00:00:13 | * | XAMPP joined #nimrod |
01:02:49 | * | XAMPP quit (Ping timeout: 246 seconds) |
01:09:38 | * | XAMPP joined #nimrod |
01:25:43 | * | q66 quit (Quit: Quit) |
02:00:08 | * | Amrykid quit (Changing host) |
02:00:08 | * | Amrykid joined #nimrod |
06:40:43 | * | llm joined #nimrod |
06:42:59 | llm | so i saw that nimrod got floatChecks for detecting NaN and inf - is there also an check for denormal floats? |
06:48:10 | llm | info about denormals: http://en.wikipedia.org/wiki/Denormal_number - i've hit them some time in an 3d simulation project - see the Wikipedia note on Performance (sometimes 100x slower....) |
06:49:37 | llm | and a nice stackoverflow article with C code example: http://stackoverflow.com/questions/9314534/why-does-changing-0-1f-to-0-slow-down-performance-by-10x/9314926#9314926 |
07:17:19 | llm | see https://github.com/D-Programming-Language/phobos/blob/master/std/math.d isSubnormal implementation |
07:59:56 | Araq | llm: I'd like to implement these checks but I don't know how to do that portably :-) |
08:12:41 | llm | i don't known on what platforms IEEE 754-2008 is also working - ARM i think |
08:13:25 | llm | or what you mean with portably |
08:23:39 | Araq | I mean it's in C99 dunno about C++ |
08:23:44 | Araq | but meh, I'll find out |
08:23:50 | Araq | make a feature request please |
08:24:06 | Araq | math.sin et. al. need these checks too anyway |
08:25:19 | llm | bbl - where to add a feature request? |
08:27:16 | Araq | bug report, I'll mark it as feature request |
08:28:54 | Araq | I have to go too, see you later |
08:56:16 | * | Trix[a]r_za is now known as Trixar_za |
09:22:36 | * | Trixar_za is now known as Trix[a]r_za |
09:34:24 | * | Trix[a]r_za is now known as Trixar_za |
09:35:28 | Trixar_za | I should stop committing untested code. But using git is so much easier when your test system can pull your changes from git :( |
10:41:21 | * | q66 joined #nimrod |
10:54:14 | * | Trixar_za is now known as Trix[a]r_za |
11:07:05 | Araq | Trix[a]r_za: we have exactly the same problem :D |
12:46:01 | dom96 | Damn. I slept way too long |
12:56:05 | * | Roin is now known as math|Roin |
13:19:19 | fowl | does ref object of PSomeOtherRefObject work |
13:19:30 | fowl | also morning |
13:23:49 | dom96 | Hrm. |
13:23:54 | dom96 | Interesting question |
13:24:24 | dom96 | In that case you would be making a ref to an object which inherits from a ref |
13:24:26 | dom96 | I think |
13:24:51 | dom96 | ref object of TSomeOtherObject makes more sense |
13:35:11 | fowl | hmm |
13:35:17 | fowl | im trying to refactor too many things at once |
14:30:08 | * | llm quit (Ping timeout: 256 seconds) |
14:41:28 | shevy | hehe |
16:08:29 | * | shevy quit (Ping timeout: 244 seconds) |
16:15:14 | dom96 | What key combination should commenting code be? |
16:18:00 | * | math|Roin is now known as Roin |
16:18:51 | fowl | control / |
16:21:12 | dom96 | hrm, I quite like that. |
16:21:42 | * | shevy joined #nimrod |
16:22:00 | dom96 | Interesting. Currently that key combination is equivalent to Ctrl + A in the GtkSourceView. |
16:39:00 | dom96 | Well this feels like a slap in the face: http://developer.gnome.org/gtk3/3.4/GtkRecentManager.html |
16:39:12 | dom96 | I implemented that myself for Aporia. |
16:41:29 | Araq | and you did it better as it's implemented in nimrod |
16:42:30 | dom96 | heh |
16:43:42 | Araq | imo you should use as few of GTK as possible ;-) |
17:13:46 | dom96 | Araq: What are your opinions on that comment shortcut? |
17:14:29 | Araq | it is fine |
17:47:23 | fowl | there is no string format like printf is there |
17:47:39 | fowl | that i can use %f |
17:47:42 | fowl | etc |
17:47:51 | Araq | true |
17:48:03 | Araq | you could wrap C's |
17:48:16 | Araq | but it's unsafe :P |
17:48:47 | Araq | and nimrod's strutils.format is better designed |
19:15:18 | * | llm joined #nimrod |
19:26:28 | * | llm quit (Quit: ChatZilla 0.9.88.2 [Firefox 14.0.1/20120713134347]) |
20:15:34 | dom96 | I just realised that I haven't tested staticExec yet |
20:15:42 | dom96 | And I really wanted that feature |
20:18:06 | fowl | what is staticExec |
20:18:25 | dom96 | http://build.nimrod-code.org/docs/system.html#618 |
20:18:55 | fowl | ah i see |
20:19:06 | fowl | why is gorge an alias lol |
20:19:19 | dom96 | Because it sounds cool :P |
20:21:27 | Araq | because language designers want to have some fun :P |
20:21:44 | Araq | people didn't like my 'gorge' and preferred 'staticExec' |
20:22:01 | Araq | so we have both now :D |
20:22:38 | dom96 | You know comment toggle is actually quite tricky. |
20:23:21 | dom96 | Gtk doesn't make it much simpler heh |
20:24:35 | fowl | dom96: when i do stringstream.setPosition(0) it will go to -1 if the string is empty |
20:25:01 | fowl | min(0, str.len-1) |
20:25:20 | fowl | er min(pos, str.len-1) |
20:26:53 | dom96 | I shall fix it |
20:27:51 | Araq | system.clamp ftw |
20:28:40 | dom96 | :) |
20:35:51 | Araq | what about: x >< a..b |
20:36:01 | Araq | for clamp(x, a, b) ? |
20:37:34 | Araq | looks cooler |
20:38:00 | dom96 | sure |
20:38:07 | Araq | x \/ a..b |
20:38:30 | Araq | let x "fall into" a..b ? |
20:39:07 | * | Araq likes ascii art |
20:42:15 | dom96 | fowl: There you go. |
20:47:30 | fowl | coo |
21:11:50 | Araq | damn I'm torn between "term rewriting macros" and "compiler as a service" |
21:12:18 | dom96 | Flip a coin... oh wait, no. I want "compiler as a service" |
21:12:31 | Araq | I have only time for 1 |
21:14:15 | Araq | I think I will do TRM as CAS has already been done before :P |
21:14:27 | Araq | TRM will make Nimrod truly unique |
21:14:35 | Araq | much better for marketing |
21:14:53 | Araq | whereas CAS is only useful for Nimrod programmers :P |
21:46:25 | * | Trix[a]r_za is now known as Trixar_za |
21:47:46 | * | XAMPP quit (Ping timeout: 246 seconds) |
21:48:32 | dom96 | Araq: Fine, go with TRM |
21:48:38 | dom96 | And we will then impress Reddit :P |
21:48:54 | * | XAMPP_ joined #nimrod |
21:49:26 | Araq | ugh, I won't show up on Reddit |
21:49:33 | dom96 | Yes, you will. |
21:49:38 | Araq | I'll use another nick |
21:49:46 | Araq | so that I can troll |
21:49:58 | Trixar_za | Funny enough, me too |
21:50:02 | dom96 | lol |
21:50:03 | Trixar_za | And astroturf |
21:50:20 | dom96 | Fine then. I'll be the mature one. |
21:50:27 | * | dom96 shakes head in disappointment |
21:50:36 | dom96 | :P |
21:50:37 | Trixar_za | We know you do it too |
21:53:11 | dom96 | maybe |
21:54:12 | dom96 | Araq: oh, did you receive your gittip money? |
21:54:39 | Araq | dunno |
21:56:49 | Araq | "Araq receives $3.00 per week." |
21:57:05 | Araq | wasn't that like $5 ? |
21:57:08 | dom96 | no |
21:57:47 | Araq | damn |
21:57:51 | Araq | well it works |
21:57:55 | Araq | I'm now rich |
21:58:01 | Araq | "Your balance is $6.00." |
21:58:59 | dom96 | lol |
21:59:10 | dom96 | Now send $1 to me :P |
22:00:09 | * | XAMPP_ quit (Quit: Leaving) |
22:26:52 | * | XAMPP joined #nimrod |
22:27:29 | Araq | oh Trixar_za, did you notice? there is now a strutils.format() ... |
22:27:46 | Trixar_za | :| |
22:27:53 | Araq | had to add it as the % [] doesn't work with varargs properly ... |
22:28:08 | Araq | but % is not deprecated |
22:30:37 | Trixar_za | Ok, so best of both worlds |
22:30:49 | Trixar_za | rather than instant shift from one to the other |
22:31:15 | Araq | well I didn't make it worse (I hope) |
22:31:31 | Araq | but it's annoying that the varargs don't work properly with % |
22:32:49 | Araq | and I might even make string interpolation builtin |
22:33:07 | Araq | echo i"x is $x" |
22:34:30 | dom96 | Should `%` use openarray anyway? |
22:34:35 | dom96 | *Shouldn't |
22:34:59 | Araq | it does |
22:35:08 | Araq | the problem is that [] constructs an array |
22:35:19 | Araq | so everything in it should be of the same type |
22:35:48 | Trixar_za | I don't mind improvements personally. Just as long as (some) backward compatibility is around afterward. |
22:35:56 | Araq | x.format("$# $#", 13, 5.0) # works |
22:35:57 | dom96 | You want varargs' ability to specify `$` for converting everything to a string? |
22:36:00 | Trixar_za | format isn't a bad idea in python 3k |
22:36:12 | Trixar_za | The problem I have with it is that they enforce it |
22:36:27 | Araq | er |
22:36:32 | Araq | format("$# $#", 13, 5.0) # works |
22:36:43 | Araq | "$# $#" % [13, 5.0] # does not |
22:36:52 | Araq | and can't easily work ... :-/ |
22:37:57 | Trixar_za | Hmmm, I'm weird. Women that are interested in me I have no interest in. But just have one that shows no interest in me at all and I'm curious to why :/ |
22:38:25 | Trixar_za | Sorry, just thinking of this Annette "Ice" Carstens at the show I was at tonight |
22:41:37 | Araq | well that surely was helpful feedback :P |
22:45:49 | Trixar_za | lol. I |
22:46:01 | Trixar_za | I'm easy, just don't make something complicated and I'm happy |
22:47:38 | Araq | ha, the language is already quite complex :P |
22:47:44 | Araq | more than you think it is :P |
22:47:58 | Araq | but it's well hidden |
22:48:37 | Araq | and you need to put the complexity into the language anyway to make the resulting *programs* easier ;-) |
22:49:33 | Trixar_za | Several layers between something and the underlying gears that make it work |
22:50:02 | Trixar_za | Like clock. It's simple to use and set, but the underlying systems are quite complex |
22:50:07 | Trixar_za | a& |
22:50:09 | Trixar_za | a* |
22:50:16 | * | Trixar_za may have had a few drinks to drink |
23:08:43 | Araq | good night |
23:09:54 | Trixar_za | Night Araq |
23:23:07 | dom96 | I like that I can do: (addr foo).blah() |
23:23:10 | dom96 | It looks so stylish |