<< 07-04-2018 >>
00:16:45*arnetheduck_ joined #nim
00:16:45*arnetheduck quit (Read error: Connection reset by peer)
00:25:02*xkapastel quit (Quit: Connection closed for inactivity)
00:31:18*aguspiza quit (Ping timeout: 256 seconds)
01:07:54*S1t1Schu joined #nim
01:11:27*S1tiSchu quit (Ping timeout: 240 seconds)
02:31:12*arnetheduck_ quit (Remote host closed the connection)
02:33:16*arnetheduck joined #nim
02:39:09*Snircle joined #nim
02:52:19*athenot quit (Remote host closed the connection)
02:52:35libman_re "COBOL backend" joke - as much as I hate Java, it would be the most logical backend to add, because Android. But I think Nim should focus on doing fewer things better.
02:52:58*athenot joined #nim
03:00:44*ldlework quit (Ping timeout: 276 seconds)
03:05:44*ldlework joined #nim
03:41:25*dddddd quit (Remote host closed the connection)
03:51:44*Lord_Nightmare2 joined #nim
03:52:42*Lord_Nightmare quit (Ping timeout: 260 seconds)
03:52:45*Lord_Nightmare2 is now known as Lord_Nightmare
04:08:44*S1tiSchu joined #nim
04:12:03*S1t1Schu quit (Ping timeout: 260 seconds)
04:21:07*SitiSchu joined #nim
04:21:57*S1tiSchu quit (Ping timeout: 264 seconds)
04:36:57*arnetheduck quit (Ping timeout: 240 seconds)
04:41:49*arnetheduck joined #nim
05:16:12*vlad1777d joined #nim
05:29:07*yglukhov joined #nim
05:33:43*yglukhov quit (Ping timeout: 265 seconds)
05:33:57*vlad1777d quit (Ping timeout: 240 seconds)
05:58:44*ldlework quit (Changing host)
05:58:44*ldlework joined #nim
06:35:01*nsf joined #nim
06:36:56*r3d9u11 joined #nim
07:06:38*rokups joined #nim
07:09:50*max3 quit (Quit: Connection closed for inactivity)
07:12:48*Snircle quit (Quit: Textual IRC Client: www.textualapp.com)
07:36:29*miran joined #nim
07:38:29*madmalik joined #nim
07:48:48Araqhttps://www.joelonsoftware.com/2000/08/09/the-joel-test-12-steps-to-better-code/
07:49:03Araq"Do you fix bugs before writing new code?"
07:56:04arnetheduckis it considered a bug when async generates bogus / dead code? https://pastebin.com/fNricQcw - note the extra call to complete after `goto beforeret`
07:57:32arnetheduckthis is from compiling https://github.com/nim-lang/Nim/blob/72d9485e8e0b839c3832ecb3b949e116fe9ed062/tests/async/tasync_traceback.nim..
08:03:03*gmpreussner quit (Ping timeout: 260 seconds)
08:03:26*athenot quit (Ping timeout: 255 seconds)
08:05:26*gmpreussner joined #nim
08:17:41FromGitter<data-man> @Araq: Do you have scripts or parameters to convert sqlite by c2nim?
08:21:55*r3d9u11 quit (Ping timeout: 265 seconds)
08:26:23*xet7 joined #nim
08:31:01arnetheduckalso, afaict it seems that the snippet `(*(*colonenvP_).colonup_).retFuture1` is generated from the same node in the ast - ie the same node instance is reference in two places in the generated AST (leading to it having the same id - or two parents) - is this / should this be allowed in macros
08:31:05arnetheduck?
08:33:32*xet7 quit (Remote host closed the connection)
08:35:13*xet7 joined #nim
08:37:48FromGitter<alehander42> so, one of my side project ideas is a smaller nim linter
08:38:19FromGitter<alehander42> I already have some simple checks working, but I'd like some opinions on e.g. what would be the most useful checks to implement
08:41:00*aguspiza joined #nim
08:42:03FromGitter<nathandaly> Hey, I want to change the foreground colour of terminal echo, there is a setForgroundColor in the terminal module but it expects file as the first param
08:42:23FromGitter<k0pernicus> Checked the issues, and wondered if the `easy` tag is for newbies?
08:42:51FromGitter<nathandaly> In bash it's a hex/escii code wrapped around the string
08:43:37FromGitter<data-man> @nathandaly: Use styledEcho
08:44:13FromGitter<mratsim> @k0pernicus in the stdlib I would say yes, in the compiler I would say it’s a good first issue for someone experimented otherwise but wants to do a first dive in the compiler
08:44:50FromGitter<alehander42> I think stdout is the file, but yeah styledEcho should be ok
08:45:02FromGitter<data-man> @alehander42: About #7508 ⏎ The current mangling schema does not allow to reconstruct the original name. Your PR also does not improve the situation. ⏎ I've some ideas for a new schema that will allow a demangling.
08:45:08FromGitter<nathandaly> Ah ok thanks for the tip
08:45:34FromGitter<Vindaar> @nathandaly @alehander42 yeah, in general if the `terminal` procs expect a file and you just want to print to the terminal, `stdout` is correct
08:45:59FromGitter<nathandaly> Ah ok ty
08:46:04FromGitter<alehander42> @data-man when is a variable *not* represented as `original_hash` ?
08:46:16FromGitter<alehander42> `original_name & hash` *
08:46:29FromGitter<data-man> Without hashing
08:48:33FromGitter<alehander42> I mean, isn't the original name almost always `mangled.rsplit('_', 1)[0]`
08:48:42FromGitter<alehander42> currently ?
08:49:07FromGitter<alehander42> (the part between the last `_`)
08:49:11FromGitter<alehander42> before*
08:51:47FromGitter<k0pernicus> Thanks @mratsim ;-)
08:52:44FromGitter<data-man> @alehander42: We need something like this: https://github.com/dlang/dmd/blob/master/src/dmd/dmangle.d
08:53:44FromGitter<alehander42> @data-man still, can you give me one example when the original variable name is not == `mangled.rsplit('_', 1)[0]`
08:54:37FromGitter<alehander42> I've used that for a long time for functions/variables , even with a mapping with name => operator and it always seemed enough
08:57:09FromGitter<alehander42> afaik Araq had some reasons why he wanted different mangling scheme than D
08:57:58FromGitter<data-man> Hashing is slow and increase binary file
08:58:23FromGitter<alehander42> but maybe it was for the older D scheme
08:59:34FromGitter<alehander42> @data-man that seems true ⏎ but still I disagree that one can reconstruct the original name, in my PR that part should be working fine in most(all?) cases (I have to just fix the split a bit)
08:59:48*bauk joined #nim
09:00:50FromGitter<alehander42> now, if you're talking about deconstructing arg types for functions from the mangled name, that's something I've wondered about
09:01:35FromGitter<alehander42> which leads me to another question related to my nim-linter idea, is it possible to get some type info for a module currently using the compiler API
09:01:45FromGitter<data-man> https://github.com/dlang/druntime/blob/master/src/core/demangle.d
09:03:56*bauk left #nim (#nim)
09:05:07FromGitter<alehander42> (tagging @Araq on this question too) what I did in past was passing a source through parse and sem and then analyzing somehow modules / methods in ModuleGraph, but I remember they didn't have sufficient type info in them
09:05:52FromGitter<alehander42> I think @zah told me before one needs to add something to the compiler if I want to somehow "export" types for outside analysis
09:07:42FromGitter<k0pernicus> I am trying to understand the behaviour of `spawn` and `parallel`. ⏎ If I understood the manual, the `spawn` behavior (without `parallel`) is to pass the task to the thread pool. ⏎ Or, using `parallel` and `spawn`, the task will be processed using another core, right?
09:08:13FromGitter<data-man> @alehander42: I want to add more info for -d:debugSigHashes
09:10:50FromGitter<alehander42> how does `debugSigHashes` work? I cant find docs
09:11:10FromGitter<data-man> Undocumented :-D
09:11:33FromGitter<alehander42> :D but what is the difference with normal sighashes
09:12:23FromGitter<data-man> Sighashes stored to db
09:12:44FromGitter<data-man> Very slow
09:13:16FromGitter<alehander42> in a db? does the db include other info corresponding to them?
09:13:59FromGitter<data-man> For types only
09:15:27FromGitter<k0pernicus> Also, can we pass anonymous as argument for `spawn`? ⏎ Instead of `spawn my_func(x)`, just write `spawn proc (x: uint32) = …` ?
09:16:27FromGitter<data-man> If store for any sym, then this db can be used for autocompletion, e.g.
09:18:01FromGitter<alehander42> @data-man sounds interesting, i'll test it later
09:18:09FromGitter<alehander42> so i just need to pass this option
09:18:19FromGitter<alehander42> and it will create the db in the same dir ? or nimcache
09:19:52FromGitter<data-man> In compiler dir (for me)
09:20:39FromGitter<data-man> compiler sources
09:37:44*SenasOzys quit (Read error: Connection reset by peer)
09:38:13*SenasOzys joined #nim
10:09:03FromGitter<k0pernicus> For this code, ⏎ ⏎ ```code paste, see link``` ⏎ ⏎ I have the following error: `expression [… ] is of type ‘FlowVar[…]’ and has to be discarded` [https://gitter.im/nim-lang/Nim?at=5ac898bf109bb0433298f8fb]
10:10:14FromGitter<dom96> Spawn returns a special value that can be used to tell if the thread finished executing
10:10:19FromGitter<k0pernicus> I don’t understand this error. ⏎ Does `spawn` returns anything? ⏎ Or do I need to return something in my anonymous procedure?
10:11:10FromGitter<dom96> You can assign the value that spawn returns to a variable, or discard it using the discard keyword
10:11:19FromGitter<k0pernicus> Thanks @dom96, but why I don’t have this error on the following code: ⏎ ⏎ ```proc print_parallel(r: uint32) = ⏎ for x in 0..r: ⏎ spawn my_proc(x)``` ⏎ ⏎ ? [https://gitter.im/nim-lang/Nim?at=5ac89947270d7d3708a0c21f]
10:11:39FromGitter<k0pernicus> with `proc my_proc(x: uint32) = echo x`
10:12:10FromGitter<dom96> Hrm. Maybe there is some issue passing an anonymous proc to spawn
10:12:40FromGitter<k0pernicus> Ok
10:13:07FromGitter<k0pernicus> I will create an issue about that - maybe there is something wrong in my code...
10:13:10FromGitter<k0pernicus> Thanks @dom96 ;-)
10:14:31dom96maybe you could gist all your code?
10:14:36dom96that might show the problem
10:14:42FromGitter<k0pernicus> Oh right, sorry
10:17:13FromGitter<k0pernicus> This is the gist: https://gist.github.com/k0pernicus/0870a3975091cbeac16b5ec9b27bc95d
10:18:39FromGitter<k0pernicus> (Forgot to import the `threadpool` module in the gist, sorry for that)
10:20:52dom96No worries :)
10:22:06dom96yeah, report it
10:22:17dom96If I add a discard I get an internal compiler error :)
10:22:26FromGitter<k0pernicus> Same thing here, yep
10:22:39FromGitter<k0pernicus> Ok, I will create an issue about that
10:24:11dom96thanks!
10:27:59FromGitter<k0pernicus> Reported here: https://github.com/nim-lang/Nim/issues/7535
10:30:33FromGitter<gogolxdong> @shashlick Have you tried transforming webkit to nim?
10:31:39dom96k0pernicus: nice, you can use Nim syntax highlighting on GitHub too :)
10:32:23FromGitter<k0pernicus> @dom96 Updated ;-) Thanks for the tip
10:33:18dom96but yeah, I wouldn't be surprised if anon procs are disallowed in that context for simplicity :)
10:34:24FromGitter<k0pernicus> Ok… Or… Do I need to call the anonymous procedure?
10:34:38FromGitter<k0pernicus> like `(proc (x: uint32) = echo x)()` ?
10:35:56dom96hm, perhaps
10:36:11FromGitter<k0pernicus> Oh yeah - I gave the procedure pointer, instead of calling it -_-
10:36:14FromGitter<k0pernicus> It works fine now
10:36:18FromGitter<k0pernicus> Sorry for that...
10:36:34dom96error could be better :)
10:36:40dom96and you did find a crash
10:36:49FromGitter<k0pernicus> Yep, it’s true...
10:37:11FromGitter<k0pernicus> I will update the issue
10:40:22FromGitter<k0pernicus> @dom96 Do you want me to change the title ?
10:40:42dom96no, it's fine
10:45:39*Vladar joined #nim
10:57:00*SitiSchu quit (Read error: Connection reset by peer)
11:06:47federico3is it just my nimble or https://github.com/watzon/github-api-nim cannot be imported after nimble install?
11:12:09dom96what's your import line?
11:17:31federico3urgh: const a = "foo"; assert a[0..^0][^1] == '\0'
11:18:37federico3dom96: import github_api , as in the readme
11:19:12dom96well, look at the source
11:19:18dom96of the package
11:19:19dom96that can't work
11:20:30federico3the sources contain a github_api directory with various files in it - indeed it looks wrong but I was wondering why the author uses that
11:24:16dom96Seems the author prefers crystal now :/
11:31:38FromGitter<data-man> @dom96: Which sqlite3 version in dlls.zip? (the latest version is 3.23.0)
11:31:56dom96no idea
11:33:41*sleepyqt joined #nim
11:37:47FromGitter<data-man> !eval import sqlite3; echo $libversion_number()
11:37:50NimBot3011000
11:38:00*djellemah joined #nim
11:38:03FromGitter<data-man> :(
11:39:52YardanicoThis is not windows dll version though
11:40:06Yardanico@data-man this is sqlite3 version installed in docker container on nim playground
11:40:18Yardanico(NimBot uses nim playground for eval)
11:40:32federico3any way to perform autovivification in JSON? e.g. create a tree with myjson["a"]["b"]["c"] = value
11:40:55FromGitter<data-man> I know. It's also bad.
11:41:08Yardanico@data-man: you can make a PR to nim playground :)
11:43:45FromGitter<data-man> I have not finished a new wrapper :)
11:48:21federico3a{"a", "b", "c"} = newJInt(2) # works!
12:01:59FromGitter<zacharycarter> sorry
12:02:03FromGitter<zacharycarter> what's up with the playground?
12:02:11FromGitter<zacharycarter> I'm working on a new version of the project now so any feedback is welcome
12:02:36*xkapastel joined #nim
12:02:46FromGitter<data-man> @zacharycarter: Old sqlite3
12:03:38dom96data-man: why does that matter?
12:14:44FromGitter<data-man> @dom96: For the playground maybe it's not important. What matter in software updating?
12:17:46FromGitter<alehander42> @data-man so I used the flag
12:18:02FromGitter<alehander42> And I couldn't see a db file
12:18:12FromGitter<alehander42> What kind of name I am looking for
12:18:53FromGitter<alehander42> I can just look through the flag source actually
12:19:17FromGitter<data-man> sighashes.db
12:21:25*Trustable joined #nim
12:21:41*xet7 quit (Remote host closed the connection)
12:22:47FromGitter<data-man> But there's only the type name and hash :)
12:23:31*xet7 joined #nim
12:27:11FromGitter<zacharycarter> I can update the sqlite3 version on the sandbox image
12:27:15FromGitter<zacharycarter> shouldn't be an issue
12:28:26FromGitter<alehander42> huh
12:28:33FromGitter<alehander42> i got the nim forum email in my spam :(
12:28:49FromGitter<alehander42> finally a member tho
12:30:43FromGitter<alehander42> ahhhh
12:30:50FromGitter<alehander42> I have to compile the compiler itself with that flag
12:30:52dom96hrm, turns out I was wrong about the cause of that memory leak
12:30:53FromGitter<alehander42> makes sense
12:33:42FromGitter<alehander42> ok, that's pretty useful
12:33:47FromGitter<alehander42> but very slow indeed
12:34:38FromGitter<alehander42> actually I am almost sure with non-db one-write at the end of compilation it can be nice
12:36:29FromGitter<alehander42> I've tried similar stuff with db patching the compiler and I've always found storing it in internal data and spewing it out in some reasonable format at the end seems much faster
12:37:17*madmalik quit (Quit: Connection closed for inactivity)
12:39:55FromGitter<data-man> @alehander42: I replaced sighashes.db to ```:memory:```
12:41:29FromGitter<alehander42> how much seconds does it take for a small file?
12:41:36FromGitter<alehander42> we can benchmark, I am replcaing it with a table
12:45:18*vegax87 quit (Changing host)
12:45:18*vegax87 joined #nim
12:45:18*vegax87 quit (Changing host)
12:45:18*vegax87 joined #nim
12:45:23FromGitter<alehander42> why does addQuitProc not accept ` <proc (){.locks: 0.}>`
12:46:09FromGitter<data-man> ``````
12:47:16FromGitter<alehander42> bravo
12:47:19FromGitter<alehander42> wow, indeed
12:52:06FromGitter<alehander42> @data-man can you access a memory-db after the compilation finished
12:53:19FromGitter<alehander42> it takes a min 30 for me with file db
12:53:55FromGitter<alehander42> no, 1 minute
12:54:13FromGitter<alehander42> and 1 second with table + text save
12:54:23FromGitter<alehander42> I guess csv to sqlite will take max 1-2 secs more
12:55:37FromGitter<data-man> @alehander42: That's why I want to use sqlite3_backup
12:55:57FromGitter<alehander42> but I just don't get why one needs sqlite3 for this
12:56:18FromGitter<alehander42> you need it to query the data after compilation has ended, yes
12:56:30FromGitter<alehander42> but it's overkill to insert it during compilation
12:59:26FromGitter<data-man> There is a check for hashes uniqueness
13:00:31FromGitter<alehander42> which you can do with tables :D
13:01:24FromGitter<data-man> for sqlite's db there is sqlitebrowser :)
13:01:39FromGitter<alehander42> but that's my point
13:02:12FromGitter<alehander42> you collect it, in the end you DO save your data in sqlite, but in only one try
13:04:04FromGitter<alehander42> maybe with memory sqlite + backup this would be the same speed, we have to compare
13:06:49FromGitter<alehander42> that can fix some of my typing info needs, that's why I am excited about it :D
13:09:17FromGitter<zacharycarter> not sure if this is relevant or not to your guys's discussion but
13:09:25FromGitter<zacharycarter> https://github.com/bluenote10/NimData
13:10:42FromGitter<alehander42> wow interesting
13:12:53FromGitter<alehander42> @data-man it seems it will all work in <2-3 secs
13:12:58FromGitter<alehander42> for a small program on my machine
13:13:23FromGitter<alehander42> collect info => dump to sqlite currently with temp csv, index by hash
13:13:57FromGitter<alehander42> but I have to benchmark on compiling bigger programs
13:14:03FromGitter<data-man> @alehander42: NimData is great library. But the compiler can't depend on it.
13:14:42FromGitter<alehander42> @data-man I don't use nimdata
13:16:58FromGitter<alehander42> I'll post a commit in a minute
13:17:15FromGitter<data-man> @alehander42: I thought that your 'wow interesting' was about this lib :)
13:17:29FromGitter<alehander42> :D it is interesting in itself , not for this case
13:19:20FromGitter<narimiran> @zacharycarter @bluenote10 said he is also working on more traditional (column based) lib: https://github.com/bluenote10/kadro
13:20:13FromGitter<data-man> @alehander42: How about zero_like_db? ;)
13:20:15*SenasOzys_ joined #nim
13:22:37*SenasOzys quit (Ping timeout: 260 seconds)
13:25:50*yglukhov joined #nim
13:25:51FromGitter<alehander42> if you're talking about zero_functional, these days @michael72 is mostly developing/maintaining it :D good thing your remind me
13:26:10*yglukhov quit (Remote host closed the connection)
13:26:20FromGitter<alehander42> yeah one can do a simple "data structures easily saveable as db-s" thing sounds nice
13:26:47*yglukhov joined #nim
13:27:40FromGitter<data-man> Of course, I'm a generator of ideas. :-D
13:27:50FromGitter<alehander42> :D idea-man
13:28:46FromGitter<data-man> It's too late to change the nickname.
13:32:53dom96Anybody on Linux/Windows want to see if this is reproducible? https://github.com/nim-lang/Nim/issues/7532#issue-312147042
13:33:02dom96Note: It allocates memory quickly so be ready to kill it.
13:33:22*yglukhov quit (Ping timeout: 264 seconds)
13:34:12FromGitter<mratsim> cute names: retFuture255001
13:34:31dom96macro code :D
13:34:39FromGitter<mratsim> https://imgs.xkcd.com/comics/wisdom_of_the_ancients.png
13:40:28Yardanicodom96, I'll try now
13:40:56dom96thx
13:41:31Yardanicodom96, yes, it allocates memory really fast
13:41:43dom96Linux right?
13:41:46Yardanicoyes
13:42:33dom96great
13:44:24FromGitter<alehander42> @data-man https://github.com/alehander42/Nim/commit/abc969e755c261657e5dc9b82ef9a17d17ae00a8
13:44:31FromGitter<alehander42> this seems to work quickly for me
13:45:21FromGitter<alehander42> there must be a way to directly dump a lot of rows into sqlite3 with the API but not sure how
13:47:35*JacobEdelman quit (Ping timeout: 256 seconds)
13:47:54*JacobEdelman joined #nim
13:49:05dom96trying out this dumpNumberOfInstances feature
13:49:06dom96[Heap] string: #56852; bytes: 9833659
13:49:23dom96I guess that means there are 56,852 strings allocated
13:50:38FromGitter<data-man> @alehander42: Nice! The general idea is to save more information from PSym.
13:51:28FromGitter<alehander42> yeah, for example it seems this doesn't save info for *all* hashes
13:53:03*miran quit (Ping timeout: 260 seconds)
13:54:23FromGitter<alehander42> do you know why? e.g. where should I add code if I want to save the type for function name hashes
13:55:42FromGitter<alehander42> the forum editor is nice , full markdown support, didn't expect it
14:00:12FromGitter<data-man> To hashTree proc
14:01:36*r3d9u11 joined #nim
14:04:43FromGitter<alehander42> why a custom syntax for links
14:04:45FromGitter<alehander42> why
14:04:51FromGitter<alehander42> nimforum
14:05:15FromGitter<alehander42> ah they're directly detected
14:05:16FromGitter<alehander42> weird
14:05:22dom96"custom"?
14:05:27dom96It's restructured text
14:05:51dom96As for rst vs markdown, you can argue with Araq about that :)
14:05:59FromGitter<alehander42> ah right
14:06:05FromGitter<alehander42> I didn't even notice :D
14:12:01*xkapastel quit (Quit: Connection closed for inactivity)
14:12:02*arecaceae quit (Read error: Connection reset by peer)
14:12:27*arecaceae joined #nim
14:20:25*jaco60 joined #nim
14:46:46shashlickgogolxdong: I haven't actually looked at webkit - what exactly are you trying to do?
14:47:12FromGitter<gogolxdong> I am trying to build a nimkit :)
14:49:53FromGitter<gogolxdong> wrote a .cfg file ,doesn't work. in chrome source code WebKit/Source/
14:50:15FromGitter<gogolxdong> D:\third_party\src\github.com\chromium\chromium\third_party\WebKit\Source
14:51:52FromGitter<gogolxdong> native nim dom node manuplation.
14:52:00shashlickcan you share your cfg
14:53:37FromGitter<gogolxdong> D:\third_party\src\github.com\chromium\chromium\third_party\WebKit\Source\core\dom
14:53:52FromGitter<gogolxdong> ```code paste, see link``` [https://gitter.im/nim-lang/Nim?at=5ac8db80080a38505308eb4f]
14:53:53*r3d9u11 quit (Ping timeout: 256 seconds)
14:54:21FromGitter<gogolxdong> probably non-sense.
14:54:47shashlickwell, which header file are you trying to wrap?
14:55:10FromGitter<gogolxdong> all under dom/
14:57:10shashlickokay and do you end up with a dom.dll or something when webkit is built?
14:58:04FromGitter<gogolxdong> didn't build wekit, just cloned the source codes
14:59:14FromGitter<gogolxdong> there is an empty nimkit directory.
14:59:20*nsf quit (Quit: WeeChat 2.0.1)
14:59:40shashlickwell, you need to tell nimgen which header files to work on
15:00:08shashlickthere's a variety in this dom directory - which one do you include? which functions are useful for you?
15:01:37FromGitter<gogolxdong> I am trying to migrate the whole dom part to nim.
15:04:18FromGitter<gogolxdong> I remembered you gave me a libvirt transformation cfg file.
15:05:57shashlickya but you were going to use libvirt with the DLL files, here you want to convert to Nim which is completely different
15:06:45FromGitter<gogolxdong> you mean unfeasible through nimgen?
15:08:01shashlickwell, I don't know if c2nim can convert C++ code 100% so you'd need to run through each file and hand translate things c2nim couldn't do
15:08:08shashlicknimgen just automates c2nim
15:09:29shashlickI also looked at my Chrome install, I don't see any webkit.dll or dom.dll or anything - it's a 50MB chrome.dll and 75MB chrome_child.dll - probably everything compiled in
15:09:31FromGitter<gogolxdong> How about show me an example of Node.h?
15:10:13FromGitter<gogolxdong> I pulled the whole source code of chrome.
15:11:06FromGitter<gogolxdong> and yes ,c2nim cannot convert 100%
15:11:30shashlicklet me get something started and share with you
15:15:16*dddddd joined #nim
15:18:08shashlickgogolxdong: are you saying you want to wrap the c++ code and use procs in Nim or convert the c++ code into Nim?
15:22:40FromGitter<gogolxdong> convert
15:23:44FromGitter<gogolxdong> leave me message ,have to go to bed , thanks in advance.
15:24:51shashlickokay will share what I come up with for Node.h
15:25:46*hal joined #nim
15:32:03halHello. I want to add a new proc toFILETIME(unix: int64, ns: int): FILETIME to the winlean module. How do I identify the correct place in winlean.nim to put that proc?
15:36:11Yardanicodom96, btw, did you try different GC's?
15:36:16Yardanicomark&sweep, gc-v2, boehm?
15:36:19dom96no
15:47:12haldom96: winlean has proc rdFileTime*(f: FILETIME): int64, would that be a good place to add toFILETIME? It is the reverse operation.
15:47:46hallooking at rdFileTime should the name of the new proc be toFileTime (instead of toFILETIME)?
15:47:47dom96hal: You can just put it anywhere in winlean
15:47:50dom96it doesn't matter much
15:48:07dom96usually new things are added to the bottom
16:05:48*littleli quit (Remote host closed the connection)
16:16:38shashlickgogolxdong: suddenly a bunch of stuff is missing from github.com - the WebKit/Source directory disappeared
16:19:13shashlickweechat is working super nicely - have irc, gitter and matrix all flowing in, and in #nim, messages from gitter and discord are being automatically fixed to show up from the nickname instead of FromGitter or FromDiscord
16:22:36*xkapastel joined #nim
16:40:43*leorize quit (Ping timeout: 265 seconds)
16:42:19*leorize joined #nim
16:48:13haldom96: GULPF: may I ask you to have a look at my current toFILETIME https://github.com/hendrikgit/Nim/commit/de8bb136aa9063ec6b46cc603ae898fe5e58765d
17:09:02*max3 joined #nim
17:11:04halNevermind, it needs a lot of changes I think. Don't look at it yet.
17:11:54FromGitter<samdmarshall> Hey, so i'm having a weird problem with the `httpclient` class; i'm trying to write a small webpage validator to lint my website by traversing links on the page. one of the links ends up being a redirect, that gets redirected a couple of times. I know the maxredirect of the client class is 5, but so far it just seems to hang indefinitely??
17:12:02FromGitter<samdmarshall> ```code paste, see link``` [https://gitter.im/nim-lang/Nim?at=5ac8fbe26d7e07082bcc4b7d]
17:13:38dom96interesting
17:14:43dom96ahh yes
17:14:47dom96I thought I fixed this :(
17:15:06FromGitter<Yardanico> @samdmarshall just FYI: httpclient is a module, not a class :) and HttpClient is an object
17:16:22dom96samdmarshall: I pushed a fix
17:16:38dom96https://github.com/nim-lang/Nim/commit/992300b30057e2b3b489b90fb0e7011607e3e8cf
17:17:42FromGitter<samdmarshall> oh sweet, let me build dev and give it a go
17:18:32dom96well, I made it stop going into an infinite loop
17:18:38dom96There might still be some other problem :)
17:20:20FromGitter<samdmarshall> that's fine, if it doesn't go into that infinite loop then I can start debugging the rest
17:21:50FromGitter<samdmarshall> but in this case, that was actually all my code was doing, performing a head request then looking at the response code. but since it was trying to handle the redirect automatically, I couldn't look at that first response code I got
17:40:40FromGitter<samdmarshall> @dom96 thanks! that worked perfectly.
17:41:07dom96great :)
17:48:19*SenasOzys_ quit (Quit: Leaving)
17:48:39*SenasOzys joined #nim
17:50:05*rockcavera joined #nim
17:55:07*SenasOzys quit (Read error: Connection reset by peer)
17:55:30*SenasOzys joined #nim
18:03:56*SenasOzys quit (Quit: Leaving)
18:04:20*SenasOzys joined #nim
18:07:22*SenasOzys quit (Client Quit)
18:07:40*SenasOzys joined #nim
18:12:22*sleepyqt quit (Ping timeout: 264 seconds)
18:19:18*miran joined #nim
18:28:13halWhat style of calling procs, I mean dot or bracket like t.toUnix or toUnix(t), is preferred for the standard library?
18:30:03dom96the former
18:31:37*r3d9u11 joined #nim
18:34:30*r3d9u11 quit (Remote host closed the connection)
18:46:16haldom96: I found out that the handle returned by findFirstFile is not valid for SetFileTime, I get error 6 invalid handle. I'll have to use CreateFile to get the handle after all
18:46:42dom96that's weird, are you sure you're passing it correctly?
18:53:26halI'm doing that the same way as in getLastModificationTime, so I would think so yes. Windows api documentation says FindFirstFile returns a search handle and some file info, that is what is used for getLastModificationTime. CreateFile returns a handle that can also be used to modify.
18:57:07halI need to do something else now but I'll try to make it work with CreateFile and get back to you. By the end of this I might actually know how to use the win api :)
19:10:43dom96ahh cool
19:10:50dom96Best way to learn about the inner workings of Windows :)
19:24:17*dddddd quit (Ping timeout: 260 seconds)
19:28:53*sleepyqt joined #nim
19:31:04*shashlick quit (Read error: Connection reset by peer)
19:31:36*shashlick joined #nim
19:34:51*rokups quit (Quit: Connection closed for inactivity)
19:43:06*dabbb joined #nim
19:51:43*dddddd joined #nim
20:06:11*sleepyqt quit (Quit: Leaving)
20:15:37*dabbb quit (Quit: Page closed)
20:23:57*aguspiza quit (Ping timeout: 245 seconds)
20:26:08YardanicoAraq, what is your opinion about syntax skins? will you remove them?
20:41:16*SenasOzys quit (Remote host closed the connection)
20:41:34*aguspiza joined #nim
20:41:40*SenasOzys joined #nim
21:20:46*craigger quit (Ping timeout: 264 seconds)
21:21:43*craigger joined #nim
21:23:29*jaco60 quit (Quit: My MacBook has gone to sleep. ZZZzzz…)
21:28:16*miran quit (Quit: Konversation terminated!)
21:30:09*Vladar quit (Quit: Leaving)
21:32:21*nsf joined #nim
21:51:44*dddddd quit (Remote host closed the connection)
22:33:47*nsf quit (Quit: WeeChat 2.0.1)
22:41:33*aguspiza quit (Ping timeout: 260 seconds)
22:43:30*rockcavera quit (Remote host closed the connection)
22:44:25*xet7 quit (Quit: Leaving)
22:58:03FromGitter<Varriount> hal: Ugh, getLastModificationTime needs to be reimplemented. FindFirstFile is not meant to be used like that.
23:02:59FromGitter<Varriount> Besides, getFileInfo is better
23:08:26dom96Make a PR ;)
23:23:25FromGitter<diegogub> hello, Is it possible to cross-compile from OSX to Linux?
23:28:37FromGitter<gogolxdong> @shashlick , I have the code ,how do I give you?
23:30:20shashlickOfficial code is at https://chromium.googlesource.com/chromium
23:30:31*SenasOzys quit (Remote host closed the connection)
23:30:33shashlickBut I don't see the same directory structure there
23:30:54*SenasOzys joined #nim
23:34:19FromGitter<gogolxdong> Is that a coincidence?WebKit/Source was removed couple of hours ago.
23:36:16FromGitter<gogolxdong> (https://files.gitter.im/nim-lang/Nim/VCqk/image.png)
23:45:50FromGitter<gogolxdong> https://gist.github.com/gogolxdong/0bbd3bc8a7369b3d3c5d01257f9a2977#file-gistfile1-txt Node.h
23:47:49FromGitter<gogolxdong> Node.cpp https://gist.github.com/gogolxdong/03624d33f7094f10fcc0e74a55c04e14
23:56:40shashlickI know, I was working on the files and it suddenly disappeared
23:58:46shashlickIs it any different from the main webkit source?
23:59:07shashlickhttps://github.com/WebKit/webkit/tree/master/Source/WebCore/dom
23:59:10FromGitter<gogolxdong> not sure, I can make a git repository.