|
i thought compilers all give up after a couple loops of that?
|
#
?
Jul 20, 2016 00:18
|
|
|
|
| # ? May 24, 2024 06:21 |
|
|
hobbesmaster posted:i thought compilers all give up after a couple loops of that?
|
|
#
?
Jul 20, 2016 00:20
|
|
|
nah thats not even the hosed up code i was thinking of. i cant find the exact one im thinking of but found plenty of fun results like http://tgceec.tumblr.com/
|
|
#
?
Jul 20, 2016 01:01
|
|
|
if C# had anonymous classes I don't see what the point of moq would be most of the time
|
|
#
?
Jul 20, 2016 01:18
|
|
|
c# does have anonymous classes
|
|
#
?
Jul 20, 2016 05:23
|
|
|
c# has anonymous types but not anonymous classes like in Java an anonymous class like in Java will let you create an implementation for an interface and do stuff w/ it or return it from a function anonymous classes are dumb and not a good feature imo
|
|
#
?
Jul 20, 2016 05:35
|
|
|
# of times I missed anonymous classes in C#: 0
|
|
#
?
Jul 20, 2016 05:38
|
|
|
anonymous classes were mostly obsoleted by lambdas. they're still nice to have but i can see why you wouldn't miss them in a language that already had lambdas
|
|
#
?
Jul 20, 2016 08:21
|
|
|
VikingofRock posted:Wasn't there some discussion about allowing people to write their own error messages that the compiler would emit when a template substitution fails? Or did I dream that? YeOldeButchere posted:if not don't worry, you can probably accomplish something close enough using templates! Ding ding ding ding*. Right now you can do a lot of playing around with SFINAE, instantion stack cutting and static (compile-time) asserts to get semi-reasonable error messages, but it takes time, is unobvious and fiddly. (Plus there is always a problem where users want different error messages than developers of the library) The ~great future~ theoretically holds concepts and easy-to-get sane error messages, but concepts have been the "next great thing" since C++11 and even the core workings have changed couple of times based on feedback from the committee. Basically they can either be highly complex, moderately complex and really hard to get into the language in backward compatible way or not really useful. I admit I have given up on watching what the final form will be until they actually get voted into draft. ------- * Actually you don't want to write messages when template substitution fails, but rather when resolution fails, because reasons.
|
|
#
?
Jul 20, 2016 08:32
|
|
|
I liked the class-like concepts proposal, that doubled as adapters. I forget who proposed it and I can't find the draft anywhere, but it was the best idea imo, gently caress concepts lite, I don't care if they are easy for compiler writers
|
|
#
?
Jul 20, 2016 09:59
|
|
|
hobbesmaster posted:i thought compilers all give up after a couple loops of that? Running out of stack space counts as giving up right?
|
|
#
?
Jul 20, 2016 11:19
|
|
|
gonadic io posted:Running out of stack space counts as giving up right? IIRC, last time I tried code like that, Clang exited cleanly with nice error message. Admittedly I haven't tried other compilers, but that counts.
|
|
#
?
Jul 20, 2016 12:58
|
|
|
GameCube posted:should i bother asking the old guys if there was a reason they didn't just use Moq instead of writing every loving mock class by hand or should i just assume ignorance. what possible reason could you have for NOT using Moq it's probably ignorance, but if you have a mock more complicated than a simple return value + verify then it's usually easier to write it out by hand anyway
|
|
#
?
Jul 20, 2016 13:46
|
|
|
lay some sick yospinions on me about contracts can i use them for input validation, or should i keep using if (condition) throw Exception? if not, what are they for i don't get it
|
|
#
?
Jul 20, 2016 16:01
|
|
|
~Coxy posted:it's probably ignorance, but if you have a mock more complicated than a simple return value + verify then it's usually easier to write it out by hand anyway not when there's like... a billion freaking methods and you only need to customize like one of them. frick
|
|
#
?
Jul 20, 2016 16:13
|
|
|
oh. moq doesn't support monoandroid. gently caress.
|
|
#
?
Jul 20, 2016 16:16
|
|
|
ok i got a weird bug and i can't find a single thing about it online: if the page is zoomed at all in chrome it completely breaks all of our javascript, what the hell? it's not even trowing exceptions it's like the javascript just plain doesn't execute works fine zoomed in ie too
|
|
#
?
Jul 20, 2016 19:57
|
|
|
chome is bad. stop using chome.
|
|
#
?
Jul 20, 2016 19:59
|
|
|
oh man now i gotta write unit tests against some concurrent poo poo. this is gonna be fun. i am not sure what irony level i am on right now
|
|
#
?
Jul 20, 2016 20:44
|
|
|
Shaggar posted:chome is bad. stop using chome. ie8 is worse
|
|
#
?
Jul 20, 2016 20:46
|
|
|
Shaggar posted:chome is bad. stop using chome. careful with that edge
|
|
#
?
Jul 20, 2016 20:49
|
|
|
GameCube posted:oh man now i gotta write unit tests against some concurrent poo poo. this is gonna be fun. i am not sure what irony level i am on right now if you can mock the scheduling so you can make sure to test edge cases then it's not that bad and actually useful if not then lol your tests are worthless and non-repeatable and just lol
|
|
#
?
Jul 20, 2016 21:10
|
|
|
HoboMan posted:ie8 is worse use ie11 or edge
|
|
#
?
Jul 20, 2016 21:55
|
|
|
Bloody posted:lay some sick yospinions on me about contracts i have concluded that if/throw is a more convenient approach in c# because the contracts stuff freaked out and broke and wanted some plugin or something
|
|
#
?
Jul 20, 2016 21:57
|
|
|
if(nullcheck) throw ArgumentNullException forever.
|
|
#
?
Jul 20, 2016 22:25
|
|
|
There should be a Java compiler plugin that automatically inserts null checks for every parameter not annotated @Nullable
|
|
#
?
Jul 20, 2016 22:38
|
|
|
for the first time in my professional career, i have found a use case for currying of course it's in my unit tests so it's still not Production Currying. but still
|
|
#
?
Jul 21, 2016 01:24
|
|
|
Mr Dog posted:Barycentric rendering is poo poo-slow yes. yeah turns out solving the equation of a plane for Z is easier and faster than doing wacky barycentric calculations
|
|
#
?
Jul 21, 2016 01:38
|
|
|
lol http://stackstatus.net/post/147710624694/outage-postmortem-july-20-2016
|
|
#
?
Jul 21, 2016 01:51
|
|
|
yo luigi thirty (and other people doing neat games/graphics rendering stuff), did you have a strong knowledge of geometry going into your project, or did you learn as you go? i just make simple 2d games and have never written a custom renderer or anything, but have found myself busting out various googled equations i barely understand for things like physics calculations and animations and stuff, and i wish i had a better grasp on this stuff (especially since there are very few good general-purpose math libraries for javascript). i'd like to be able to do more complex stuff in my games but i'm scared off by some of the math, so i'm wondering if there are any good courses or tutorials or anything or if my best bet is to just continue trying to turn http://math.stackexchange.com/ answers into code
|
|
#
?
Jul 21, 2016 01:53
|
|
|
i switched out of the computer graphics course in school because after the first week everyone started busting out words like "vector" and "matrix". i thought "heh i got this" but then i realized they were not talking about science fiction
|
|
#
?
Jul 21, 2016 02:12
|
|
|
Nice. I ran into something like this when working on code that turns the BBCode-like tags into HTML. If you opened a bunch of identical tags, then closed one of them, the code found itself in O(2n) land.
|
|
#
?
Jul 21, 2016 02:22
|
|
|
abraham linksys posted:yo luigi thirty (and other people doing neat games/graphics rendering stuff), did you have a strong knowledge of geometry going into your project, or did you learn as you go? um i'm learning as i go geometry is hard
|
|
#
?
Jul 21, 2016 03:12
|
|
|
I got A's in four calculus courses in college and a C- in linear algebra, so I'm mostly copying and pasting!
|
|
#
?
Jul 21, 2016 03:29
|
|
|
i made some other optimizations in addition to ejecting barycentricism and now the renderer is INFINITELY FAST
|
|
#
?
Jul 21, 2016 03:44
|
|
|
linear algebra owns
|
|
#
?
Jul 21, 2016 04:36
|
|
|
linear algebra suffuses and perfuses DOOM's program code. or just any sort of real time computer graphics application in general really who says math can't be metal as hell
|
|
#
?
Jul 21, 2016 04:38
|
|
|
linear algebra is basically everything. Blas rules everything around me
|
|
#
?
Jul 21, 2016 04:40
|
|
|
i'm a for reals mathamagician, ama
|
|
#
?
Jul 21, 2016 04:43
|
|
|
|
| # ? May 24, 2024 06:21 |
|
|
Mr Dog posted:linear algebra suffuses and perfuses DOOM's program code. or just any sort of real time computer graphics application in general really i mean, carmack is like a genius savant in his domain which doesnt really disprove the idea that its extremely hard. no one could get the sort of performance doom got without specialty hardware with no OS in the way before doom came along. the moment someone starts talking about matrices and vectors my eyes glaze over. its so out of my wheelhouse. the last math i took was calc 1 my freshman year of college. i skipped every day because it was at 8am, showed up only for quizzes and tests, and just barely got the required credits with a c in the course. i dont remember jack poo poo from that course. i have to google whenever i need to remember what a logarithm is.
|
|
#
?
Jul 21, 2016 04:48
|
|
