|
also when you do give me wrong data, which there's no compiler-enforced way of checking, i'll probably appear to work fine and you'll only find out when a client gets garbage data and complains if you're lucky, or when you get hacked due to buffer overflow/memory access errors if you're unlucky
|
# ? Jun 17, 2015 23:44 |
|
|
# ? May 23, 2024 16:23 |
|
I'm a C API, and this is Jackass
|
# ? Jun 17, 2015 23:49 |
|
rrrrrrrrrrrt posted:maybe i'm stupid but how would tco be unpythonic when it's basically an implementation optimization? iirc the main argument against it is it fucks up the stack trace which merits consideration, but doesn't seem so bad if you require programmers to mark the tail calls, which is the best way to do it anyway
|
# ? Jun 17, 2015 23:53 |
|
MALE SHOEGAZE posted:how come c apis are so hard a C API doesn't have to be so hard it just looks like the Wayland people just didn't bother to abstract their wire protocol from their API they should have designed their API in terms of windows/surfaces/layers and events, and then defined the mapping between that API and their wire protocol separately (even if the mapping is 1:1). that way people who just want to use the API don't have to even read about protocol details. the API there reminds me a bit of the Amiga/SunView/Xt APIs where you only have a couple of “do_something(int action, ...)” functions in the whole API and everything else is done by calling those with different combinations of varargs. contrast to the classic Mac OS or Win16/Win32 APIs (which copied this style from the Mac), where everything you can do has a separate and clear and much more directly object-oriented API.
|
# ? Jun 18, 2015 00:57 |
|
Rust seems like a really neat lang but is it as thin on "batteries included" or libraries as it seems?
|
# ? Jun 18, 2015 01:04 |
|
rrrrrrrrrrrt posted:Rust seems like a really neat lang but is it as thin on "batteries included" or libraries as it seems? yes. hopefully that part gets better
|
# ? Jun 18, 2015 01:08 |
|
does anybody actually use julia yet
|
# ? Jun 18, 2015 05:13 |
|
yet?
|
# ? Jun 18, 2015 07:49 |
|
I keep asking in financial land, most people haven't heard of it. I was pointed to OpenCPU for R which is pretty nifty, one ETF company adores it.
|
# ? Jun 18, 2015 08:38 |
|
https://onedrive.live.com/view.aspx?resid=1E5AA35A965D3234!26479&ithint=file%2cdocx&app=Word&authkey=!AHbAQ1i_GgwNxJY silverlight program manager says .NET is dying. time to pack it up gents
|
# ? Jun 18, 2015 18:05 |
|
gonadic io posted:hi i'm a c api, please give me three void pointers, and their lengths. i will fill one of them with data and read from the other two. if any of the lengths are wrong, or any of the data uninitialised, or the wrong type pre-casting, you're absolutely hosed. text me
|
# ? Jun 18, 2015 18:07 |
|
pram posted:https://onedrive.live.com/view.aspx?resid=1E5AA35A965D3234!26479&ithint=file%2cdocx&app=Word&authkey=!AHbAQ1i_GgwNxJY lmao quote:If you’re into client-side web development there’s a huge boom in “HTML5” technologies and Javascript skills have never been in more demand. For server-side web development Ruby and Python are still going strong and Node.js is picking up serious steam.
|
# ? Jun 18, 2015 18:20 |
|
Blinkz0rz posted:lmao when was this written? 2006? "HTML5". node js picking up steam!
|
# ? Jun 18, 2015 18:24 |
|
well poo poo now what do we learn, Haskell???
|
# ? Jun 18, 2015 18:26 |
|
pram posted:https://onedrive.live.com/view.aspx?resid=1E5AA35A965D3234!26479&ithint=file%2cdocx&app=Word&authkey=!AHbAQ1i_GgwNxJY gently caress yeah
|
# ? Jun 18, 2015 18:34 |
|
you know what i just noticed. windows 8 was released october 2012 lol
|
# ? Jun 18, 2015 18:41 |
|
wasnt there talk in this or a parallel thread re: how those graphs change if you add .net
|
# ? Jun 18, 2015 18:44 |
|
is tiobe still dumb and utterly meaningless or have they improved it
|
# ? Jun 18, 2015 19:05 |
|
MALE SHOEGAZE posted:how come c apis are so hard because they are made by the kind of people who decide to use C
|
# ? Jun 18, 2015 19:13 |
|
pram posted:you know what i just noticed. windows 8 was released october 2012 lol my decision to move to the part of my team that works in c++ looks better all the time on the other hand, c++
|
# ? Jun 18, 2015 20:08 |
|
Dessert Rose posted:my decision to move to the part of my team that works in c++ looks better all the time if you reject every bad practice inherited from C, C++ becomes awesome
|
# ? Jun 18, 2015 21:19 |
|
Zlodo posted:if you reject every bad practice inherited from C, C++ becomes awesome unfortunately this is impossible because c++ doesn't provide any alternative to the few parts of c that are actually bad, such as the textual-inclusion header system and the awful type declaration syntax
|
# ? Jun 18, 2015 22:14 |
|
idk the type decl syntax doesn't seem super awful to me. auto goes a long way like my team is actively moving code to modern c++ standards as we touch it, everyone involved is super happy about all the new features. i'm learning tons from massive code reviews of my teammates modernizing 10-year-old code (and we're even finding bugs when we do it!) in a couple of years i hope to be almost as good in c++ as i am with c#. it already feels much easier to express ideas in the language than it did six months ago so i think it's really helping me grow
|
# ? Jun 19, 2015 00:38 |
|
Dessert Rose posted:idk the type decl syntax doesn't seem super awful to me. auto goes a long way I wanna make this transition too for some reason but I have no real opportunity to besides trying to get good at modern C++ on my own time and gently caress that really. I haven't touched C++ since visual studio 6 and MFC
|
# ? Jun 19, 2015 01:43 |
|
it's worth getting good enough at it to find a job in it, and of course once you work in something every day you get better at it way way faster. c++ is going to be around for a long long time. rust wants to replace it but that isnt going to happen for decades. by the time rust gains any foothold worth speaking about all the useful concepts will be glued onto c++. it also has the added bonus that most new cs grads are going to look down their noses at c++ jobs because it's old and icky and doesn't shield them from all that nasty "memory management" stuff that they have gotten to essentially ignore. it's also super useful for embedded programming because you only pay for the abstractions you actually use (in theory, anyway), so you get to pick and choose what you want to spend your precious storage/memory/cpu cycles on, whereas using c you have had the choice made for you: you get nothing, and you'll like it you could make worse investments of your time than learning c++ if you want to advance your career and your understanding of programming.
|
# ? Jun 19, 2015 01:55 |
|
also all the really fun poo poo is written in c++, assuming you think optimizing your string classes to save 8 bytes per instance is fun
|
# ? Jun 19, 2015 01:56 |
|
Dessert Rose posted:also all the really fun poo poo is written in c++, assuming you think optimizing your string classes to save 8 bytes per instance is fun sproingggg im hard
|
# ? Jun 19, 2015 02:07 |
Soricidus posted:unfortunately this is impossible because c++ doesn't provide any alternative to the few parts of c that are actually bad, such as the textual-inclusion header system and the awful type declaration syntax actually, if you just wait until c++17 TR1,
|
|
# ? Jun 19, 2015 02:16 |
|
The best use of cpp is to write better languages to use
|
# ? Jun 19, 2015 02:21 |
|
VikingofRock posted:actually, if you just wait until c++17 TR1, and then wait for your compiler to implement it, (ms' compiler is just now finishing up c++11 support. constexpr is still unimplemented, lol)
|
# ? Jun 19, 2015 02:22 |
Dessert Rose posted:(ms' compiler is just now finishing up c++11 support. constexpr is still unimplemented, lol) Oh my god that's humiliating. I just checked and gcc has had constexpr support since 2011; clang since 2012.
|
|
# ? Jun 19, 2015 02:32 |
|
VikingofRock posted:Oh my god that's humiliating. I just checked and gcc has had constexpr support since 2011; clang since 2012. i think we got variadic templates, like, this year
|
# ? Jun 19, 2015 02:39 |
|
Maybe use a real compiler and not Microsofts abortion
|
# ? Jun 19, 2015 03:32 |
|
i thought i heard ms is going to incorporate clang into vs somehow
|
# ? Jun 19, 2015 03:36 |
|
Valeyard posted:try out RequireJS and let us know how it works because i want to know too i ended up doing this with gulp, which i like a lot. basically it has a plugin to grab all your bower deps from the json file and then you can concat/minify them code:
|
# ? Jun 19, 2015 04:01 |
|
require.js has a tool that will do that for you based on ur requires + other stuff.
|
# ? Jun 19, 2015 04:03 |
|
Bloody posted:Maybe use a real compiler and not Microsofts abortion short of this happening: fritz posted:i thought i heard ms is going to incorporate clang into vs somehow i'm pretty much stuck using our own tools
|
# ? Jun 19, 2015 04:13 |
|
Dessert Rose posted:short of this happening: Oh you work for Microsoft
|
# ? Jun 19, 2015 04:17 |
|
Bloody posted:Oh you work for Microsoft yeah, you can update your yospos.xls
|
# ? Jun 19, 2015 04:19 |
|
|
# ? May 23, 2024 16:23 |
|
browserify is pretty good if you're already using gulp and minifying all js to a single file. you can pretend its a real language with imports
|
# ? Jun 19, 2015 04:25 |