| 00:00:35 | * | superfunc joined #nimrod |
| 00:02:54 | superfunc | dom96: how did your finals go man? |
| 00:03:32 | dom96 | superfunc: could've gone better, but went well enough I guess, thanks. Still got one more exam left though! |
| 00:04:12 | superfunc | dom96: lol, that's pretty much how I felt about mine afterwards |
| 00:05:32 | dom96 | It's annoying realising the silly mistakes you make right after the exam, but humans are prone to such mistakes unfortunately heh. |
| 00:06:03 | Varriount | dom96: That's why I shove any exam-related thought out of my head after I've taken the exam. |
| 00:06:07 | superfunc | yeah, I yolo'd my OS exam super hard |
| 00:06:49 | * | DAddYE quit (Remote host closed the connection) |
| 00:13:03 | Varriount | dom96: Regarding https://github.com/Araq/Nimrod/issues/1151 , isn't that what 'is' does? |
| 00:13:35 | dom96 | Varriount: It doesn't work for two typedescs |
| 00:14:47 | Varriount | On another topic, how does one do explicit value comparisons, vs address comparisons? |
| 00:15:06 | Varriount | If addr x == addr y? |
| 00:16:15 | EXetoC | dom96: so, patch 'is' then? |
| 00:16:34 | dom96 | EXetoC: I dunno. Ask Araq. |
| 00:16:34 | EXetoC | rather than adding == |
| 00:20:18 | dom96 | 'night |
| 00:26:51 | superfunc | later |
| 00:27:48 | superfunc | can a typeclass define an iterator? |
| 00:28:22 | flaviu1 | Yes, but I'm not sure how well it works |
| 00:28:35 | superfunc | time to find out lol |
| 00:28:44 | flaviu1 | the manual says to add `for item in x: item is MyType` |
| 00:29:02 | flaviu1 | But procedure style might work too, IDK |
| 00:31:41 | superfunc | didn't see that, thanks for pointing it out |
| 00:41:35 | superfunc | it is way to damn hot to code here |
| 00:42:22 | superfunc | 93° |
| 00:45:56 | * | Jesin quit (Quit: Leaving) |
| 00:46:12 | Varriount | I don't understand why 'is' is needed - couldn't it be replaced with 'type(foo) == FooType'? |
| 00:46:25 | flaviu1 | subtyping too |
| 00:46:38 | Varriount | Hm, good point. |
| 00:46:52 | Varriount | flaviu1: But what if you want strict typing comparison? |
| 00:47:17 | flaviu1 | I don't really know, maybe FooType == BarType? |
| 00:47:46 | EXetoC | has anyone actually needed that yet? |
| 00:47:59 | Varriount | Not really. |
| 00:48:05 | EXetoC | superfunc: take of your clothes and sit on some ice |
| 00:48:16 | EXetoC | cus the world depends on our nimrod code. get to work |
| 00:54:27 | * | joelmo quit (Quit: Connection closed for inactivity) |
| 01:09:32 | * | superfunc quit (Quit: Page closed) |
| 01:10:03 | * | superfunc joined #nimrod |
| 01:27:31 | * | freezerburnv joined #nimrod |
| 01:29:32 | * | hoverbear joined #nimrod |
| 01:36:49 | * | bjz joined #nimrod |
| 01:37:55 | * | superfunc quit (Ping timeout: 246 seconds) |
| 01:39:18 | * | q66 quit (Quit: Leaving) |
| 01:54:08 | zshazz | BTW, Araq, I just need to say that for working with random numbers you really need unsigned ints. It's not bad to implement their support properly in an RNG interface since they are, in fact, necessary for correct code (or, at least, from what I can see) |
| 01:54:43 | zshazz | (that's in response to a comment you made awhile ago) |
| 01:55:24 | zshazz | signed ints simply can't accomplish everything unless you give some nice things up or you do very strange things that make maintanence a nightmare |
| 01:57:39 | * | brson quit (Quit: leaving) |
| 01:58:05 | zshazz | for instance, if you want a random number in range[ low(IntType)+2 .. high(IntType)-2 ] ... it can't be done without an unsigned int reliably (or, at least, I cannot fathom a reasonable to read algorithm that can) |
| 01:59:02 | * | brson joined #nimrod |
| 01:59:25 | flaviu1 | He's asleep right now, but he reads the logs |
| 01:59:52 | zshazz | Yeah, I suspected so, but I figured I'd say it anyway |
| 02:01:43 | * | brson_ joined #nimrod |
| 02:01:44 | zshazz | He has always seemed really against the idea of unsigned ints (understandably so, because they're a common source of bugs). Basically I dismissed nimrod ~2 years ago because of its lack of unsigned int support and the insistence that it was unnecessary to support. Plus it seems like he still holds the belief that they shouldn't be supported (but at least some progress has apparently been made) |
| 02:05:18 | * | brson quit (Ping timeout: 240 seconds) |
| 02:05:36 | EXetoC | I don't recall him saying that they shouldn't even be supported |
| 02:07:25 | zshazz | To a question asked about how to make my code support uints properly for a client, he stated "tell the client he's a fool" ... If that's not a vehement denial of support, I've no idea what is. |
| 02:09:42 | EXetoC | I don't know what he was on about. the goal is to discourage using them, and not to actually omit support for some operations |
| 02:11:23 | flaviu1 | Better than java, at least |
| 02:11:44 | zshazz | *shrug* I seem to always remember him posting negative things about uints but I've never seen anything from him admitting that they're necessary. Perhaps he's concentrating too hard on "discouraging" them. |
| 02:12:03 | zshazz | flaviu1, Hell yes. |
| 02:18:36 | EXetoC | he claims that signed integers are generally better, and that people often use them even when it's not necessaryy |
| 02:19:22 | EXetoC | what is it that doesn't work? |
| 02:19:48 | flaviu1 | Java has just signed ints |
| 02:20:37 | flaviu1 | The only unsigned data type is a char, and thats completely useless |
| 02:20:53 | flaviu1 | for arithmetic, I mean. Also, no bitshifts |
| 02:27:03 | zshazz | EXetoC: Pretty much the example I gave ... I can't think of a reasonable/readable way to make a random number in the range [ low(IntType)+2 .. high(IntType)-2 ] without the unsigned version of IntType. There probably is some way if you don't mind relying on overflow behavior (but that's much uglier than using a uint) |
| 02:28:09 | flaviu1 | uhh, cast the bits? |
| 02:28:25 | zshazz | I'd like a readable, reliable way to cast to the unsigned version of the signed type |
| 02:29:45 | zshazz | Right now I've got a bunch of `when sizeof(IntType) == 8: ...` things to make it work, but an official `unsigned(IntType)` would be nice (I guess I'll end up writing that myself, though) |
| 02:30:50 | flaviu1 | What about a macro? |
| 02:30:58 | EXetoC | a template seems fine |
| 02:31:11 | zshazz | Yeah, I had a template in mind. |
| 02:32:04 | zshazz | though I think it could be done even with a generic proc, so that's what I'll use first |
| 02:32:51 | EXetoC | nope can't return a type from one |
| 02:33:18 | EXetoC | or can you? with compileTime. I wouldn't do it though |
| 02:35:13 | zshazz | It probably would work if I'm doing something like `var dist = unsigned(high(range) - low(range))` (though, we're right back to overflow behavior but it's probably unavoidable in this case) |
| 02:35:30 | zshazz | and by `range` I mean `aRange` |
| 02:36:00 | * | brson_ quit (Quit: leaving) |
| 02:36:14 | * | brson joined #nimrod |
| 02:36:57 | EXetoC | I don't think the type can be bound in the body |
| 02:37:08 | Varriount | zshazz: I just want to say, thank you for all the effort. A good PRNG library is quite handy. |
| 02:37:11 | EXetoC | anyway, looks good. bye |
| 02:38:16 | zshazz | Varriount: No problem. I'll probably post what I have tomorrow (and put it on babel, of course). I wish I had more to show you guys but I'm still learning the language as I go :) |
| 02:49:23 | * | saml_ quit (Quit: Leaving) |
| 02:53:54 | * | kemet joined #nimrod |
| 03:09:13 | * | brson quit (Quit: leaving) |
| 03:10:22 | * | def-_ joined #nimrod |
| 03:13:56 | * | def- quit (Ping timeout: 252 seconds) |
| 03:18:59 | * | kemet quit (Ping timeout: 252 seconds) |
| 03:28:36 | * | zshazz_ joined #nimrod |
| 03:32:33 | * | zshazz quit (Ping timeout: 252 seconds) |
| 03:37:26 | * | superfunc joined #nimrod |
| 03:39:49 | * | freezerburnv quit (Quit: freezerburnv) |
| 03:40:21 | * | Skrylar joined #nimrod |
| 03:43:03 | * | darkf quit (Read error: Connection reset by peer) |
| 03:43:11 | * | flaviu1 quit (Ping timeout: 252 seconds) |
| 03:43:33 | * | darkf joined #nimrod |
| 04:32:17 | * | BitPuffin quit (Ping timeout: 245 seconds) |
| 04:37:07 | zshazz_ | Are there any library functions to copy the bytes from one openArray to another regardless of their integral type? (eg put the bytes from an int64 into 2 int32 cells and so on) |
| 04:47:44 | fowl | not in the std lib |
| 05:21:20 | * | BitPuffin joined #nimrod |
| 05:55:37 | * | hoverbear quit (Ping timeout: 245 seconds) |
| 07:21:42 | araq | zshazz_: unsigned *operations* are important and Nimrod got them from day 1 in the form of +% operators. But encoding it via the type leads to exactly the bloat that you encountered. Unsigned then becomes a first class citizen, you need to be able to parseUInt them etc... it's code bloat. |
| 07:22:45 | araq | Just imagine we'd also have unsigned floating point numbers, the type zoo is incredible already. |
| 07:29:00 | zshazz_ | I don't think there'd be a point to having unsigned floating point numbers. OTOH, working with implementing an RNG makes me notice how necessary unsigned ints are. |
| 07:30:26 | zshazz_ | Things like necessitating the use of ze (as opposed to just asking for the size of thing you want) means you have to be even more careful or you'll accidentally `or` with a bunch of 1s because the sign got extended when you casted somewhere |
| 07:31:16 | araq | well guess what, I added 'uint' back because 'ze' etc. are too error prone... ;-) |
| 07:31:43 | araq | but this wasn't immediately obvious from the approach and had to be discovered :P |
| 07:31:57 | zshazz_ | But OTOH, putting all this hard work with "code bloat" means that user code is clean and crisp and just works. I'd rather put the hard work in first than deal with it later. But I do get your point with that. |
| 07:32:28 | zshazz_ | Yeah, I almost thing I should have implemented ISAAC with uint32s at this point, but I've gotten all the kinks worked out at this point, so it'd be mostly pointless |
| 07:33:18 | zshazz_ | Anyway, I've been working on this for 17 hours straight, basically... So I'm gonna head to bed. Take care :) |
| 07:34:10 | * | zshazz_ quit (Quit: Leaving) |
| 07:35:37 | fowl | o.o |
| 07:50:14 | * | Mat3 joined #nimrod |
| 07:50:17 | Mat3 | hello |
| 07:50:21 | Mat3 | araq: http://www.eecs.berkeley.edu/~nishant/papers/Chlorophyll.pdf |
| 07:50:28 | * | Skrylar quit (Ping timeout: 260 seconds) |
| 07:50:34 | Mat3 | ^porbably it is of interest for you |
| 07:51:50 | Mat3 | ciao |
| 07:51:54 | * | Mat3 quit (Client Quit) |
| 08:05:02 | * | kunev joined #nimrod |
| 09:00:23 | * | freezerburnv joined #nimrod |
| 09:05:08 | * | freezerburnv quit (Ping timeout: 254 seconds) |
| 09:31:57 | araq | kmp_int32 __kmpc_omp_task_with_deps(ident_t loc_ref, kmp_int32 gtid, kmp_task_t new_task, kmp_int32 ndeps, kmp_depend_info_t dep_list, kmp_int32 ndeps_noalias, kmp_depend_info_t noalias_dep_list) |
| 09:32:13 | araq | for all the underscore lovers among you ... ;-) |
| 09:32:45 | araq | reads like shit IMO |
| 09:50:02 | * | kunev quit (Read error: Connection reset by peer) |
| 09:51:30 | * | kunev joined #nimrod |
| 10:00:28 | fowl | deadCodeElim made my 988K binary only 720K! 2815 line project |
| 10:13:37 | araq | 720K is still pretty big |
| 10:20:11 | fowl | it is a server and client in one |
| 10:23:53 | fowl | gn |
| 10:52:15 | clone1018 | Hey guys I'm trying to build from source and I'm currently stuck here, https://gist.github.com/clone1018/631780a9dd8ebf456e08 is master not stable? |
| 11:02:09 | * | darkf quit (Read error: Connection reset by peer) |
| 11:02:56 | * | darkf joined #nimrod |
| 11:10:44 | dom96 | clone1018: Try the devel branch |
| 11:11:10 | clone1018 | I just went with the latest packaged releases |
| 11:11:19 | clone1018 | I'll try that next time though |
| 11:23:40 | dom96 | araq: Did you fix the forum compilation issues yet? |
| 11:30:54 | * | kemet joined #nimrod |
| 11:53:46 | * | ehaliewicz quit (Remote host closed the connection) |
| 12:00:49 | * | freezerburnv joined #nimrod |
| 12:05:09 | * | freezerburnv quit (Ping timeout: 252 seconds) |
| 12:13:49 | * | kemet quit (Ping timeout: 276 seconds) |
| 12:17:43 | * | untitaker quit (Ping timeout: 276 seconds) |
| 12:24:14 | * | untitaker joined #nimrod |
| 12:41:21 | * | flaviu1 joined #nimrod |
| 12:45:52 | * | BitPuffin quit (Ping timeout: 260 seconds) |
| 12:52:01 | * | superfunc quit (Ping timeout: 246 seconds) |
| 13:07:34 | * | Varriount|Mobile joined #nimrod |
| 13:24:59 | EXetoC | who wrapped gtk2? |
| 13:36:22 | * | freezerburnv joined #nimrod |
| 13:36:48 | * | darkf quit (Quit: Leaving) |
| 13:45:50 | * | freezerburnv quit (Quit: freezerburnv) |
| 14:28:21 | * | BitPuffin joined #nimrod |
| 14:28:57 | * | Mat3 joined #nimrod |
| 14:29:07 | Mat3 | hi |
| 14:32:04 | * | xenagi joined #nimrod |
| 14:33:56 | * | BitPuffin quit (Ping timeout: 260 seconds) |
| 14:35:55 | * | Mat3 quit (Quit: Verlassend) |
| 14:42:58 | * | bjz quit (Ping timeout: 240 seconds) |
| 14:54:00 | * | freezerburnv joined #nimrod |
| 15:04:24 | * | askatasuna joined #nimrod |
| 15:06:10 | * | jfilstrup joined #nimrod |
| 15:07:52 | jfilstrup | exit |
| 15:07:58 | * | jfilstrup quit (Client Quit) |
| 15:08:40 | * | superfunc joined #nimrod |
| 15:20:03 | * | kunev quit (Quit: leaving) |
| 15:25:43 | * | Matthias247 joined #nimrod |
| 15:49:07 | * | Johz joined #nimrod |
| 16:10:25 | * | brson joined #nimrod |
| 16:15:02 | * | hoverbear joined #nimrod |
| 16:23:42 | * | hoverbear quit () |
| 16:55:40 | * | vendethiel joined #nimrod |
| 17:10:08 | * | q66 joined #nimrod |
| 17:10:08 | * | q66 quit (Changing host) |
| 17:10:08 | * | q66 joined #nimrod |
| 17:19:56 | * | Trustable joined #nimrod |
| 17:28:55 | * | askatasuna quit (Ping timeout: 252 seconds) |
| 17:35:15 | * | DAddYE_ joined #nimrod |
| 17:47:31 | * | Mat3 joined #nimrod |
| 17:50:03 | araq | hi Mat3 your link is interesting |
| 17:50:36 | araq | but I can't see it working out in practice, the 18bit architecture is simply too weird |
| 17:51:26 | araq | and emulating essential stuff like 32bit ops will lose against implementing them directly in hardware |
| 17:51:32 | Mat3 | hello araq |
| 17:52:02 | * | Xuerian left #nimrod (#nimrod) |
| 17:53:00 | Mat3 | well, to my knowledge green array is allready working on a 32 bit version |
| 17:54:07 | Mat3 | despcie that, the approach applied also to other array architectures (xmos, propeller for example) |
| 17:54:14 | Mat3 | ^despice |
| 17:56:39 | araq | ah thats good news |
| 17:59:44 | Mat3 | another suspect of research (in the Forth community) seem to be region-based memory managment |
| 18:03:36 | Mat3 | by the way: Does one know an effective hash algorithm for fixed-sized strings which maps well to small table sizes (hash values should be lesser than 12 bit) ? |
| 18:08:34 | Matthias247 | 18bit architecture? what's that? some dsp? |
| 18:08:45 | Varriount|Mobile | Synthsis... Isn't that when a compiler puts small operations together in such a way as to mimic (with a more optimized version) the intent of the code it's given? |
| 18:11:11 | Mat3 | Matthias247: No, it is an asynchron field processor from Green Arrays |
| 18:12:24 | * | enurlyx joined #nimrod |
| 18:12:40 | * | EXetoC quit (Quit: WeeChat 0.4.3) |
| 18:12:55 | enurlyx | Hello |
| 18:13:05 | Mat3 | hello enurlyx |
| 18:13:19 | * | EXetoC joined #nimrod |
| 18:14:34 | enurlyx | I want to write a macro, that creates wrapper templates for the functions in the vtable. Is it possible to parse the typedefinition for a given type at compiletime? |
| 18:14:57 | enurlyx | Vtable of COM Objects |
| 18:17:23 | araq | enurlyx: vtable of COM objects sounds like fun but with the knowledge of where the C++ compiler puts the virtuals (the offsets) it should be doable |
| 18:18:45 | Varriount|Mobile | enurlyx: Would the type definitions be known to be COM objects ahead of time? |
| 18:18:46 | Matthias247 | Mat3: ah ok. I'm not familiar with Green Arrays |
| 18:19:02 | enurlyx | At the moment i hava Direct2D wrapper done with c2nim |
| 18:19:25 | Varriount|Mobile | enurlyx: Could you use a pragma macro, a la importc? |
| 18:19:35 | enurlyx | And i need to call a function in this way: d2d2Factory.lpVtbl.CreateSomething(d2dFactroy... |
| 18:20:03 | enurlyx | I want macros to call it this way: d2dFactroy.CreateSomething(... |
| 18:20:27 | enurlyx | I mean i want a macro that generates all the templates for all the functions in the vtable |
| 18:21:41 | Varriount|Mobile | enurly: A pragma macro should do, I think. Just have it return the templates, in addition to the original type definition |
| 18:22:00 | Varriount|Mobile | * enurlyx |
| 18:22:04 | * | dLog_ joined #nimrod |
| 18:22:12 | enurlyx | i have to check it |
| 18:22:33 | araq | enurlyx: first simple step is to do this: |
| 18:24:46 | araq | template `%`(a, b: expr, c: varargs[expr]): expr = a.lpVtbl.b(a, c) |
| 18:24:57 | araq | but nah, screw this, doesn't work |
| 18:24:58 | * | dLog quit (Ping timeout: 240 seconds) |
| 18:33:45 | enurlyx | ? |
| 18:35:18 | enurlyx | But can i call a macro with a type? And then extract the structure of the type? (Like reflection) |
| 18:35:45 | araq | sure |
| 18:35:49 | araq | mymacro: |
| 18:35:55 | araq | type Foo = object |
| 18:35:57 | araq | ... |
| 18:38:31 | enurlyx | Mh I miss the point how to extract the information |
| 18:39:15 | enurlyx | Another question: I did modify c2nim to handle anonymous structs and unions. |
| 18:39:24 | araq | cool! |
| 18:39:26 | enurlyx | Do there exists tests? |
| 18:39:30 | EXetoC | you parse the AST that is passed in |
| 18:40:28 | araq | enurlyx: c2nim/tests but unfortunately the desired output has not been specified, so you need to check it manually |
| 18:40:30 | EXetoC | but applying it to individual types instead simplifies chaining |
| 18:40:33 | araq | patches are welcome of course |
| 18:41:05 | araq | EXetoC: no, one doesn't *parse* the AST, it already is parsed, you simply need to process it, perhaps recursively |
| 18:41:20 | araq | for many people that seems to be too hard already |
| 18:46:52 | * | Mat3 created a hash function which only involve two binary operations and which gives near perfect results |
| 18:48:10 | def-_ | Mat3: how? |
| 18:50:03 | * | bobsuncle joined #nimrod |
| 18:52:03 | OrionPK | forum needs to be moved from the right-nav in nimrod-lang.org |
| 18:52:12 | OrionPK | it keeps getting further down the page every time a news item is added -_- |
| 18:52:49 | bobsuncle | hey guys.. I've recently discovered Nimrod and I'm interesting in starting an ambitious project with it.. quick question: what would be the preferred way to embed an HTML/CSS+JS rendering engine into a cross-platform GUI app? |
| 18:53:06 | bobsuncle | there are no QtWebKit bindings, no any other HTML engine bindings that I can see |
| 18:53:59 | bobsuncle | essentially, I wouldn't need all the WebKit features, but just local HTML rendering.. latest HTML5/CSS3/JS support would be a plus |
| 18:54:41 | bobsuncle | is there something lightweight I could embed and it wouldn't be a big effort to create bindings to? |
| 18:55:03 | Mat3 | def-_: it is a modificated checksum generator which I know since 30 years from listings in a some home-computer magazine |
| 18:55:26 | Mat3 | ^in a home-computer magazine |
| 18:56:04 | Matthias247 | bobsuncle: as you need to integrate a whole browser (webkit, blink, ...) it won't really be lightweight |
| 18:57:09 | enurlyx | araq, i still do not get the type into the macro. I only get an Ident Node with the name of the type, but I need the Type AST, so that i can extract the functions of the vtable and generate a template for each of them. Maybe you can give me an other hint :D |
| 18:58:18 | araq | enurlyx: well methods/procs are not attached to types, so that won't work even with the planned type API that is currently missing indeed |
| 18:58:26 | araq | bobsuncle: hi, welcome |
| 18:58:44 | araq | I once wrapped GTK's browser and created a demo with it, but lost the code |
| 18:59:11 | Matthias247 | take a look at chromium embedding framework, node-webkit, atom-shell and probably a view others. |
| 18:59:18 | bobsuncle | Matthias247: I guess, I was just hoping there would be an embeddable browser out there that wouldn't be massive.. there is https://code.google.com/p/chromiumembedded/, and there are bindings for Go.. would it be a big undertaking to create Nimrod bindings? |
| 19:00:18 | araq | bobsuncle: that depends on whether it exposes a C API |
| 19:00:42 | Matthias247 | don't know which ones are easier to use/wrap |
| 19:00:43 | araq | if it doesn't things are harder but then I can't see how the Go wrapper can work without an intermediate C API |
| 19:01:29 | * | zshazz joined #nimrod |
| 19:01:52 | Matthias247 | afaik canonical is also working on a webkit wrapper as an alternative to qtwebkit. but i don't remember the name |
| 19:02:01 | bobsuncle | araq: indeed it does.. CEF exposes C and C++ APIs |
| 19:04:10 | enurlyx | araq, why are functions not part of the type API? a = proc (...) {.stdcall.} I think this is usefull |
| 19:08:04 | araq | enurlyx: I don't follow you. proc *types* are of course part of the type's AST |
| 19:08:31 | araq | bobsuncle: well that's "simply" a matter of running c2nim over the C API then |
| 19:11:54 | bobsuncle | araq: Hmm right.. I'll take a look at that, thanks! |
| 19:12:03 | * | ehaliewicz joined #nimrod |
| 19:12:49 | enurlyx | araq, you wrote methods/procs are not attached to types and wont work with the planned type APIs. If i understand you right, a procvar (functionpointer), will work with the API. |
| 19:13:10 | araq | yes |
| 19:13:17 | enurlyx | For now i think i can do my macro by just putting the whole type definion into the macro call :) |
| 19:13:33 | araq | that's what I tried to tell you :P |
| 19:13:42 | enurlyx | ahhh. i am so slow XD |
| 19:19:13 | * | kunev joined #nimrod |
| 19:33:21 | * | enurlyx quit (Quit: Nettalk6 - www.ntalk.de) |
| 19:35:12 | * | kunev quit (Ping timeout: 245 seconds) |
| 19:35:17 | * | bobsuncle quit (Quit: Page closed) |
| 19:57:21 | dom96 | If you guys see bobsuncle again, give him this https://github.com/Varriount/nimrod-libcef |
| 19:57:57 | zshazz | Are there any ways to check if code is being compiled for release vs debug (it's for a `when`) |
| 19:58:33 | dom96 | when defined(release) |
| 19:58:54 | zshazz | thanks |
| 20:05:54 | def-_ | Does anyone know if this ambiguity error I get is intentional, and if so, how to fix it? https://gist.github.com/def-/cb05159fe18e3c3dfe12 |
| 20:06:32 | dom96 | Yes, you can't overload on return types. |
| 20:09:15 | dom96 | You need different types for those params to fix it. |
| 20:09:37 | def-_ | That's a problem |
| 20:13:08 | dom96 | Only way I can think of getting this to work is to write a `:=` macro and then invoke it like so: var fr: Frequency := 1 / Time(1) |
| 20:13:31 | dom96 | The macro can then transform the `/` to an appropriate proc. |
| 20:13:54 | def-_ | That's funny, I already have "Frequency := 1 / Time" in my code as a macro |
| 20:14:34 | dom96 | there you go :P |
| 20:14:45 | def-_ | But for something else! |
| 20:16:17 | dom96 | You could come up with a different operator. |
| 20:18:06 | def-_ | But that would be harder to write because you have to remember another operator |
| 20:18:16 | def-_ | Thanks anyway for the suggestions, dom96 |
| 20:18:55 | dom96 | np. Maybe araq can suggest something better. |
| 20:19:21 | Mat3 | h,mm, I remember ':=' can be used as macro name but not '=' as replacement for '==' |
| 20:20:36 | * | Johz quit (Quit: Leaving) |
| 20:22:17 | * | kunev joined #nimrod |
| 20:23:36 | * | xenagi quit (Quit: Leaving) |
| 20:26:56 | * | kunev quit (Ping timeout: 260 seconds) |
| 20:31:21 | zshazz | Ok, I probably can't make any more progress on the RNG stuff. So I'm going to go ahead and release what I have and someone else can finish it later on, if they wish: https://gist.github.com/Zshazz/37d6205ddb7e4c526705 |
| 20:31:34 | zshazz | (I'll post it on the forums too because it's a secret gist) |
| 20:33:08 | dom96 | zshazz: put it in its own repo please! |
| 20:33:19 | * | Raynes quit (*.net *.split) |
| 20:33:19 | * | eximiuswastaken quit (*.net *.split) |
| 20:33:20 | * | skroll quit (*.net *.split) |
| 20:33:20 | * | tumak_ quit (*.net *.split) |
| 20:33:20 | * | jez0990 quit (*.net *.split) |
| 20:33:21 | * | bstrie quit (*.net *.split) |
| 20:33:22 | dom96 | and submit it to babel |
| 20:33:24 | zshazz | It doesn't work, unfortunately |
| 20:33:29 | * | tumak joined #nimrod |
| 20:33:31 | * | bstrie joined #nimrod |
| 20:33:31 | * | Raynes joined #nimrod |
| 20:33:31 | * | Raynes quit (Changing host) |
| 20:33:31 | * | Raynes joined #nimrod |
| 20:33:34 | * | skroll joined #nimrod |
| 20:33:35 | * | eximiuswastaken joined #nimrod |
| 20:33:38 | zshazz | It can't be used until it's finished (you can see why it doesn't work in user.nim) |
| 20:33:57 | zshazz | Otherwise, I would. |
| 20:33:59 | * | jez0990 joined #nimrod |
| 20:34:19 | * | freezerburnv quit (Ping timeout: 276 seconds) |
| 20:34:40 | dom96 | I think even if it's broken, putting it in its own repo is still good. |
| 20:34:49 | dom96 | Makes it easier for people to fork and fix it |
| 20:35:05 | zshazz | Ok, I'll do it. Give me a few minutes |
| 20:36:36 | dom96 | zshazz: also, try writing 'mixin shoveEachTo' in your reset proc |
| 20:37:28 | dom96 | maybe that will fix it |
| 20:40:59 | zshazz | Thanks for the advice but it appears to not work still |
| 20:41:16 | Mat3 | ciao |
| 20:41:28 | * | Mat3 quit (Quit: Verlassend) |
| 20:41:40 | zshazz | (I'm on 0.9.4 though, so maybe it would work in the git release) |
| 20:44:56 | * | Trustable1 joined #nimrod |
| 20:48:18 | * | freezerburnv joined #nimrod |
| 20:49:57 | dom96 | it's possible |
| 20:53:20 | * | freezerburnv quit (Quit: freezerburnv) |
| 20:54:23 | * | Trustable quit (*.net *.split) |
| 20:55:36 | * | brson quit (Quit: Lost terminal) |
| 20:55:44 | * | brson joined #nimrod |
| 20:57:51 | zshazz | Unfinished work here: https://github.com/Zshazz/Ramrod |
| 21:01:16 | * | noam quit (Ping timeout: 260 seconds) |
| 21:02:12 | araq | zshazz: thanks for all your hard work |
| 21:02:42 | araq | I'll see what I can do for your API design problems but right now I'm too tired to even look at your problem, sorry |
| 21:03:17 | zshazz | That's fine, I understand :) |
| 21:03:38 | Varriount|Mobile | zshazz: I'll see if I can help get it running (as son as I recover from this cold) |
| 21:03:53 | Varriount|Mobile | *soon |
| 21:04:20 | zshazz | Varriount|Mobile: Get well soon :) |
| 21:04:56 | Varriount|Mobile | zshazz: Thanks |
| 21:06:25 | * | vendethiel quit (Quit: q+) |
| 21:07:03 | * | noam joined #nimrod |
| 21:11:39 | * | noam quit (Client Quit) |
| 21:14:03 | * | BitPuffin joined #nimrod |
| 21:21:25 | * | noam joined #nimrod |
| 21:21:29 | * | noam quit (Read error: Connection reset by peer) |
| 21:22:50 | * | noam joined #nimrod |
| 21:25:29 | * | superfunc quit (Ping timeout: 265 seconds) |
| 21:33:37 | * | Matthias247 quit (Read error: Connection reset by peer) |
| 21:39:35 | * | Trustable1 quit (Quit: Leaving) |
| 21:45:19 | * | Varriount|Mobile quit (Read error: Connection reset by peer) |
| 21:45:35 | * | Varriount|Mobile joined #nimrod |
| 21:47:07 | * | Changaco left #nimrod (#nimrod) |
| 22:06:21 | * | brson quit (Read error: Connection reset by peer) |
| 22:06:28 | * | brson joined #nimrod |
| 22:09:53 | * | hoverbear joined #nimrod |
| 22:11:53 | * | brson quit (Ping timeout: 265 seconds) |
| 22:28:19 | * | noam quit (Read error: Connection reset by peer) |
| 22:29:14 | * | noam joined #nimrod |
| 22:49:55 | * | brson joined #nimrod |
| 23:03:19 | * | Jesin joined #nimrod |
| 23:04:57 | * | freezerburnv joined #nimrod |
| 23:30:08 | * | hoverbear quit () |
| 23:31:43 | * | darkf joined #nimrod |
| 23:32:19 | * | xenagi joined #nimrod |
| 23:57:12 | * | Varriount|Mobile quit (Read error: Connection reset by peer) |