|
Shameproof posted:I fiddled around with Reactive Extensions for C# and it's pretty neato. It changes event subscriptions into variables that sort of represent a garden hose that sprays events. You screw different nozzles onto the garden hose. are they anything like golang's channels?
|
#
?
Dec 5, 2012 23:29
|
|
|
|
| # ? Jun 2, 2024 23:29 |
|
|
Cocoa Crispies posted:assuming it's the same poo poo as ironruby: typical microsoft project follow-through (lack thereof) that and it didn't see much adoption as the java ports. third party library components tend to be the missing feature.
|
|
#
?
Dec 5, 2012 23:29
|
|
|
Now I want to make my pi-calculus language again. tef, fancy writing me a parser because I am lazy?
|
|
#
?
Dec 5, 2012 23:33
|
|
|
WHOIS John Galt posted:are they anything like golang's channels? iirc rx is billed as 'linq for events'. if you imagine that iterators are 'pull based' computation, you can sorta imagine that observers are 'push based'. this is the whole 'the iterator is the dual of the observer' (and for those following at home, one of the big differences between python and ruby) anyway, you can do similar sorts of operations on both. what rx does is allow you to use the linq methods to chain up stuff for observers. or i'm talking out of my butt
|
|
#
?
Dec 5, 2012 23:36
|
|
|
Zombywuf posted:Now I want to make my pi-calculus language again. tef, fancy writing me a parser because I am lazy? sure. write me the grammar in prolog and i'll send you a parser
|
|
#
?
Dec 5, 2012 23:36
|
|
|
Zombywuf posted:Now I want to make my pi-calculus language again. tef, fancy writing me a parser because I am lazy? make it a toenail clipping language
|
|
#
?
Dec 5, 2012 23:38
|
|
|
Linq is the best
|
|
#
?
Dec 5, 2012 23:41
|
|
|
Notorious b.s.d. posted:threading isn't just broken, it's so broken than Google, Inc. couldn't fix it with practically infinite budget: unladen swallow first abandoned plans to extract the GIL, and then scaled back plans for JIT or GC fixes, then just gave up entirely on all their project goals there are other reasons why python isn't a great language for just in time compilation. unladen swallow was a small experiment that failed, and was anticipated heavily by readers of reddit and hacker news, waiting for some new aspect of the gil to complain about. quote:the alternative plan was "Go." Google found it easier to develop a brand new compiled language in-house than to fix the most obvious bugs in python this post seems to imply that you think google is making these decisions at a very high level. i was under the impression that a bunch of old plan9 diehards working at google, hated using c++ and python to write simple network services. so they took a bit of plan9 c, limbo, and alef and mashed it together with a static type system built around interfaces. quote:also: garbage collection. every python program leaks memory because python uses refcounting. python uses refcounting because guido didn't want to break c module compatibility in like 1996. also c coders are terrible and it turns out refcounting and a big lock means they don't have to think too hard about making code that works. it turns out people other than web developers use python, and as such are wedded to c-module compatibility. breaking the abi is telling your users and developers to go gently caress themselves. then again, you can look at ruby for the success of getting c programmers to write gc-friendly code http://timetobleed.com/the-broken-promises-of-mrireeyarv/ (i.e by sprinkling magic volatile dust over the code)
|
|
#
?
Dec 5, 2012 23:44
|
|
|
Jonny 290 posted:regardless i think that we all can agree that the answer to "what is perl 5's replacement?" will never be "perl 6" dont be mean :L
|
|
#
?
Dec 5, 2012 23:47
|
|
|
tef posted:then again, you can look at ruby for the success of getting c programmers to write gc-friendly code http://timetobleed.com/the-broken-promises-of-mrireeyarv/ (i.e by sprinkling magic volatile dust over the code) the worst parts of ruby & python: they're written in c by c programmers
|
|
#
?
Dec 5, 2012 23:50
|
|
|
perl 5 is used in enterprise. it will never die. nor will the local copy ever be upgraded.
|
|
#
?
Dec 5, 2012 23:51
|
|
|
Cocoa Crispies posted:the worst parts of ruby & python: they're written in c by c programmers ergo pypy is the best 
|
|
#
?
Dec 5, 2012 23:51
|
|
|
Use rubinius then
|
|
#
?
Dec 5, 2012 23:52
|
|
|
Nomnom Cookie posted:I don't like python...it just doesn't get me hot like Scala can
|
|
#
?
Dec 5, 2012 23:53
|
|
|
Also theres no official spec for ruby? So just copy and replace some lisp poo poo and claim its a new ruby
|
|
#
?
Dec 5, 2012 23:53
|
|
|
0xB16B00B5 posted:Also theres no official spec for ruby? So just copy and replace some lisp poo poo and claim its a new ruby there is a standard for ruby 1.8 you still have to read the ruby source code to understand what it does
|
|
#
?
Dec 5, 2012 23:55
|
|
|
gucci void main posted:intellij idea 12 is out!!! 
|
|
#
?
Dec 5, 2012 23:56
|
|
|
0xB16B00B5 posted:Use rubinius then bundle exec rake 2.97s user 0.21s system 99% cpu 3.199 total RBXOPT=-X19 bundle exec rake 6.27s user 0.29s system 126% cpu 5.197 total that's p. decent, only about half as fast
|
|
#
?
Dec 6, 2012 00:02
|
|
|
tef posted:perl 5 is used in enterprise. it will never die.
|
|
#
?
Dec 6, 2012 00:08
|
|
|
tef posted:ergo pypy is the best i want to write a perl 6 implementation in perl6 and call it pepe :3 but that would take years so, no
|
|
#
?
Dec 6, 2012 00:09
|
|
|
gucci void main posted:intellij idea 12 is out!!! paying money for a java ide
|
|
#
?
Dec 6, 2012 00:10
|
|
|
Gazpacho posted:good, perl 6 is madness ya i know! its like, perl 5 plus built in moose and like, some better syntax and poo poo. crazy
|
|
#
?
Dec 6, 2012 00:10
|
|
|
TiMBuS posted:i want to write a perl 6 implementation in perl6 and call it pepe :3 you could write it in a subset of perl first  and then you could compile it to an all new intermeditary format without a vm for it.
|
|
#
?
Dec 6, 2012 00:11
|
|
|
haaaaa 
|
|
#
?
Dec 6, 2012 00:13
|
|
|
TiMBuS posted:haaaaa do it. you've found your life's purpose. nothing else you in do in your life could possibly be more important than this.
|
|
#
?
Dec 6, 2012 00:14
|
|
|
Jerry SanDisky posted:paying money for a java ide Oracled once again
|
|
#
?
Dec 6, 2012 00:15
|
|
|
TiMBuS posted:haaaaa twenty years from now people will reverently quote your listserv posts in their v-mail sigs and on Google++. there will be bugs, theorems, and folklore with your name. your dumb family and stuff will wither and decay, but your legacy of implementing perl6 in perl6 will live on. do it you pussy
|
|
#
?
Dec 6, 2012 00:17
|
|
|
WHOIS John Galt posted:do it. you've found your life's purpose. nothing else you in do in your life could possibly be more important than this. yes. you could call the language 'Not Yet Perl', (NYP), and call the the vm after that april fools joke about a vm for perl and python, what was it called.
|
|
#
?
Dec 6, 2012 00:17
|
|
|
TiMBuS posted:dont be mean :L  disclaimer: i have one
|
|
#
?
Dec 6, 2012 00:21
|
|
|
now it doeosnt take a fool to know youre making a joke at me, horse man and tef, but i could swear that if i read between the lines.. maybe youre trying to encourage me??Jonny 290 posted:
a really neat well organized set of tools?
|
|
#
?
Dec 6, 2012 00:24
|
|
|
it wasnt really a nuanced barb
|
|
#
?
Dec 6, 2012 00:25
|
|
|
Jonny 290 posted:it wasnt really a nuanced barb neither is your mother
|
|
#
?
Dec 6, 2012 00:25
|
|
|
tef posted:you could write it in a subset of perl first so glad #cobol talked me out of contributing to parrot
|
|
#
?
Dec 6, 2012 00:28
|
|
|
Jonny 290 posted:neither is your mother my name is also barb
|
|
#
?
Dec 6, 2012 00:30
|
|
|
writin state machines at work left n right, life owns
|
|
#
?
Dec 6, 2012 00:31
|
|
|
TiMBuS posted:haaaaa that is the good news. even when/if perl 6 is ready, no-one will use it. people hate complete breaks! ruby 1.9 changed a bunch of stuff for unicode and didn't really make things easier, just possible, because japanese users have to deal with hosed up systems (lots of mixed encodings, even in file paths). ruby 2.0 was close to adding a new feature which would have broken lots of things, and it got to the point where some of the non-mri implementations were calling for mutiny. meanwhile, python3 didn't really change much in total (with some of the changes backported), but changed the language and the abi at the same time. this created a new chicken and egg problem, so third party library authors had to be hounded into porting their code across, in order to entice users. people hate backwards and forwards incompatibility. when you write a new incompatible version of your language, you are forking it. migration is hindered when you user base rely on third party components. often it is just as easy to move to a different language than to move to a newer version. (c.f that boeing study arguing that changing more than 25% is about as buggy as replacing wholesale)
|
|
#
?
Dec 6, 2012 00:32
|
|
|
Otto Skorzeny posted:so glad #cobol talked me out of contributing to parrot turing tar pits are everywhere in programming language implementation
|
|
#
?
Dec 6, 2012 00:33
|
|
|
inspired by earlier python chat made a markov chain generator for jerkcity dialog here sample output: quote:BILLIONS TO YOUR WITNESS I KEEP IN FACT POKER IS JERKCITY GETS YOU TALENTLESS FAT GUY ONLY SEE SSH AND WE NEED SOMEONE TO CLINTON'S COCK FOLLOWED CLOSELY AND MELTING INTO THE F)(*DV:#JDFLS
|
|
#
?
Dec 6, 2012 00:37
|
|
|
tef posted:that is the good news. even when/if perl 6 is ready, no-one will use it. perl 6 isnt designed to be an upgrade path from 5. its also not designed to replace 5. backkwards compat is just going to be done by embedding perl5 in it (this is sort of already done)
|
|
#
?
Dec 6, 2012 00:38
|
|
|
|
| # ? Jun 2, 2024 23:29 |
|
|
tef posted:that is the good news. even when/if perl 6 is ready, no-one will use it. the only thing in ruby 1.9 they added that i know i hate is the symbol: value hash, because a lot of times you end up seeing symbol: :symbol and it looks ugly as sin. i don't know why anyone uses it when :symbol => value is so much clearer and you only use like two more characters
|
|
#
?
Dec 6, 2012 00:39
|
|
