 joined #nim
18:31:43*LittleZeeZee joined #nim
18:32:15*LittleZeeZee quit (Client Quit)
18:33:09*libman quit (Read error: Connection reset by peer)
18:37:40*irrequietus joined #nim
18:41:03*Ven_ joined #nim
18:45:01dom96Where do we stand with our release tarballs?
18:45:08dom96Araq_: federico3?
18:48:11dom96Anybody free (and on Windows) that could test our installers?
18:48:18federico3dom96: apart from potentially removing the html files, the build of a binary pkg from the source tarballs fails when trying to build the doc page for "coro", but I can work around it. AFAIK there are no other blockers
18:48:47Araq_dom96: will test them very soon
18:48:56dom96Here is a link to the installers: https://gitlab.com/nim-lang/nim/builds/4497667/artifacts/browse/build/
18:49:07dom96federico3: what's the cause of the error?
18:49:19dom96could we fix it instead of working around it? :)
18:50:07federico3coro seems to import from lib/arch/arch.nim
18:50:27*enthus1ast joined #nim
18:50:36federico3https://github.com/nim-lang/Nim/blob/devel/lib/arch/arch.nim#L44 and reaches this Unsupported architecture
18:51:55*libman joined #nim
18:52:42federico3looks like the fix for https://github.com/nim-lang/Nim/issues/3245 disabled compiling coro but maybe the doc page should also implement the same check?
18:53:28dom96sure
18:53:40dom96just add when defined(nimdoc) or similar wherever required
18:54:30*Ven_ quit (Ping timeout: 264 seconds)
18:54:33federico3besides, the naming is a bit misleading
19:01:14*Ven_ joined #nim
19:07:57*FreezerburnV joined #nim
19:09:41*Ven_ quit (Ping timeout: 240 seconds)
19:10:48federico3dom96: ideally "nim doc2" should still run against https://github.com/nim-lang/Nim/blob/devel/lib/pure/coro.nim\#L10 and return a doc page that goes "sorry, Nim was build without coroutine support"
19:11:17federico3I'm not sure if we can prevent nim doc2 from trying to import arch
19:11:28*Jesin joined #nim
19:12:56dom96federico3: add a when defined(nimdoc) into arch.nim
19:13:54federico3be aware that coro.nim is not displaying the error "Coroutines require -d:nimCoroutines". under nim doc2
19:14:39federico3I'm getting the error stright from arch
19:15:04dom96I don't think I understand what the problem is
19:16:58federico3http://pastebin.com/URJhJ9hc that's what I'm getting
19:20:34federico3puzzling
19:21:35federico3somehow nim doc2 is importing arch.nim before reading anything from coro.nim
19:21:51*Ven_ joined #nim
19:22:35*Ven_ quit (Read error: Connection reset by peer)
19:22:38*libman quit (Quit: Leaving.)
19:22:52*libman joined #nim
19:23:00cheatfatedom96, what test do you need?
19:23:05cheatfateexactly?
19:23:09cheatfatejust install and run?
19:23:12cheatfateor run tests?
19:23:44dom96install and make sure everything works
19:23:47dom96Not necessarily tests
19:23:52dom96just compile some things
19:24:12*brechtm joined #nim
19:26:18*enthus1ast quit (Ping timeout: 264 seconds)
19:26:54dom96federico3: interesting
19:27:01*libman quit (Ping timeout: 240 seconds)
19:27:27federico3dom96: looks like changes to arch.nim are not affecting nim doc2 at all - I guess I should rebuild nim, but why?
19:28:03*libman joined #nim
19:28:05dom96that suggests that Nim is picking a different stdlib on your system
19:28:45federico3could it be picking up an object file instead?
19:29:51cheatfatedom96, x32 is not proper name because there no such arch... x86 is a proper name
19:30:27federico3aha, I got something
19:30:42dom96cheatfate: too bad, we've been using that name for too long
19:30:54dom96And... it's really a minor thing
19:31:25cheatfatedom96, windows 7 x86
19:32:02cheatfatedom96, see this text in installer 'Current PATH length(0) too long to modify in NSIS; set manually if needed'
19:32:25federico3found 2 minor regressions
19:32:47dom96cheatfate: is your PATH super long? :)
19:33:02cheatfateand also dialog box "Error HTTP/1.1 404 Not Found" with buttons Retry, Abort, Ignore
19:33:06cheatfatewhat i must to do?
19:33:52cheatfatedom96, SYSTEM PATH = %SystemRoot%\system32;%SystemRoot%;%SystemRoot%\System32\Wbem;%SYSTEMROOT%\System32\WindowsPowerShell\v1.0\;C:\Program Files\Git\cmd
19:34:20cheatfatedom96, USER PATH is not exist at all... this is my test machine no soft installed so there no PATH by default for User
19:35:21dom96I guess it's completely broken huh? :(
19:36:03cheatfatei don't have vm for x64
19:36:59*brechtm quit (Remote host closed the connection)
19:41:16*Ven_ joined #nim
19:43:04Araq_dom96: is it too late to announce Aporia as "slim editor with compile support" instead of IDE on the website?
19:43:36federico3dom96: aha! nimCoroutines is defined while going through arch.nim during "nim doc2"
19:44:55Araq_the installer adds C:\Nim64test\\bin;C:\Nim64test\\dist\mingw\bin to the path
19:44:59*yglukhov joined #nim
19:45:02Araq_note the double backslashes
19:45:05Araq_that's a bug
19:45:39dom96it's not finding docs.zip
19:45:46dom96federico3: nice
19:45:55dom96Araq_: yes
19:47:13federico3lib/pure/coro.nimcfg actually contains -d:nimCoroutines and nim doc2 is reading it and setting it.
19:47:47dom96Araq_: I installed it successfully, and it changed the path correctly for me too
19:48:54Araq_maybe I had a trailing slash in my path description adn you didn't
19:49:55dom96what I wonder is why it adds ...\dist\babel to the path
19:51:26federico3dom96: anyhow, even if I can set an error in arch.nim that still breaks the build. I would need a "stop processing this file here" pragma
19:52:44dom96wrap the file in a `when defined(nimdoc)`?
19:54:43*Ven_ quit (Ping timeout: 272 seconds)
19:55:45federico3...and I still don't know why nim doc2 is trying to import arch.nim
19:56:26federico3even if I put the whole arch.nim under when not defined(nimdoc), nim doc2 fails later while parsing gc.nim and gc_common.nim ...
19:56:54federico3apparently it's because of that "-d:nimCoroutines" inside lib/pure/coro.nimcfg
19:57:20dom96Araq_: Just FYI I'm fixing the NSIS script
19:57:38Araq_oh ok, then I will proceed with my other points
19:59:24*enthus1ast joined #nim
20:00:59*Ven_ joined #nim
20:06:08*desophos joined #nim
20:06:34federico3and even if we remove it nim doc2 will fail in a different spot
20:09:22*Ven_ quit (Ping timeout: 244 seconds)
20:18:19*enthus1ast quit (Quit: Leaving.)
20:20:47dom96Araq_: What should we do regarding coro doc gen failing?
20:20:48Araq_federico3: fail with what error message?
20:21:02*Ven_ joined #nim
20:22:36*Ven_ quit (Read error: Connection reset by peer)
20:23:16federico3Araq_: the original error was http://pastebin.com/URJhJ9hc
20:24:33*Trustable joined #nim
20:24:43Araq_and now it is?
20:24:48*enthus1ast joined #nim
20:25:23Araq_dom96: docs have no been uploaded so that's why their download fails, I hope you are aware
20:26:00federico3so far the only way to avoid the failure is to remove -d:nimCoroutines from coro.nimcfg and put the whole coro.nim under when not defined(nimdoc):
20:26:06federico3...looks pretty invasive
20:26:26dom96Araq_: yeah
20:29:12dom96grrr, one of GitLab's shared runners is down causing the build to stall
20:31:35*nsf joined #nim
20:35:18*aziz quit (Quit: Ex-Chat)
20:40:52*Ven_ joined #nim
20:43:29*chemist69 quit (Ping timeout: 272 seconds)
20:45:32*Calinou quit (Ping timeout: 244 seconds)
20:47:34*pregressive quit (Read error: Connection reset by peer)
20:50:24*Ven_ quit (Quit: My MacBook has gone to sleep. ZZZzzz…)
20:56:18*Ven_ joined #nim
20:56:35*pregressive joined #nim
20:56:56cheatfatedom96, why windows installer needs administrative rights?
20:57:11cheatfateits not a very good decision because installer don't have signature...
20:57:34dom96because it changes the PATH?
20:57:59dom96also, it silently fails without admin privs
20:58:34cheatfatedom96, UAC will prompt and this prompt will be bad looking because installer is not signed
20:58:55cheatfatealso windows has user path variable which can be changed without administrative privileges
20:59:25cheatfatePATH=[SYSTEM PATH] + [USER PATH]
20:59:32*gokr joined #nim
20:59:50cheatfatesystem path can be modified only with administrative privileges, but user path can be modified freely
20:59:55Araq_cheatfate: yes yes we know
21:00:12Araq_the default installation dest is c:\ because mingw doesn't like spaces in the path
21:00:25Araq_and every other default might have spaces on some version of windows
21:00:38Araq_and that's why it requires admin rights iirc
21:00:45cheatfateAraq_, PROGRA~1?
21:00:53Araq_yeah right
21:00:53dom96oh, is that why the default installation is into C:\Nim?
21:00:56dom96ugh
21:01:07Araq_well mingw evolved
21:01:14Araq_we can try again with spaces :-)
21:01:43dom96UAC doesn't look "ugly" when there is no code signing
21:01:52dom96I've installed plenty of things that weren't signed
21:02:18*mitai quit (Ping timeout: 264 seconds)
21:02:19cheatfatedom96, is it too hard to get certificate for person?
21:02:51dom96I have no idea, but I bet it is
21:03:20flyxgetting a cert for Windows installer sounds definitely lake wasting time that could be spent at better things.
21:03:32flyx*like
21:04:16cheatfatedom96, its same procedure like you want to get certificate for site, only difference - price is much higher
21:05:11flyxeveryone just clicks „yes“ at UAC messages anyway.
21:05:17dom96omg http://i.imgur.com/IrE4unQ.png
21:06:06def-dom96: well, escape the space?
21:06:36cheatfateflyx, not everyone...
21:07:01dom96def-: I take full responsibility: https://github.com/nim-lang/nimble/blob/master/src/nimblepkg/tools.nim#L24
21:07:07*dom96 facepalms
21:07:32*elrood quit (Quit: Leaving)
21:08:26cheatfate:)
21:10:00*chemist69 joined #nim
21:11:11dom96Aporia is also buggy as hell
21:13:26dom96oh right, the file just needs admin privs to be edited
21:14:03*Calinou joined #nim
21:16:34dom96It seems that the x86 installer downloaded a 64bit mingw
21:18:52*Ven_ quit (Quit: My MacBook has gone to sleep. ZZZzzz…)
21:27:20dom96Araq_: do you remember where ${mingw} gets expanded?
21:27:42dom96also, do you specify the nimble version somewhere or does it use the latest tag automagically?
21:28:24Araq_latest tag autom
21:28:42Araq_it is passed via 'koch'
21:29:42dom96okay, I'll tag a new version then
21:30:56Araq_ah fuck
21:30:59Araq_ exec(("tools" / "niminst" / "niminst --var:version=$# --var:mingw=mingw$#" &
21:31:00Araq_ " nsis compiler/installer.ini") % [VersionAsString, $(sizeof(pointer)*8)])
21:31:17Araq_koch uses its own pointer size to guide the NSIS generation :P
21:31:21Araq_how smart
21:31:40Araq_on the other hand. ... our build environment does rebuild koch, rightß
21:31:42Araq_?
21:31:51yglukhovquick thought: instead of "import mylib.a, mylib.b, mylib.c" allow "import mylib.[a, b, c]"
21:32:14Araq_yglukhov: planned once import foo.bar means what we want it to mean
21:32:24yglukhovok
21:33:15dom96That has been on my wishlist for a while, ever since I saw ooc doing it.
21:34:02dom96Araq_: it doesn't rebuild koch
21:34:06*gokr quit (Ping timeout: 244 seconds)
21:34:10dom96Easy fix then
21:34:16Araq_yup
21:34:56dom96Unless the nim built via c sources can't build koch...
21:38:19Araq_well it can build a working compiler with can then build koch
21:41:43Araq_not that koch uses anything fancy
21:41:51Araq_it should always compile
21:42:33dom96okay
21:42:40dom96Everything is pushed
21:42:43dom96CI is running
21:44:05Araq_including my changes?
21:46:06dom96no, but I just want to make sure the nsis installer works ok
21:48:14Araq_I can already update the version and rebuild the docs and upload them, right?
21:49:02dom96sure
21:49:11dom96well
21:49:17dom96I guess we need to fix coro?
21:49:58federico3dom96: is there any starting point for "nim doc" to generate docs for a package? e.g. nim doc *.nim ?
21:50:35dom96ehhh, not entirely sure what you mean
21:50:50dom96By package do you mean a nimble package?
21:50:54federico3yep
21:51:01dom96no
21:51:27Araq_there is a newish --project feature for nim doc2
21:51:37Araq_but I never used it
21:51:48federico3yep, but there's a warning on it
21:52:00Araq_should be easy enough to bring into shape though
21:52:16federico3is nim doc ever following files/dirs by itself or can I just call it with find | xargs ?
21:53:02Araq_as I said, --project or similar generates docs for every module belonging to the package
21:53:16Araq_if you have a module that uses everything in that package
21:53:33Araq_which you kind of should have for testing anyway
21:54:10federico3aha https://github.com/nim-lang/Nim/issues/4027
21:54:52Araq_well I think I implemented that and didn't know about the issue
21:56:55dom96sooo, where are we with tarballs?
21:59:42*brson quit (Ping timeout: 276 seconds)
22:01:54*Demon_Fox quit (Remote host closed the connection)
22:04:52*mitai joined #nim
22:06:10*brson joined #nim
22:07:29dom96ugh, still installed mingw 64 bit
22:08:25*pregressive quit (Remote host closed the connection)
22:09:23*pie__ joined #nim
22:12:20*pie___ quit (Ping timeout: 244 seconds)
22:12:24dom96ugh
22:31:46*xet7 quit (Quit: Leaving)
22:34:45*yglukhov quit (Remote host closed the connection)
22:35:52*bjz quit (Quit: My MacBook Pro has gone to sleep. ZZZzzz…)
22:37:45*yglukhov joined #nim
22:46:01*yglukhov quit (Ping timeout: 240 seconds)
22:51:42*yglukhov joined #nim
22:56:15*yglukhov quit (Ping timeout: 244 seconds)
22:58:49*Jesin quit (Quit: brb, rebooting)
23:01:55*yglukhov joined #nim
23:02:08*bjz joined #nim
23:02:17*Jesin joined #nim
23:03:16*irrequietus quit ()
23:06:39*yglukhov quit (Ping timeout: 276 seconds)
23:07:25*dddddd joined #nim
23:07:51dom96okay, so the installer works now
23:08:15dom96https://gitlab.com/nim-lang/nim/builds/4545023/artifacts/browse/build/
23:09:54dom96Araq_: What else is needed?
23:13:00*yglukhov joined #nim
23:17:01*yglukhov quit (Ping timeout: 240 seconds)
23:21:33*bjz quit (Quit: My MacBook Pro has gone to sleep. ZZZzzz…)
23:23:00*yglukhov joined #nim
23:25:07dom96Release delayed until tomorrow.
23:25:09dom96Good night
23:25:36Araq_yup
23:25:38Araq_good night
23:27:01*yglukhov quit (Ping timeout: 240 seconds)
23:29:06*yglukhov joined #nim
23:34:40*LeNsTR is now known as lenstr
23:36:25*Trustable quit (Remote host closed the connection)
23:43:03*yglukhov quit (Ping timeout: 276 seconds)
23:48:15*yglukhov joined #nim