|
i'll be honest if python's standard library wasn't such a loving disorganised mess I'd probably use python more yes let's mix every identifier capitalisation style ever. also let's have "pickle" and "cpickle" and the only difference between the two is that the second has a faster(?) underlying implementation. ok chief. also the GIL is frankly shameful. basically what i'm saying is that i'm a mong
|
#
¿
Apr 29, 2012 18:27
|
|
|
|
| # ¿ May 3, 2024 17:04 |
|
|
quote:UML  quote:Haskell Ah yes, functional programming. I think I'll bake a cake. How do you think I formulate this problem in my head: 1. Grab bowl and ingredients, start the oven heating up in the background 2. Add flour, sugar, eggs, etc to bowl 3. Stir until consistent 4. Wait for oven to get hot if it isn't already 5. Insert dough into oven, wait X minutes or: Let F(tau, xi, aleph, B, W, w1, w2, n) be a regular functional of the third kind from the domain of all functions on the finite Abelian field blah blah blah blah blah p.s. monads yeah no
|
|
#
¿
Apr 30, 2012 10:44
|
|
|
um doesn't setting a condition variable require a syscall?  0MQ looks kinda kewl tho
|
|
#
¿
Apr 30, 2012 11:11
|
|
|
oh wow Python 3 is actually quite nice.
|
|
#
¿
Apr 30, 2012 19:16
|
|
|
I was gonna write some code, but then I got high I was gonna wait for Eclipse to load, but then I got high My test cases are broken, and I know why Because I got high, because I got high, because I got high
|
|
#
¿
Apr 30, 2012 19:22
|
|
|
uh but the one thing I don't like is how they deprecated formatting using the % operator "string literal".format(blah, blah) just looks hella gross. It's like how Ruby lets you do stuff like 1234.butt() and then acts like that's a good thing.
|
|
#
¿
Apr 30, 2012 19:23
|
|
|
Fren posted:how is it any different from "butts".upper() it isn't, both are terrible.
|
|
#
¿
Apr 30, 2012 19:53
|
|
|
bitch you best step off i'mma loving cut you
|
|
#
¿
May 1, 2012 22:01
|
|
|
Hey AV go read the following poo poo on Wikipedia Data structures: - Dynamic arrays - Linked lists - Hash tables - Heaps (n.b. not 'the dynamic memory pool', the data structure) - Binary trees (and the main ways to make them not suck: AVL Trees and Red-Black trees. Do not attempt to implement the latter, it is painful). Algorithms: - Binary search - Heapsort (you know about heaps now so this is ez) - Quicksort - Insertion sort (token 'suboptimal' example that's fast on 'small' lists) Concurrent programming: - Threads - Mutexes - Condition variables Maybe optionally learn how malloc()/free() actually work as well and maybe what a cache line is. Congratulations, that probably took you a week or two and this is like 99% of the actual theory that you need to know for professional programming, not including writing more than a few thousand LOC without it turning into an incomprehensible soup, and that's craft and not science so it simply takes practice. Also everyone disagrees on how best to do it and everyone thinks everyone else sucks at it anyway. Oh yeah Wikipedia goes out of its way to explain things in the worst way imaginable so maybe read this poo poo somewhere else idk.
|
|
#
¿
May 5, 2012 23:40
|
|
|
ppp posted:it just works. but i have no clue how *le random flash 0day exploit, steals my le ramz* - failcat reddit raus
|
|
#
¿
May 16, 2012 14:28
|
|
|
kill everybody
|
|
#
¿
May 17, 2012 02:22
|
|
|
So I just learned that Python iterators throw an exception to indicate end of sequence what the gently caress kind of amateur hour horse poo poo is this
|
|
#
¿
May 17, 2012 15:40
|
|
|
It wouldn't have been hard to retrofit a non-retarded behaviour in Python 3, either, just add a has_next or __has_next__ or whatever method to the iterator protocol and then keep the existing StopIteration behaviour as a genuine "you did something you weren't supposed to do, bonehead" exception, albeit a poorly named one. Even a __boolean__ coercion, while kinda terrible, would have been better than just leaving that wtf in its current form.
|
|
#
¿
May 17, 2012 16:01
|
|
|
Shaggar posted:java is the only language i've used that does exceptions correctly. I know I'm probably being trolled but this is actually true Checked exceptions are an excellent thing when they're (a) optional and (b) structured correctly (read: have a common base class if methods tend to throw three or four of them). Java has a useful level of covariance and contravariance as well, which is nice. I just don't like duck typing in general. I mean, strong typing turns a big class of runtime errors into compile errors for not too much hassle, those types give you hyperlinks around the codebase for free, and it's just easier to see what's going on and what the collaborators are in general. I'm trying to fix some stupid loving Python library at the moment and having types in the signatures would be really loving useful, since I can look at a line of code like butt.poke() and have a reasonable chance of actually finding where the implementation of poke() is located. Basically Python is cute but I'm not convinced it scales. Sometimes repetition is a necessary price for clarity, as long as the compiler enforces the correctness of that repetition I don't see what's wrong with it.
|
|
#
¿
May 17, 2012 17:04
|
|
|
there is no way in hell i would use c++ for anything willingly. ever c is an altogether different story optional as in some exceptions can be checked and some exceptions can be unchecked because they represent different categories of exceptional condition and in either case exceptions should be exceptional. i.e. it should be possible to use a program in every possible intended way without exceptions being thrown inside it, anything else is just incredibly lazy poo poo design. also not all large software is middleware and separation of concerns is a thing so i'd rather not learn how an entire gigantic codebase works if i just want to alter some small piece of it (cute is probably an overly negative and dismissive adjective to use to describe a programming language. pretend I said "elegant")
|
|
#
¿
May 17, 2012 17:19
|
|
|
if you are iterating over something that's simultaneously changing while you're iterating over it then an iterator is probably not exactly the tool you want. no commonly used I/O system lets you send something to someone and then subsequently change your mind and un-send it. also if you really wanted to abuse iterators for I/O (let's make these completely different things look exactly the same) then you'd probably make next() block or something and only make has_next() return false on EOF. if it isn't changing while you're iterating over it then constant data can't very well have a race condition can it (please don't bring up concurrent splay trees or something stupid like that) C++ (ew) makes iterators pointer-shaped. Java's iterator interface uses hasNext() and next(). Literally every other imperative language with iterators manages to do this just fine without needing to use nonlocal jumps. Stop making up ridiculous contrived what-if cases to justify this design decision please.
|
|
#
¿
May 18, 2012 03:00
|
|
|
tef is like the ultimate YOSPOS hipster yeah i'm into some p obscure software engineering philosophies you've probably never heard of them i'll leave you with this: the same people who propose new langauges/operating systems to solve all the ills of the programming world are the same ones who propose forming new political parties to solve the ills of the physical one people's front of judea etc
|
|
#
¿
May 19, 2012 19:27
|
|
|
I think Cincinnatus was the most well-known example, though that wasn't so much a case of "Yo you guys suck and i'mma fix your poo poo" as "omg we suck plz fix our poo poo". twice. but yeah SOPA thread is thataway
|
|
#
¿
May 19, 2012 21:38
|
|
|
embedded programming is really fun and everyone should try it sometime provided that you get to use nice hardware anyway. which is approximately never. so ok, stay out it's terrible (i'm lucky, i get to use nice hardware)
|
|
#
¿
May 21, 2012 01:41
|
|
|
univ. of edinburgh has the computer science ("informatics") building in the liberal arts campus and all the other engineering stuff way outside town in the science campus i think they have the right idea
|
|
#
¿
May 23, 2012 11:11
|
|
|
If you develop Win32 software using anything whatsoever other than the WDK then you're doing it wrong visual studio was good back in the Visual C++ 6.0 days and then it got overrun by stupid training wheels for baby languages Sapozhnik fucked around with this message at 14:23 on May 27, 2012 |
|
#
¿
May 27, 2012 14:21
|
|
|
perl is awful, never write anything new in perl that doesn't fit on your screen
|
|
#
¿
May 28, 2012 12:50
|
|
|
Scheme looks really neat, i like simple and elegant things so i think i might like Scheme. YOSPOS plz tell me why Scheme sucks because I don't want to waste a year of my life learning it only to inevitably come to the conclusion that it sucks when i could spend that time eating lard and watching anime instead
|
|
#
¿
Jun 28, 2012 14:09
|
|
|
GNU make owns. The whole "ASCII TABs serve a distinct and crucial purpose in the grammar of this language and ok that was kind of a bad idea in hindsight but I've already got ten users now so welp" thing is unfortunate, but basically anyone who attempts to make an alternative to make because "make sucks" usually doesn't actually understand make and just creates something far less flexible and not that much easier to use. Write a small set of macros to drive the whole process in a master GNUmakefile and then make some Module.mk files or whatever in your source tree that look like code:
|
|
#
¿
Jul 14, 2012 13:47
|
|
|
also is there a word for when someone uses an imperative language to do declarative things (e.g. Scons, writing key-value configuration files in Lua) because holy poo poo i hate that. HTML templating languages are acceptable but every other such usage needs to die.
|
|
#
¿
Jul 14, 2012 13:48
|
|
|
Internaut! posted:I know there's all sorts of object relational frameworks and approaches out there, what's considered today's best of breed in c++/java land? none, they're all horrible. ORMs are like C++ in the sense that it looks great when your project is still in its nascent state, but by the time you realise how truly hosed you are it's impossible to get rid of. personally i did a thing in Java where I can just define Java interfaces with some light annotations and then there's this thing that takes the Class of that interface and gives you back a proxy object. Method calls on that proxy object get transparently turned into stored proc calls and result sets get turned into Java scalars or Lists of beans or w/e. imho this mini-framework seriously fucen owns for my particular application (a butt store) but it certainly isn't a one-size-fits all kind of deal. The best part of it is that there is no code generation involved. C# has no equivalent functionality, which is a bit of a pity. Sapozhnik fucked around with this message at 17:30 on Jul 15, 2012 |
|
#
¿
Jul 15, 2012 17:27
|
|
|
Sounds like a wordier version of my solution tbh. This doesn't need mapping files, there's these things called annotations that have been around for almost ten years now that do the job just fine. I kinda sidestep the transaction thing by just running each SP in its own transaction but this could be integrated fairly easily into JTA i suppose. Spring has seriously gone off the deep end lately, though. I mean first you have this AOP insanity which means that changing the name a method can change its semantics in a manner that's defined in some file on the other side of the Then they decided that autowiring is awesome because a bloo bloo bloo typing is hard. Except now you have objects of class X connected to objects of class Y and you have absolutely no idea how it got that way short of searching for all implementations of an interface within your codebase or something. I have no idea why they'd do either of these things, it's like they're actually trying to make your codebase incomprehensible so that when everything finally goes completely pete tong you have to hire their consultants to help you... oh. Basically writing my own YAML-based IoC thing to replace Spring in my project is now sitting on my to-do list.
|
|
#
¿
Jul 15, 2012 17:42
|
|
|
Oh yes I forgot this is Java where you need a framework to abstract out what framework framework you are using. I have a front servlet that has a hard dependency on Free marker and a bidirectional URL mapping scheme that I wrote, I bet it gives people like that conniption fits Also stop saying "c/c++" please they are not the same thing. "As a man of wealth and taste I enjoy drinking fine scotch/my own urine"
|
|
#
¿
Jul 16, 2012 10:33
|
|
|
ESC? oh, you mean Ctrl-[  (Ctrl-P triggers VIM's built-in fake intellisense thing though and it is tres annoying)
|
|
#
¿
Jul 17, 2012 23:24
|
|
|
i like how everyone itt likes to periodically jizz over the D Language Forum's performance, and my Java web thing is pretty much that fast and i don't even do anything terribly clever with it (there's a bunch of memcached hooks in the code but i haven't actually got around to writing a Cache impl that actually remembers anything. i prob never will because tef says cache invalidation is Hard and tef is always right) postgres and the jvm ft loving w
|
|
#
¿
Jul 22, 2012 15:18
|
|
|
I keep thinking that one of these days there'll be an OOP anti-fad and we can all get that nonsense behind us but nope, 30 years in and here we are. Still, we are at least kinda sorta coming around to the idea that C++ is a Bad Programming Language, so there may be some hope yet.
|
|
#
¿
Jul 24, 2012 16:21
|
|
|
How to do Java web deployment (for babbys) 1. mvn package 2. Copy target/butt-store.war to ur live Tomcat instance's webapps directory I mean I'm not sure how that could possibly be any more straightforward, it really is p sweet. You can't SCP it straight to the webapps directory, you have to SCP it somewhere else and move it first (otherwise it'll try to deploy a partially-transferred WAR and that way lies 503 errors and lots of frantic scrambling). But apart from that it's p sweet. You might be thinking of that old shambling EJB poo poo but lmao if you think anyone uses that anymore. I don't even see the point in using JSP these days, because even with the JSTL it's wordy as fuc and Freemarker is just better. Java owns, Postgres owns. Kill all MySQLs ever, it is an embarassment.
|
|
#
¿
Jul 25, 2012 10:28
|
|
|
homercles posted:$ mvn package apparently so because it does that once ever and then it caches it all in your ~/.m2 directory?
|
|
#
¿
Jul 25, 2012 12:11
|
|
|
I can't get the tomcat admin app to work on my live server lol. and i don't particularly want to risk extended downtime by loving around with it right now. it's some class path poo poo (needs a special class loader to gain super admin powers and the option that says it enables that doesn't actually seem to enable it) so I just push things using SCP/mv like a big god drat idiot deployment over HTTPS to an admin web service would be p sw8 tho.
|
|
#
¿
Jul 25, 2012 16:57
|
|
|
Currently looking at maybe throwing ZeroMQ into the mix so that I can have a middle tier (i have two HTTP things, wanna add a middle tier to do caching and pub/sub between them). ZeroMQ looks really nice but the ~hilariously zany~ documentation is kind of annoying. "hello sir can i interest you in building your business on top of a middleware package that was written by a gigantic man child?" does anyone have any experience with it?
|
|
#
¿
Jul 27, 2012 12:22
|
|
|
pub-sub doesn't really map well to rest
|
|
#
¿
Jul 30, 2012 16:43
|
|
|
tinselt0wn posted:why would you waste time compiling code when scripting has surpassed all compiled code except for a few edge cases that require speed compiled code tends to be statically typed, static typing catches typos
|
|
#
¿
Jul 30, 2012 23:06
|
|
|
i should write a high frequency in prolog
|
|
#
¿
Jul 31, 2012 01:02
|
|
|
tef posted:this is where the tight coupling happens - without a common message format, it is hard to build reusable middleware. without self descriptive messages, it is hard to build loosely coupled components. Neither ZMQ nor HTTP has a defined payload format, what makes them different?
|
|
#
¿
Aug 4, 2012 17:15
|
|
|
|
| # ¿ May 3, 2024 17:04 |
|
|
don't learn to program in java
|
|
#
¿
Aug 5, 2012 13:39
|
|