|
if you're using C# and like linq/IEnumerable, get MoreLinq for a bunch of crap the standard library is missing
|
# ? Jul 9, 2016 14:13 |
|
|
# ? May 11, 2024 10:17 |
|
We have this weird homebrewn DSL for our test environment. It would be "fine" if it wasn't for the fact that it doesn't have stuff like functions. Instead you have to execute other scripts and pass all parameters as globals. Oh and of course all scripts modifies other globals so you have no idea what's going on
|
# ? Jul 9, 2016 14:17 |
|
comedyblissoption posted:if you're using C# and like linq/IEnumerable, get MoreLinq for a bunch of crap the standard library is missing also peep the (builtin) parallel linq poo poo if you haven't already
|
# ? Jul 9, 2016 14:40 |
|
comedyblissoption posted:if you're using C# and like linq/IEnumerable, get MoreLinq for a bunch of crap the standard library is missing i implemented .Mode() using (abusing) .GroupBy() yesterday so thanks
|
# ? Jul 9, 2016 15:07 |
|
on the subject of weird pointless heap allocations you can create in .NET without noticing, i have recently learnt that in F#code:
code:
https://dotnetfiddle.net/QkEJiM -> choose "view IL" in the menu at the top right basically the answer to 90% of questions about f# and performance seems to be "inline that poo poo"
|
# ? Jul 9, 2016 15:24 |
|
in f# if your let/use binding has no parameters it will make a thing and if it has parameters it will define a function. it's a pretty straightforward rule.
|
# ? Jul 9, 2016 15:35 |
|
Bloody posted:what is worse to work with: plangers or people who smugly insist that c++ is the best because performance c++ users
|
# ? Jul 9, 2016 16:07 |
|
Wheany posted:Theory: let's use a dsl so that non-coders can write <whatever code> this why carmack abandoned quake-c and scriptable engines - they made quakec so "the coders didn't have to write the code and the level designers could" - they found that they spent most of the time debugging the quakec code - ah gently caress it, let's write our logic in the same language so we have the tools or "let's write a programming language for non programmers" is doomed from the outset with a patronising attitude and a naivety of the problem to solve
|
# ? Jul 9, 2016 16:10 |
|
comedyblissoption posted:c++ ... holy gently caress at the syntax
|
# ? Jul 9, 2016 16:28 |
|
tef posted:c++ users they're like the lisp users of systems programming no matter what you do, it could always be done faster in C++ any sufficiently complicated abstraction could be made easy by using an esoteric feature of the language out of common usage that sort of person who tries to win arguments with the entirety of the language at their fingertips but unwilling to concede that production use requires subsetting such a beast of a language. you can tell c++ has the lisp problem because heavy users come down with a bad case of making a new programming language that's just a little bit different sure, it is limiting to bring in constraints and work in a less powerful language but with constraints come focus many people think the measure of strength of a language is the lack of barriers, or a pay-as-you-go mentality for features. after living with the wreckage of this mentality, i think language designers have a responsibility to limit and not empower programmers, there are enough reckless languages already. i mean pay-as-you-go works out in your favor when you don't use a feature, but when 90% of you code does maybe that whole "no-gc" thing doesn't work out a programming language is kinda an executable manifesto on how the author believes you should be programming, and when you don't limit it in scope you punt that decision to the developers and this is how you end up with 60 page coding conventions on how to write acceptable C++
|
# ? Jul 9, 2016 17:57 |
|
devs: let's spend a lot of time removing runtime tagging of data for speed ops: gently caress you too buddy
|
# ? Jul 9, 2016 17:58 |
|
|
# ? Jul 9, 2016 18:01 |
|
so ue4 uses c++.. is it appropriate there?
|
# ? Jul 9, 2016 18:02 |
|
echinopsis posted:so ue4 uses c++.. is it appropriate there? probably, because there isn't a great option for gamedev because performance is the most user-facing feature i mean i'm not saying responsiveness isn't important but i've found that is a bit more architectural than a always down to language choice alone. but a whole bunch of gamedev stuff is done in c# or even javascript atop (like unity). going down to a language like C++ is fine, and often the entirety of a project means that it is reasonable to write at that level of control and responsibility but in the end poo poo gets glued together to make stuff happen, so the idea that instead of "let's write our own special dsl to let people avoid coding" is bad but the idea of "well, let's do the heavy lifting in c++ and punt out to lua, c#, js" is very, very good
|
# ? Jul 9, 2016 18:11 |
|
when i'm talking about c++ users I mean the people who think all of their code is important and thus must be written in a macho, powerful, unconstrained, heavy duty language because if it isn't performance critical or heavy lifting, why would anyone look up to them
|
# ? Jul 9, 2016 18:12 |
|
lol at anyone who thinks their code is important
|
# ? Jul 9, 2016 18:16 |
|
yeah c++ definitely has a machismo thing going on c is still the only game in town for embdev though.
|
# ? Jul 9, 2016 18:17 |
|
GameCube posted:wheres ur github https://github.com/Luigi30/polygon?files=1 The code is mega uggo tho (and you need Open Watcom to compile it)
|
# ? Jul 9, 2016 18:25 |
tef posted:when i'm talking about c++ users I mean the people who think all of their code is important and thus must be written in a macho, powerful, unconstrained, heavy duty language because if it isn't performance critical or heavy lifting, why would anyone look up to them That type of person sounds annoying, but for what it's worth lots of people use C++ for totally different reasons. It's a very popular language in the physics community (probably second to python) and I've never met anyone who uses it out of a machismo thing. A lot of people legitimately do need the speed (especially simulation writers), and there are a lot of physics libraries written in it (most notably ROOT for particle physics). Also everyone's advisors know it, so you can ask them coding questions whereas if you use something like Java you are on your own. (Also lol at using Windows in physics academia, so the entirely of the .NET languages are no-gos from the start.)
|
|
# ? Jul 9, 2016 18:51 |
|
Luigi Thirty posted:https://github.com/Luigi30/polygon?files=1 this code is actually very clean. you must have autism, congrats
|
# ? Jul 9, 2016 19:24 |
|
we do c++ in part b/c we need to run our computational stuff on a bunch of platforms and the intersection of supported languages are c and c++
|
# ? Jul 9, 2016 19:39 |
|
AWWNAW posted:this code is actually very clean. you must have autism, congrats Object arrays as parameters are not very forgivable though. C++ code:
|
# ? Jul 9, 2016 20:18 |
|
VikingofRock posted:That type of person sounds annoying, but for what it's worth lots of people use C++ for totally different reasons. It's a very popular language in the physics community (probably second to python) and I've never met anyone who uses it out of a machismo thing. A lot of people legitimately do need the speed (especially simulation writers), and there are a lot of physics libraries written in it (most notably ROOT for particle physics). Also everyone's advisors know it, so you can ask them coding questions whereas if you use something like Java you are on your own. (Also lol at using Windows in physics academia, so the entirely of the .NET languages are no-gos from the start.) i write physics sims in c++ and god do i hate it i hope rust eats a significant chunk of c++'s niche. not in physics (which probably wont happen), in general.
|
# ? Jul 9, 2016 20:25 |
|
anybody who calls themselves a c++ expert is a liar
|
# ? Jul 9, 2016 20:31 |
is there even anything better than saturday afternoon spent trying to plot data from a binary file that has no accompanying information concerning the specifics of the data contained
|
|
# ? Jul 9, 2016 20:35 |
im slowly starting to hate whatever the gently caress python pretends to be doing with memory, and all the means to benchmark it that ive come across insofar
|
|
# ? Jul 9, 2016 22:12 |
|
c++ is good
|
# ? Jul 9, 2016 22:16 |
great, the most usable python profiler, apparently, doesnt work on x64
|
|
# ? Jul 9, 2016 22:20 |
|
i've been seeing a lot of scala job postings around so i've started lookng into scala and it looks fairly ok to me. don't see a lot of love for it on the internet though. what are its hidden horrors?
|
# ? Jul 9, 2016 22:44 |
|
NihilCredo posted:i've been seeing a lot of scala job postings around so i've started lookng into scala and it looks fairly ok to me. don't see a lot of love for it on the internet though. what are its hidden horrors? jvm, akka takes a while to get used to, community isn't really that big and there's only one big corporate "sponsor" - they used to be called typesafe and renamed recently iirc - and outside of akka I don't think their stuff is that great
|
# ? Jul 9, 2016 22:53 |
|
NihilCredo posted:i've been seeing a lot of scala job postings around so i've started lookng into scala and it looks fairly ok to me. don't see a lot of love for it on the internet though. what are its hidden horrors? sulk did scala in a previous job and hated it
|
# ? Jul 9, 2016 22:55 |
|
uncurable mlady posted:jvm, akka takes a while to get used to, community isn't really that big and there's only one big corporate "sponsor" - they used to be called typesafe and renamed recently iirc - and outside of akka I don't think their stuff is that great scala suffers a bit from being a hybrid in that the language has too many features. make sure you're in a team with code review though and you'll be fine e: also ask in the interview about scala implicits and if they don't immediately spit on the floor (or in your face) then walk out.
|
# ? Jul 9, 2016 23:02 |
|
I don't really have much animus against scala other than a generic preference for declarative over OO and a dislike of the jvm also procedural programming is good imo death to state kitten emergency fucked around with this message at 23:06 on Jul 9, 2016 |
# ? Jul 9, 2016 23:03 |
|
that said you can use scala in an almost purely functional style so that's why I don't hate it
|
# ? Jul 9, 2016 23:08 |
|
i'm reading about vector graphics since i'm doing wireframe polygon stuff and lolquote:There are several methods for producing the clock rates we need. Atari used Binary Rate Multipliers (BRMs). A BRM is a counter that divides the input clock by a digital number. Although the pulses it produces are not guaranteed to be evenly distributed through the counting cycle they will be close enough for our purpose.
|
# ? Jul 10, 2016 00:33 |
|
fritz posted:sulk did scala in a previous job and hated it scala is piss that said it has a lot of capabilities as a language as gonadic io said, but the level of disparity in code complexity can be very great. i don't see much reason in using it over java 8 or maybe even clojure if you had the option because scala's tooling is awful.
|
# ? Jul 10, 2016 00:55 |
|
i've never written scala professionally, it seems like if a team is using it strictly as a nicer java it'd be okay but if they're trying to take advantage of the entire kitchen sink it hands you it'd be horrifying. y/n?
|
# ? Jul 10, 2016 01:16 |
|
scala might be good if you're using it as jvm-haskell. scala is probably bad if you're using it as java++
|
# ? Jul 10, 2016 01:49 |
|
so you're basically saying that if you don't use it for anything real, it'll be fine?
|
# ? Jul 10, 2016 01:57 |
|
|
# ? May 11, 2024 10:17 |
"my use case is the only real use case"
|
|
# ? Jul 10, 2016 02:05 |