| 00:01:06 | EXetoC | derived? isn't it all implicit? |
| 00:02:07 | EXetoC | which I think was criticized, but then someone defended that behavior I think |
| 00:02:28 | Araq | no idea what you mean |
| 00:02:44 | Araq | if you speak of some reddit discussion, I barely read those |
| 00:03:09 | filwit | gotta run to the store, be back in a few |
| 00:04:58 | EXetoC | Araq: something about things conforming accidentaly I think |
| 00:09:21 | * | darkf joined #nimrod |
| 00:14:08 | Araq | good night |
| 00:16:01 | EXetoC | bye |
| 01:03:29 | * | Varriount quit (Ping timeout: 240 seconds) |
| 01:28:36 | * | BitPuffin quit (Ping timeout: 245 seconds) |
| 01:42:11 | NimBot | Araq/Nimrod devel 370b74e Araq [+0 ±4 -0]: fixes regression: constant fac4 didn't work |
| 01:43:14 | filwit | Araq is secretly still up and hacking on Nimrod.. |
| 01:43:36 | filwit | :P |
| 01:44:59 | * | DAddYE quit (Remote host closed the connection) |
| 01:45:18 | NimBot | Araq/Nimrod devel 9029763 superfunc [+0 ±1 -0]: Fixed typo's in tutorial 2 |
| 01:45:18 | NimBot | Araq/Nimrod devel 98526cf Andreas Rumpf [+0 ±1 -0]: Merge pull request #876 from superfunc/devel... 2 more lines |
| 01:45:33 | * | DAddYE joined #nimrod |
| 01:46:20 | NimBot | Araq/Nimrod devel ab62556 EXetoC [+0 ±2 -0]: A first attempt at fixing the MongoDB modules. |
| 01:46:20 | NimBot | Araq/Nimrod devel af613f4 Andreas Rumpf [+0 ±2 -0]: Merge pull request #875 from EXetoC/mongo... 2 more lines |
| 01:49:57 | * | DAddYE quit (Ping timeout: 260 seconds) |
| 01:52:29 | Araq | filwit: your PR makes no sense |
| 01:52:44 | * | brson quit (Quit: leaving) |
| 01:52:54 | filwit | okay, how so? what is the correct solution? |
| 01:53:05 | Araq | 'Bar' and 'Baz' happen to have no children, but Foo[Bar] would have a child |
| 01:53:19 | * | brson joined #nimrod |
| 01:53:23 | filwit | one sec, let me look |
| 01:53:37 | Araq | but why should in Foo[Bar] Bar be the matching candidate? makes no sense |
| 01:54:25 | filwit | i'm not sure what you're saying exactly. How do Bar/Baz not have children. and Foo is a module in my example. |
| 01:54:28 | filwit | one sec |
| 01:54:36 | Araq | zahary really needs to looks at this code |
| 01:54:44 | Araq | but I would fix it this way: |
| 01:55:05 | Araq | let toMatch = a |
| 01:55:45 | Araq | sorry, Foo is not a module in my example |
| 01:55:55 | Araq | I need to sleep now, good night |
| 01:56:16 | filwit | night |
| 02:17:57 | * | brson quit (Ping timeout: 260 seconds) |
| 02:23:27 | * | Demos joined #nimrod |
| 02:33:27 | renesac|away | does nimrod have a power operator? |
| 02:33:38 | * | renesac|away is now known as renesac |
| 02:33:58 | renesac | I can't find a good list of defined operators |
| 02:34:16 | Demos | renesac, no idea, try `^` and `**` |
| 02:34:25 | Demos | I think `^` is avalible for overloading |
| 02:35:24 | Demos | does nimrod still have opengl bindings in the standard dist? |
| 02:36:28 | renesac | none of those two are defined |
| 02:36:40 | renesac | seems I have to import math |
| 02:36:51 | renesac | and use pow() |
| 02:37:12 | Demos | you could define ^ if you wanted I guess |
| 02:37:35 | renesac | right, but as I'm using only one time, there is no need for that |
| 02:37:58 | renesac | too bad pow only works with floats... |
| 02:38:23 | renesac | I will have to sprinkle quite a few type conversions here |
| 02:38:32 | Demos | well you could write other versions and add them to the standard lib |
| 02:38:38 | filwit | ^ |
| 02:38:58 | renesac | right, I need to write a todo to list those things |
| 02:39:07 | renesac | :P |
| 02:39:14 | Demos | anyone know if the nimrod-mode.el version on marmelade is the newest |
| 02:39:22 | Demos | M-x org-mode :D |
| 02:42:22 | renesac | templates can't have default arguments? |
| 02:45:17 | renesac | https://gist.github.com/ReneSac/715730359e06c472f683 <-- the error |
| 02:46:01 | * | DAddYE joined #nimrod |
| 02:47:54 | EXetoC | the order is wrong |
| 02:48:27 | Demos | default args must be the last argument |
| 02:48:59 | Demos | speaking of which, how does tuple/map packing and unpacking work. Are there any docs? |
| 02:49:15 | filwit | renesac, make two overloads of timeit |
| 02:50:31 | * | DAddYE quit (Ping timeout: 265 seconds) |
| 02:51:21 | renesac | filwit, yeah, that seems the way |
| 02:53:04 | renesac | but still ugly, as I have to repeat all over the default values for the arguments |
| 02:54:00 | renesac | Demos, the stmt must be the last value, so that isn't an option... |
| 02:54:49 | Demos | hm, annoying... maybe someday there will be special rules for templates |
| 02:54:59 | renesac | not only for templates |
| 02:55:07 | renesac | for functions accepting functions too |
| 02:55:28 | renesac | that must have them as the last arguments to use the do: notation |
| 02:55:29 | EXetoC | that's not a problem |
| 02:55:35 | EXetoC | yes |
| 02:56:22 | renesac | when using the do notation, let me forget the previous optional arguments |
| 02:56:37 | filwit | renesac: make internal const for the default values then |
| 02:56:50 | EXetoC | the implementer should deal with that |
| 02:56:50 | renesac | it is like I do "last_argument_name=func()" |
| 02:56:59 | EXetoC | anyway, it's not the end of the world |
| 02:57:18 | renesac | yeah, but ugly |
| 02:57:20 | renesac | ^" |
| 02:57:24 | Demos | maybe you could write a macro that generates templates to emulate default args |
| 02:57:44 | Demos | granted that would be kinda confusing code.. |
| 02:57:58 | renesac | I'm more inclined to fiddle with the grammar to fix it once for all |
| 02:58:33 | renesac | though this seems complex |
| 03:05:40 | * | DAddYE joined #nimrod |
| 03:09:54 | renesac | hum, the problem is not the grammar, but the way do: blocks are translated to the AST |
| 03:34:41 | * | ics quit (Quit: My MacBook Pro has gone to sleep. ZZZzzz…) |
| 04:07:27 | Demos | if I have proc foo(a: int) = var a = a ... is that saying "I want to copy a and then shadow it"? |
| 04:09:04 | fowl | Demos, ye |
| 04:09:15 | Demos | sweet, is it idomatic? |
| 04:09:22 | Demos | dare I say nimrodic |
| 04:09:51 | fowl | dunno |
| 04:11:05 | Demos | it seems strange that java/c#'s arrays have a fixed length. I mean if everything is heap allocated anyway and the arrays are already carrying around their length it would be reasonable to just make them "vectors" by default |
| 04:50:31 | * | Varriount joined #nimrod |
| 04:50:55 | Varriount | Sorry about getting disconnected, I had a power outage |
| 04:57:36 | * | dmac joined #nimrod |
| 05:02:02 | * | ics joined #nimrod |
| 05:05:05 | Varriount | reactormonk: I had a look at your cartesian product code, very nice! |
| 05:06:06 | Varriount | reactormonk: I hope you don't mind if I submit a PR adding comments and lengthening some variable names (I was also able to add a couple lines that shaved off some time) |
| 05:17:05 | * | Demos quit (Read error: Connection reset by peer) |
| 05:26:31 | * | xtagon quit (Read error: Connection reset by peer) |
| 05:41:15 | * | ddl_smurf quit (Quit: ddl_smurf) |
| 06:01:55 | * | BitPuffin joined #nimrod |
| 06:06:45 | * | BitPuffin quit (Ping timeout: 265 seconds) |
| 06:13:51 | reactormonk | Varriount, go for it |
| 06:20:36 | * | CarpNet quit (Ping timeout: 245 seconds) |
| 06:21:00 | * | CarpNet joined #nimrod |
| 06:42:19 | * | ics quit (Quit: My MacBook Pro has gone to sleep. ZZZzzz…) |
| 07:00:54 | * | CarpNet quit (Ping timeout: 245 seconds) |
| 07:01:20 | * | CarpNet joined #nimrod |
| 07:12:43 | * | filwit quit (Ping timeout: 260 seconds) |
| 08:19:26 | * | dmac quit (Ping timeout: 245 seconds) |
| 08:27:02 | * | DAddYE quit (Remote host closed the connection) |
| 08:54:13 | * | _davidk_ joined #nimrod |
| 09:28:52 | * | DAddYE joined #nimrod |
| 09:33:18 | * | DAddYE quit (Ping timeout: 265 seconds) |
| 10:29:31 | * | DAddYE joined #nimrod |
| 10:34:23 | * | DAddYE quit (Ping timeout: 272 seconds) |
| 11:08:21 | * | psquid joined #nimrod |
| 11:08:49 | * | Mordecai quit (Ping timeout: 245 seconds) |
| 11:11:19 | * | BitPuffin joined #nimrod |
| 11:19:00 | * | awestroke joined #nimrod |
| 11:55:42 | * | awestroke quit (Ping timeout: 265 seconds) |
| 11:55:59 | * | awestroke joined #nimrod |
| 11:59:36 | Araq | hi awestroke welcome |
| 12:02:45 | * | ddl_smurf joined #nimrod |
| 12:05:48 | * | eigenlicht quit (Ping timeout: 252 seconds) |
| 12:08:19 | * | eigenlicht joined #nimrod |
| 12:12:34 | * | eigenlicht quit (Ping timeout: 246 seconds) |
| 12:17:11 | * | eigenlicht joined #nimrod |
| 12:25:09 | * | eigenlicht quit (Ping timeout: 248 seconds) |
| 12:27:37 | * | eigenlicht joined #nimrod |
| 12:30:49 | * | DAddYE joined #nimrod |
| 12:35:31 | * | DAddYE quit (Ping timeout: 265 seconds) |
| 12:45:41 | * | eigenlicht quit (Ping timeout: 245 seconds) |
| 13:04:15 | * | darkf quit (Quit: Leaving) |
| 13:31:34 | * | DAddYE joined #nimrod |
| 13:33:18 | * | radsoc joined #nimrod |
| 13:36:26 | * | DAddYE quit (Ping timeout: 265 seconds) |
| 14:12:18 | * | [1]Endy joined #nimrod |
| 14:36:28 | OrionPK | Demos you can resize arrays |
| 14:36:52 | OrionPK | oh that was a long time ago... |
| 14:36:55 | * | OrionPK has no conception of time |
| 14:42:46 | * | eigenlicht joined #nimrod |
| 14:46:56 | * | noam joined #nimrod |
| 15:32:52 | * | DAddYE joined #nimrod |
| 15:37:16 | * | DAddYE quit (Ping timeout: 245 seconds) |
| 15:38:39 | * | aftershave joined #nimrod |
| 15:46:20 | awestroke | Araq: hello hello |
| 16:15:19 | * | BitPuffin quit (Quit: WeeChat 0.4.2) |
| 16:18:24 | * | psquid quit (Quit: work) |
| 16:22:03 | * | ics joined #nimrod |
| 16:33:32 | * | DAddYE joined #nimrod |
| 16:34:05 | * | ics quit (Quit: My MacBook Pro has gone to sleep. ZZZzzz…) |
| 16:38:33 | * | DAddYE quit (Ping timeout: 272 seconds) |
| 16:44:39 | * | brihat left #nimrod (#nimrod) |
| 16:48:34 | * | awestroke quit (Read error: Connection reset by peer) |
| 16:48:48 | * | awestroke joined #nimrod |
| 17:01:55 | * | Demos joined #nimrod |
| 17:22:43 | * | awestroke quit (Remote host closed the connection) |
| 17:29:07 | * | brson joined #nimrod |
| 17:34:10 | * | DAddYE joined #nimrod |
| 17:38:45 | * | DAddYE quit (Ping timeout: 248 seconds) |
| 17:53:41 | * | brson quit (Ping timeout: 248 seconds) |
| 17:54:39 | * | Demos quit (Ping timeout: 245 seconds) |
| 18:00:45 | * | brson joined #nimrod |
| 18:09:28 | dom96 | Ahh so those bots were from the GCHQ not the NSA. I was almost right: https://www.quakenet.org/articles/102-press-release-irc-networks-under-systematic-attack-from-governments |
| 18:10:52 | EXetoC | :o |
| 18:11:35 | * | xtagon joined #nimrod |
| 18:13:04 | * | aftershave quit (Quit: Textual IRC Client: www.textualapp.com) |
| 18:24:35 | * | DAddYE joined #nimrod |
| 18:34:04 | * | _davidk_ left #nimrod ("Leaving") |
| 18:45:41 | Varriount | Araq: When I use repr() on a a newly created sequence of integers with uninitialized entries, how does repr() know the difference bettween a 0 and a nil? |
| 18:46:11 | dom96 | Value types cannot be nil. |
| 18:53:39 | EXetoC | right. it just means that they aren't explicitly given a value, so it could be anything |
| 18:54:54 | dom96 | No, it's always initialised to binary 0. |
| 18:56:05 | EXetoC | doesn't sound like uninitialized to me, but ok |
| 18:57:55 | EXetoC | ok so he just got the semantics wrong |
| 19:05:18 | * | icebattle joined #nimrod |
| 19:05:29 | * | brson quit (Ping timeout: 272 seconds) |
| 19:05:50 | * | brson joined #nimrod |
| 19:10:11 | * | BitPuffin joined #nimrod |
| 19:10:28 | * | Demos joined #nimrod |
| 19:11:23 | EXetoC | is `&=` supposed to be exclusively for strings? should we just stick to 'add'? |
| 19:11:45 | Demos | well & works on seqs |
| 19:12:36 | EXetoC | yes, & but not &= |
| 19:13:16 | Araq | EXetoC: please add it |
| 19:13:35 | Araq | also &= should be an alias for 'write' |
| 19:18:04 | EXetoC | Araq: for every overload of add and write? |
| 19:18:32 | Araq | nah, just make it generic |
| 19:18:52 | Araq | proc &= [T](f: TFile, x: T) = write(f, x) |
| 19:18:58 | EXetoC | still for every overload then, just with less repetition |
| 19:19:09 | Araq | well also for sockets |
| 19:19:20 | Araq | alias for 'send' then |
| 19:20:28 | * | delian66 joined #nimrod |
| 19:20:37 | * | aftershave_ joined #nimrod |
| 19:24:45 | EXetoC | will do. I'd prefer not to have 'add' then because of the overlap, but it's used everywhere |
| 19:25:04 | Varriount | Araq: It's amazing how shallow copying can speed something up |
| 19:25:38 | Varriount | I spent ~2 hours last night, comparing my version of the cartesian product iterator with reactormonk's, trying to find why his was faster |
| 19:26:22 | Varriount | What made his faster was the use of a constant container and shallowCopy() |
| 19:26:32 | Demos | yeah getting the correct semantics for copying/memory layout/ownership is supah important |
| 19:27:10 | Demos | also, I thought nimrod did shallow copies by default as far as refs are concerned |
| 19:28:28 | EXetoC | or have only 'add' and `&`. |
| 19:36:29 | Demos | if I have a template foo(t: typedesc): stmt = var `t blarg`: int what is the name I should expect when calling like foo(int)? would I get an intblarg? |
| 19:38:06 | EXetoC | I think you need a macro |
| 19:38:30 | Demos | oh dear god... macros are really hard to write |
| 19:39:20 | Demos | the code I posted compiles |
| 19:39:36 | Demos | I just don't know what the actual name of the symbol is going to be, and it is hard to figure that out |
| 19:39:41 | dom96 | I disagree. Macros are pretty easy. |
| 19:39:50 | EXetoC | `t blarg`? |
| 19:40:31 | Demos | accodring the the manual the backticks can be used to generate names using the stuff you were passed |
| 19:40:36 | dom96 | Demos: Call it and see. The compiler will give you a warning about an unused variable. |
| 19:40:48 | EXetoC | but, there's whitespace there, so I don't know if it works |
| 19:41:03 | dom96 | It does. Personally I dislike that syntax. |
| 19:41:21 | EXetoC | yeah, just tested it |
| 19:41:48 | EXetoC | dom96: ugly, but sometimes necessary |
| 19:41:51 | Demos | ah it just calls it "typSceneNode" |
| 19:42:44 | Demos | or rather tblarg |
| 19:43:20 | * | filwit joined #nimrod |
| 19:43:26 | EXetoC | oh. yeah, apparently |
| 19:44:17 | Demos | is there a way to get the name of a type as it appears in the source? |
| 19:44:22 | EXetoC | it sure is hacky, but I don't know why you need it |
| 19:44:26 | Demos | oh, wait |
| 19:44:28 | Demos | I know |
| 19:44:30 | EXetoC | maybe you should document that, because it's not obvious |
| 19:44:37 | Demos | yeah |
| 19:45:34 | Demos | actually it may just be a bug |
| 19:45:50 | Demos | if I replace t: typedesc with t:expr I still get tblarg |
| 19:47:05 | EXetoC | maybe the constraint is the only difference |
| 19:47:28 | EXetoC | where the former is more restrictive |
| 19:49:55 | EXetoC | Araq: should &= be preferred then? |
| 19:50:23 | EXetoC | when invoking |
| 19:55:52 | Araq | EXetoC no ... |
| 19:55:59 | Araq | we're doing it ruby style |
| 19:56:02 | Araq | aliases ftw |
| 19:57:08 | dom96 | So much for consistency... |
| 20:01:34 | Varriount | Araq: Once again, I thank you for having a templating feature |
| 20:02:26 | Araq | you're welcome |
| 20:06:46 | EXetoC | Araq: sarcasm or not, I'm confused |
| 20:06:51 | * | Demos quit (Ping timeout: 245 seconds) |
| 20:06:55 | * | brson quit (Ping timeout: 272 seconds) |
| 20:07:45 | EXetoC | CSS sure has a lot of stuff |
| 20:07:59 | Araq | EXetoC: no sarcasm this time |
| 20:11:11 | Varriount | Araq: Does return() do anything in an iterator? |
| 20:11:34 | EXetoC | Araq: what purpose will it serve then? to aid in writing generic interfaces and to reduce noise when it really matters? |
| 20:11:53 | Araq | EXetoC: generic interface |
| 20:12:39 | dom96 | Varriount: You were right. Adding a custom field to the overlapped structure is useful, and in fact necessary. |
| 20:13:35 | Varriount | dom96: I'm glad that you found that link useful. |
| 20:15:28 | Varriount | Also, Araq, is an object yielded from an iterator copied? |
| 20:15:50 | EXetoC | Araq: ok that's fine |
| 20:16:03 | Araq | Varriount: yes |
| 20:16:10 | EXetoC | I think &= in basic{2,3}d should be *= |
| 20:17:46 | Varriount | Hm. Ok, return isn't allowed, so is there a was to stop an iterator early, without resorting to booleans and if statements? |
| 20:19:13 | Araq | Varriount: 'return' workswith closure iterators |
| 20:20:00 | Araq | EXetoC: that code is used in a commercial project irrc, so don't touch it for now |
| 20:21:18 | EXetoC | ok |
| 20:22:08 | EXetoC | Araq: a 'put' trait seems like a natural progression then, whenever that becomes a possibility |
| 20:28:52 | Varriount | Araq: How serious is a closure iterator gc bug? |
| 20:29:31 | Araq | Varriount: I will mark it in red ... |
| 20:30:03 | Varriount | Araq: It's probably a more obscure bug, since my iterator is rather complicated. |
| 20:42:52 | reactormonk | Varriount, and I'm a strong believer of the functional style ;-) |
| 20:43:00 | reactormonk | Varriount, how much faster btw? |
| 20:44:13 | Varriount | reactormonk: Well, the best time out of 100 rounds, with a thousand loops each round, was about 0.08 seconds for the initial version |
| 20:44:27 | Varriount | Sometimes it was a bit higher, 0.1 seconds |
| 20:44:32 | reactormonk | Varriount, how much _faster_, not how fast |
| 20:44:43 | Varriount | The revised version took 0.07 seconds |
| 20:44:59 | reactormonk | slap some more in there, you want at least 1s runtime ;-) |
| 20:45:21 | Varriount | I know it's not *that* much faster, but I couldn't see how to optimize it any more. |
| 20:45:41 | reactormonk | I'll take that as a compliment ;-) |
| 20:45:44 | Varriount | Also, I appear to be running into a codegen bug when trying to turn it into an iterator |
| 20:46:00 | reactormonk | huh? |
| 20:47:33 | Varriount | https://gist.github.com/Varriount/8852176 |
| 20:47:55 | Varriount | reactormonk: The generated C code isn't being generated properly, there are syntax errors. |
| 20:48:18 | Varriount | reactormonk: I also added an ordering option for results. |
| 20:48:50 | * | [1]Endy quit (Ping timeout: 252 seconds) |
| 20:48:50 | Varriount | Araq: The link I posted causes the error |
| 20:49:25 | Varriount | It seems to have something to do with the garbage collector. |
| 20:49:25 | reactormonk | Varriount, issue |
| 20:49:40 | Varriount | reactormonk: And what do I call the issue? |
| 20:49:54 | Varriount | "Mysterious bug from planet x"? |
| 20:50:07 | reactormonk | probably |
| 20:50:09 | reactormonk | # Handle corner cases |
| 20:50:10 | reactormonk | if groupsLen == 0 or groupsLen == 1: |
| 20:50:12 | reactormonk | raise newException(EOutOfRange, "Invalid args") |
| 20:50:14 | reactormonk | |
| 20:50:16 | reactormonk | ^ not sure about that one... |
| 20:50:40 | Varriount | reactormonk: What do you mean? |
| 20:50:50 | reactormonk | I'd make it a condition, aka something you can capture, but doesn't interrupt the flow and just use my flows |
| 20:51:04 | reactormonk | otherwise you force people to handle corner cases, whereas empty input just means empty output |
| 20:51:09 | reactormonk | ... which is mathematically defined. |
| 20:51:28 | Varriount | reactormonk: I had "return" before, however I was trying to get the iterator to work. |
| 20:53:27 | reactormonk | Varriount, in case of 0 elements, don't yield at all. In case of 1 element, yield the element. |
| 20:56:47 | Araq | Varriount: how can it both generate wrong C and cause GC issues? |
| 20:57:45 | Varriount | Araq: The error in the generated C is around a "asgnRefNoCycle" call |
| 20:58:47 | NimBot | Araq/Nimrod devel 40015dd Simon Hafner [+0 ±1 -0]: removed unittest from talgorithm |
| 21:04:08 | Varriount | Araq, reactormonk : Updated the gist |
| 21:04:09 | reactormonk | Araq, do you approve an indexBy[T, U](x: seq[T], index: proc(x: T): U):TTable[U,T] ? |
| 21:05:07 | reactormonk | Varriount, is it faster with shallowcopy? |
| 21:05:26 | Varriount | reactormonk: Well, yes. However I don't have to do that with an iterator. |
| 21:05:39 | reactormonk | Varriount, oh, right. |
| 21:05:43 | Varriount | Since the yielded item is copied anyway |
| 21:06:07 | Varriount | reactormonk: And in the updated procedure version of product(), I preallocate all the space needed for the result. |
| 21:06:27 | reactormonk | Varriount, cute |
| 21:06:29 | Varriount | So I don't need to use shallowcopy for that either. |
| 21:07:10 | Araq | Varriount: report the bug please |
| 21:07:49 | Varriount | Araq: Ok. |
| 21:08:39 | Varriount | reactormonk: I particularly like the usage of templates to help implement the ascending/descending result order |
| 21:09:01 | reactormonk | can I use indexBy[T, U, V](x: V[T], index: proc(x: T): U):TTable[U,T] - where V is a container that can be iterated over? |
| 21:09:10 | reactormonk | aka, defines items. |
| 21:10:14 | reactormonk | Varriount, hm, they're basically procs with scope here, but I guess it's faster here |
| 21:11:04 | reactormonk | and make stuff rather readable, which can't be said of my version |
| 21:11:18 | Varriount | reactormonk: Which was my primary goal :p |
| 21:11:27 | Araq | reactormonk: V[T] is not supported yet |
| 21:13:03 | reactormonk | Araq, so how would I go define it? |
| 21:13:32 | filwit | Araq: so it seems my macro bug is happening somewhere in vmgen:genIf:305, dest is 10 when gen is called, and the preceding unused check fails (if dest >= 0) |
| 21:13:35 | reactormonk | Varriount, btw, if you make an issue, please paste the code in there, don't link it. github cleans up gists. |
| 21:14:07 | Varriount | reactormonk: Really? So what is their expiration date? |
| 21:14:23 | reactormonk | Varriount, no idea, but I once looked up an old one and it was expired |
| 21:14:35 | filwit | Araq: i'm not really understanding how dest is used. I understand it's a register index, but why does unused check if it's above 0 then? |
| 21:14:46 | Varriount | reactormonk: The gist should last long enough for this particular bug |
| 21:15:25 | filwit | off-topic: never played the game, but this song is great: http://www.youtube.com/watch?v=8yCyysCmhlQ |
| 21:15:42 | Varriount | Also, I can't find anything on the github website/s to suggest that gists expire. |
| 21:15:53 | reactormonk | Varriount, kk |
| 21:16:27 | reactormonk | Araq, hmm, template? |
| 21:16:29 | Varriount | I have a gist from three years ago. |
| 21:16:54 | Araq | reactormonk: just use V instead of V[T] |
| 21:17:11 | Araq | T is deduced from the proc then. big deal. :P |
| 21:17:41 | reactormonk | Araq, cool |
| 21:20:28 | * | eigenlicht quit (Read error: Operation timed out) |
| 21:20:31 | * | radsoc quit (Ping timeout: 250 seconds) |
| 21:23:52 | Varriount | Araq: I don't suppose you could submit an organization application to the Google Summer of Code? |
| 21:25:12 | Varriount | Then I could have a reason to work on Nimrod all summer long |
| 21:26:41 | * | eigenlicht joined #nimrod |
| 21:28:43 | Araq | Varriount: I can try ... |
| 21:29:12 | Varriount | Araq: https://www.google-melange.com/gsoc/homepage/google/gsoc2014 |
| 21:32:35 | * | eigenlicht quit (Ping timeout: 246 seconds) |
| 21:35:46 | * | eigenlicht joined #nimrod |
| 21:40:46 | dom96 | GSOC sounds like a lot of work for the organizers. |
| 21:46:08 | dom96 | Hrm, "College". My school has "College" in the name. |
| 21:46:12 | dom96 | I wonder if I could apply. |
| 21:48:46 | reactormonk | Araq, that's one week. But if you manage to get through, you can get our own coding pet ;-) |
| 21:49:19 | Araq | good night |
| 21:49:40 | dom96 | Doesn't look like it :\ |
| 21:49:47 | reactormonk | dom96, http://pastie.org/8706267 here's the agreement |
| 21:50:06 | dom96 | "post secondary school" means post-high-school |
| 21:50:26 | reactormonk | you are not a resident or national of Cuba, Iran, Syria, Sudan, North Korea, or Myanmar (Burma); |
| 21:52:21 | dom96 | I could be a mentor though heh |
| 21:52:26 | reactormonk | dom96, sure |
| 21:52:40 | reactormonk | (a) To provide a publicly published list of project ideas from which students participating in Google Summer of Code (“Students”) will choose a project to work on for the duration of Google Summer of Code; |
| 21:52:45 | reactormonk | (b) To make available a person or persons to evaluate Student suggestions for projects should those students not find something that appeals to them on the published list and to either approve or reject those suggestions; |
| 21:52:49 | reactormonk | (c) To make available a person or persons to review the incoming Student applications targeted to the Mentor Organization and to decide which applications should be accepted; |
| 21:52:53 | reactormonk | (d) To make available a person or persons responsible for helping Students integrate with the Mentor Organization’s community; |
| 21:52:57 | reactormonk | (e) To make available a person or persons to monitor the progress of the students and mentor them as the project proceeds ("Mentor(s)"); |
| 21:53:01 | reactormonk | (f) To make available one or more alternate Mentors in the event an existing Mentor is unable to continue providing guidance to the accepted Student applicant; |
| 21:53:11 | reactormonk | .... and the written evaluation |
| 21:53:24 | dom96 | Yeah, sounds like a lot of work. |
| 21:53:34 | reactormonk | yup |
| 21:53:40 | dom96 | And all the mentoring organisation gets is $500 |
| 21:53:46 | reactormonk | and code. |
| 21:54:37 | dom96 | true. |
| 21:54:47 | reactormonk | phI||Ip, Araq mentioned he had a proper fix for the PR somewhen in this chat. |
| 21:55:11 | filwit | why you write my name so odd.. :P |
| 21:55:19 | * | aftershave_ quit (Quit: Textual IRC Client: www.textualapp.com) |
| 21:55:44 | reactormonk | filwit, just do it like everyone else and use the same nick everywhere |
| 21:55:44 | filwit | he already fixed it better, and it works now with recent changes to devel, so i closed |
| 21:56:00 | filwit | i try, but i can't usually |
| 21:56:03 | reactormonk | phI||Ip, wrong person. sorry. |
| 21:56:13 | filwit | oh, okay |
| 21:56:28 | filwit | i recently closed a PR, so thought you where talking to me about that |
| 21:56:41 | reactormonk | filwit, nah, I was. |
| 21:56:56 | filwit | oh, i see |
| 21:57:12 | reactormonk | I just hl'd the guy again and mention "wrong person" so he can go back to sleep |
| 21:57:28 | filwit | yeah i always tried to use "F i L" places (that's my nic on phoronix, steam, etc) but a lot of places don't like the spaces |
| 21:58:05 | reactormonk | dom96, now the only problem is to get Araq to sign up for that, or do it yourself. |
| 21:58:17 | reactormonk | filwit, do it R style and use dots |
| 21:58:19 | filwit | so i chose PhilipWitte (my real name) for things that didn't like the space (youtube, github, etc) until i realized 'filwit' was probably better |
| 21:58:24 | reactormonk | F.i.L |
| 21:58:36 | reactormonk | nope, freenode doesn't like it. |
| 21:58:55 | filwit | meh, filwit (Philip Witte = F i L Wit = filwit) is simpler to type anyways |
| 21:59:43 | reactormonk | :> |
| 21:59:46 | filwit | i really don't care about internet anonymity, lol |
| 22:01:21 | reactormonk | filwit, pseudonymity you say |
| 22:01:30 | reactormonk | you can look up my name pretty fast |
| 22:02:03 | filwit | you can look up almost anyone pretty fast unless they don't use tech at all |
| 22:02:12 | reactormonk | yup |
| 22:03:47 | filwit | anyone know the what a 'var T' enum looks like in the VM? nkVarTy? |
| 22:05:16 | filwit | trying that.. think it might work |
| 22:05:44 | reactormonk | lib/system.nim(1627, 7) Error: undeclared identifier: 'data' |
| 22:05:46 | reactormonk | muh |
| 22:06:30 | filwit | fixit |
| 22:06:50 | reactormonk | just did |
| 22:07:28 | * | filwit gives cookie |
| 22:07:55 | filwit | (internet cookie with virus that is) |
| 22:10:53 | EXetoC | executable cookies? |
| 22:11:40 | filwit | only if you're on Win95 of earlier |
| 22:11:45 | filwit | or* |
| 22:12:17 | EXetoC | nah I'm not |
| 22:12:36 | EXetoC | you? |
| 22:12:40 | filwit | gah, i fixed my macro bug, but Araq says it's not right... dunno why though. contemplating making a PR anyways |
| 22:13:05 | filwit | EXetoC: i use the penguin these days :) |
| 22:13:23 | filwit | though i have to stay in Win 8.1 a lot for work |
| 22:15:32 | filwit | Anyone here know much about how the VM works? |
| 22:16:34 | reactormonk | filwit, nimrod vm? |
| 22:16:43 | filwit | yes |
| 22:17:05 | filwit | specifically vmgen:genIf, concerning 'dest' register indexes |
| 22:17:59 | filwit | i can't figure out why the dest variable at vmgen:305 is set to 10 when 'gen' gets called on that line |
| 22:18:45 | filwit | but gen immediately does 'unused' for the node kinds it gets from genIf, and unused checks if dest is >= 0, which obviously fails when it's 10 |
| 22:19:49 | filwit | i'm not sure what's going on. I though dest would be a positive index, but apparently not, so i'm trying to understand what exactly it's values are used for |
| 22:19:59 | filwit | thought* |
| 22:21:06 | filwit | my "fix" for the bug basically entails expanding all assert/unused checks for my specific test if/else macro test case, and obviously that's not right |
| 22:21:34 | * | ddl_smurf quit (Quit: ddl_smurf) |
| 22:21:44 | filwit | but it works if i do that, so I'm assuming there's just a bug with how dest is being set in genIf, and I need to fix it there (then the asserts will work) |
| 22:27:10 | * | Demos joined #nimrod |
| 22:36:09 | * | io2 joined #nimrod |
| 22:39:30 | Varriount | filwit: If you want an easy-ish bug to fix, get modules with unicode names working. |
| 22:39:49 | Varriount | Currently they cause C compiler errors, because of the identifiers. |
| 22:41:27 | filwit | Varriount: not really trying to fix random bugs right now. I need this to be fixed so my macros work with devel |
| 22:42:28 | Varriount | Macros which do..? |
| 22:43:36 | reactormonk | Varriount, http://pastie.org/private/u49hwpmco7ht1lxrykn8w looks reasonable? |
| 22:43:59 | Demos | honetly, macros and templates are quite hard to deal with because there are a lot of compiler bugs and a lot of invalid code that is rejected with terrible errors |
| 22:44:11 | reactormonk | Demos, so you say basically C++? |
| 22:44:53 | Demos | no, c++ errors are long, but in the end tell you where you screwed up. Nimrod says either SIGSEGV or like a one line error that does not help you much |
| 22:44:59 | Demos | or just "Internal compiler error" |
| 22:45:06 | filwit | Varriount: OOP stuff mostly. they build types/procs based on an incoming body of code. But they'll also connect to an editor generated code and mix stuff together |
| 22:45:55 | filwit | Varriount: i'll be making a stripped down version for a standard-lib proposal, and a unique version for my game engine. |
| 22:49:18 | * | whatelse joined #nimrod |
| 22:54:34 | Demos | hm there is documentation in macros.nim that is not on the devel docs... are the devel docs just of master? |
| 22:56:31 | NimBot | Araq/Nimrod devel b23e47d Simon Hafner [+1 ±1 -0]: indexBy, which indexes a collection into a hashtable |
| 22:59:45 | reactormonk | Error: named expression expected |
| 22:59:47 | reactormonk | huh? |
| 22:59:50 | reactormonk | x.inputs.incl(TItem(typeID: id, amount = tbl[id].amount + input.amount)) |
| 23:01:22 | EXetoC | this: "amount ="? |
| 23:01:48 | EXetoC | a = instead of : |
| 23:01:49 | * | brson joined #nimrod |
| 23:07:10 | * | DAddYE_ joined #nimrod |
| 23:08:42 | * | BitPuffin quit (Quit: WeeChat 0.4.2) |
| 23:10:57 | * | DAddYE quit (Ping timeout: 260 seconds) |
| 23:15:39 | * | brson quit (Ping timeout: 272 seconds) |
| 23:16:09 | * | ddl_smurf joined #nimrod |
| 23:19:13 | * | brson joined #nimrod |
| 23:23:13 | reactormonk | EXetoC, thanks |
| 23:29:39 | Varriount | reactormonk: Yes, however if these kind of abstract types are repeatedly used in algorithms/sequtils, I would define them seperately, in a type block |
| 23:30:46 | Varriount | reactormonk: See my unfinished nimrod version of python's itertools -> https://gist.github.com/Varriount/8854701 |
| 23:37:02 | EXetoC | doesn't that usage of typedesc result in unnecessary verbosity? |
| 23:37:22 | * | xenagi joined #nimrod |
| 23:38:55 | * | xenagi quit (Client Quit) |
| 23:39:20 | * | brson quit (Quit: leaving) |
| 23:40:01 | * | brson joined #nimrod |
| 23:40:22 | EXetoC | untested I guess |
| 23:43:31 | * | xenagi joined #nimrod |
| 23:51:02 | * | DAddYE_ quit (Remote host closed the connection) |
| 23:51:38 | * | DAddYE joined #nimrod |
| 23:53:06 | * | darkf joined #nimrod |
| 23:53:39 | filwit | eating a mug full of Reeses Puff cereal.. finally found an American food I can boast about. |
| 23:55:26 | * | io2 quit (Ping timeout: 264 seconds) |