|
Internaut! posted:still usin CVS crew checkin in it's faster
|
# ¿ May 9, 2012 03:30 |
|
|
# ¿ May 9, 2024 01:02 |
|
rotor posted:the cool thing is when someone loses a bunch of work because they accidentally delete their local repo if you or somebody you employ deletes their local repo you deserve what you get `rm` is the scrubbiest command, an admission of failure
|
# ¿ May 9, 2012 15:09 |
|
rotor posted:yeah actually it is. if the network or server or power goes out with the kind of regularity that makes you plan around it, get some decent infrastructure, jesus git is that decent infrastructure
|
# ¿ May 9, 2012 15:43 |
|
Internaut! posted:and not because we can literally measure the impact to our profits when our trade data falls out of byte alignment in memory for example lmao this scrub doesn't use tools that guarantee constraints like that if poo poo can just "fall out of byte alignment" you might as well be programming in php
|
# ¿ May 9, 2012 17:56 |
|
Sweevo posted:make nodes in large data structures align on 16-byte bounderies, you can fill a cache line with one read, plus use the segment registers to point to the one you want, then optimise by hardcoding the offsets for each field sounds boring, compiler should get it
|
# ¿ May 9, 2012 18:01 |
|
Deuterieux posted:holy poo poo, my mouth dropped open when they started with making not-so-subtle "we can get you fired we know your boss" threats and her boss chimes in and is all "um... no." yeah pretty pleased with basho management these days https://twitter.com/antonyfalco/status/200034239700414464
|
# ¿ May 9, 2012 22:29 |
|
Sulk posted:1) what kind of a name is shanley 1) the name of somebody i respect a lot 2) you should never post anything on yospos either
|
# ¿ May 10, 2012 02:13 |
|
CaptainMeatpants posted:lol @ self-important manchildren unable to run companies glad you said "companies" and not "businesses" because geeklist is mos def not a business
|
# ¿ May 10, 2012 02:34 |
|
trex eaterofcadrs posted:that's actually not a horrible strategy; spin up worker vm's with huge heaps every time you need to run a job and reclaim them immediately when the job is complete. that's what unicorn (ruby app server) does
|
# ¿ May 13, 2012 04:22 |
|
rotor posted:
|
# ¿ May 13, 2012 04:30 |
|
rotor posted:the unicorn doesn't work! and for things unicorn is bad at: http://rainbows.rubyforge.org/
|
# ¿ May 13, 2012 04:31 |
|
Char posted:sure i mean if you're trapped in some sitcom version of 2007 or you habitually ignore languages because of imaginary people who use them then i guess this post makes sense
|
# ¿ May 13, 2012 13:58 |
|
Internaut! posted:do people pay cash dollars for solutions built on these things even idiot idea people with money have heard of "ruby on rails," and they don't care about the details
|
# ¿ May 13, 2012 20:08 |
|
rotor posted:So the idea that you just want to get your poo poo up and functional asap and gently caress scaling until it actually becomes an issue is a perfectly reasonable choice in some circumstances. ding ding ding
|
# ¿ May 13, 2012 22:11 |
|
Internaut! posted:ah I thought this was something you would use instead of rails
|
# ¿ May 14, 2012 04:48 |
|
Sweevo posted:wanky academic papers: the only thing haskell programmers will ever write you forgot Nomnom Cookie posted:write a monad tutorial
|
# ¿ May 14, 2012 18:13 |
|
Markov Chain Chomp posted:bytes are not integers and i will fight anyone who says otherwise yeah pretty much what's that quote about "of all the six numeric types in java none match the ones identified by aristotle" or whatever
|
# ¿ May 16, 2012 05:26 |
|
actually if you count the totally broken and stupid way numeric primitives and numeric objects work in java it has way more than six
|
# ¿ May 16, 2012 05:27 |
|
yaoi prophet posted:hurf durf i think the integers are fundamental because i'm a babby intuitionist who thinks that mathematics has to correspond to the real world lol if you don't think integers are fundamental what are you some kind of idiot who never learned to count
|
# ¿ May 16, 2012 06:19 |
|
My Linux Rig posted:The company I work at has started a new project with modules in Coldfusion it is
|
# ¿ May 16, 2012 22:27 |
|
Dr. Honked posted:nothing wrong with python, enjoy your journey into code
|
# ¿ May 17, 2012 03:59 |
|
ppp posted:just use a real compiler unironically this microsoft writing bad software that doesn't keep up to standards shouldn't be a surprise to anyone
|
# ¿ May 17, 2012 04:13 |
|
trex eaterofcadrs posted:in my head i thought "no way" can't you just pass a block to the iterator instead of doing scrub poo poo like writing your own loops? code:
|
# ¿ May 17, 2012 16:04 |
|
Shaggar posted:this syntax is so gross lots of that is comment code:
code:
|
# ¿ May 17, 2012 16:37 |
|
tef posted:maybe python isn't the best language to write million loc middleware in. who knew. no language is the best for a million loc project projects like that don't deserve to exist if you can't be succinct about your ideas then it's a bad idea
|
# ¿ May 17, 2012 17:18 |
|
Sweeper posted:Some things just require a lot of code yeah but not a million lines the examples are always poo poo like controllers for physical processes (automotive ECUs, rocketry controllers, etc) but behind the "million lines of C" is a few thousand lines of matlab or something that actually compiles to the million lines facebook compiles to 1.5gb of binary but you can bet it's not anywhere near 1.5gb of php source or even 1.5gb of C++ intermediate code what you have to do is design a language that makes solving your problem succinct and straightforward, and it turns out java and xml are not that language
|
# ¿ May 17, 2012 17:28 |
|
trex eaterofcadrs posted:if you dont write programs to write programs for you you are misusing the computer
|
# ¿ May 17, 2012 17:52 |
|
JawnV6 posted:i think single threaded, isn't the entire world also like that!??!? no, but fortunately there's abstraction layers to allow you to reason synchronously about a concurrent world threads with shared state are not that abstraction
|
# ¿ May 17, 2012 18:46 |
|
trex eaterofcadrs posted:groovy's a little better than that if we're golfing it, ruby: code:
|
# ¿ May 17, 2012 21:16 |
|
Internaut! posted:eh assuming that reads "print all members of this set" that seems like good syntax to me it implicitly calls to_a (to array) on the range 1..5, expands it out into multiple arguments, and calls puts with them code:
|
# ¿ May 17, 2012 21:34 |
|
JawnV6 posted:i fully expect these multiple statements in my high level language, possibly separated by control flow, to be atomic i want them to be useful, and it turns out that you can either not share anything (the right way) or use a bunch of janky ('cause they're easy to gently caress up) locking constructs to make them useful
|
# ¿ May 17, 2012 22:33 |
|
JawnV6 posted:lock bt[rs]l ought to be enough for anyone anyone meaning implementors of compilers erlang/otp supremacy
|
# ¿ May 17, 2012 22:47 |
|
Internaut! posted:eh I look at actors in erlang/scala/clojure/etc and without looking too closely I know if they're both easy and foolproof there must be a catch and it's almost certainly performance it probably is performance but the erlang app i work with is i/o bound not cpu bound so it's not that important
|
# ¿ May 18, 2012 00:34 |
|
Mr Dog posted: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. these are kind of the same thing though and it enables some cool stuff! the "rack" http server protocol for ruby only requires that the response body respond to "each" and yield only String values to the block passed in to each this means that regardless if your app is going to respond with "butts" or if it's going to stream a thing over a network you don't have to have the upstream app server know what's coming out (in the first case, the array of one string ["butts"]; in the second case, some kind of TCPSocket) or necessarily have your app buffer things think of it as a duck-typed http://docs.oracle.com/javase/6/docs/api/java/io/Reader.html
|
# ¿ May 18, 2012 03:06 |
|
rotor posted:finding the end of an array is an exceptional condition so what you're saying is that it works best if the enumeration loop knows what it's enumerating over, so you can have one version for arrays and other bounded collections and one for IO that reads until exhausted, but both present the same API to higher-level code so you don't have to care?
|
# ¿ May 18, 2012 15:47 |
|
Janin posted:this is how python works, fyi nah they're at a lower level if they have to worry about exceptions from the array ending or does python have some kind of __~^each^~__(array, lambda{but only one expression}) function with a terrible name and tenuous object-orientedness?
|
# ¿ May 18, 2012 17:09 |
|
Janin posted:And why would you ever want a lambda with more than one expression? If you really need to use statements in a loop for some horrible reason, just define a local procedure. Because using lambdas/blocks/anonymous functions as a building block for control structures is convenient code:
|
# ¿ May 18, 2012 18:38 |
|
Janin posted:In Python, it's more idiomatic to use existing control structures like 'if' or 'with', rather than inventing your own ad-hoc. Oh yeah, Ruby doesn't have "with" built-in; implementing it is an exercise for the reader https://github.com/meh/ruby-with/blob/master/lib/with.rb
|
# ¿ May 18, 2012 19:39 |
|
Sweeper posted:What book/resource would you recommend for learning the details of ruby? It seems like a really cool language and I only know the basics atm i never really used a book or a blog post about "metaprogramming" work with some libraries that use ruby cleverly and read their source if in doubt read the C source or the rubinius (ruby in ruby) source if still in doubt ask someone
|
# ¿ May 18, 2012 20:48 |
|
|
# ¿ May 9, 2024 01:02 |
|
gently caress ruby, i'm using objectivist-c for my new bitcoin miner
|
# ¿ May 18, 2012 21:00 |