p-lang thread: (now (have you (problems two)))

The Something Awful Forums > Discussion > Serious Hardware/Software Crap > YOSPOS > p-lang thread: (now (have you (problems two)))

«‹›1784 »

Plastic Snake: Mar 2, 2005; For Halloween or scaring people.

suffix posted:

https://aphyr.com/posts/322-call-me-maybe-mongodb-stale-reads

mongodb... is bad?!

was gonna post this myself. another loving homerun from aphyr, i wish i was 10% as smart as him

also lmao at the comments on the ticket he filed with them https://jira.mongodb.org/browse/SERVER-17975

"working as expected, wontfix"

# ? Apr 22, 2015 05:41

Adbot: ADBOT LOVES YOU

# ? Jun 1, 2024 09:23

Sapozhnik: Jan 2, 2005; Nap Ghost

read jira project name as SEVER there

# ? Apr 22, 2015 06:44

triple sulk: Sep 17, 2014

http://railsbird.tumblr.com/post/117076597115/existential-crisis-at-railsconf

lol rails dev(s) slowly realizing now that rails has been dying for a while

# ? Apr 22, 2015 14:01

Notorious b.s.d.: Jan 25, 2003; by Reene

Plastic Snake posted:

was gonna post this myself. another loving homerun from aphyr, i wish i was 10% as smart as him

it's not that aphyr is smart. i mean, he is, but i don't think that's the key to the "call me maybe" work

it's that aphyr works hard. really, really hard. he's built a huge toolbox of stuff specifically to work on these problems, himself, via relentless grind. if you look at the github committers, 50,000 lines of code changes came from aphyr alone.

i'd really like to have even half the dedication that guy carries into his work

# ? Apr 22, 2015 16:04

fritz: Jul 26, 2003

triple sulk posted:

http://railsbird.tumblr.com/post/117076597115/existential-crisis-at-railsconf

lol rails dev(s) slowly realizing now that rails has been dying for a while

waiter this omakase is stale

# ? Apr 22, 2015 16:37

crazysim: May 23, 2004; I AM SOOOOO GAY

Notorious b.s.d. posted:

it's not that aphyr is smart. i mean, he is, but i don't think that's the key to the "call me maybe" work

it's that aphyr works hard. really, really hard. he's built a huge toolbox of stuff specifically to work on these problems, himself, via relentless grind. if you look at the github committers, 50,000 lines of code changes came from aphyr alone.

i'd really like to have even half the dedication that guy carries into his work

don't just count the lines of code. think of his posts too! and it's not filled with noise. that's quite a few "lines" in itself.

# ? Apr 22, 2015 16:40

Workaday Wizard: Oct 23, 2009; by Pragmatica

his github picture sure is something

# ? Apr 22, 2015 16:42

Plastic Snake: Mar 2, 2005; For Halloween or scaring people.

Notorious b.s.d. posted:

it's not that aphyr is smart. i mean, he is, but i don't think that's the key to the "call me maybe" work

it's that aphyr works hard. really, really hard. he's built a huge toolbox of stuff specifically to work on these problems, himself, via relentless grind. if you look at the github committers, 50,000 lines of code changes came from aphyr alone.

i'd really like to have even half the dedication that guy carries into his work

Yeah that's a great point. I also wish I had 10% of the drive he does. If I didn't already work 8hrs a day on code maybe I'd be more motivated to do it outside of work too.

# ? Apr 22, 2015 16:54

Plastic Snake: Mar 2, 2005; For Halloween or scaring people.

crazysim posted:

don't just count the lines of code. think of his posts too! and it's not filled with noise. that's quite a few "lines" in itself.

Yeah he's got a real knack for technical writing. And endless patience judging by the comments on that jira ticket

# ? Apr 22, 2015 16:55

Plorkyeran: Mar 22, 2007; To Escape The Shackles Of The Old Forums, We Must Reject The Tribal Negativity He Endorsed

endless patience is pretty much a prerequisite for good technical writing on complex topics

# ? Apr 22, 2015 17:38

sarehu: Apr 20, 2007; (call/cc call/cc)

Singapore Prime Minister Lee Hsien Loong says the phrase, "learn you a Haskell for great good"

(a bit after 9:51:)

https://www.youtube.com/watch?v=49YeHvJ6yZg

sarehu fucked around with this message at 22:00 on Apr 22, 2015

# ? Apr 22, 2015 21:57

Blotto Skorzany: Nov 7, 2008; He's a PSoC, loose and runnin'
came the whisper from each lip
And he's here to do some business with
the bad ADC on his chip
bad ADC on his chiiiiip

it's a shame BONUS stopped poasting

# ? Apr 22, 2015 22:02

VikingofRock: Aug 24, 2008

Here's a cool article about the Option monad in Rust. It's nice seeing monads (and the clean code that they create) getting some play in imperative languages.

# ? Apr 30, 2015 06:13

fart simpson: Jul 2, 2005; DEATH TO AMERICA

im a monad

# ? Apr 30, 2015 07:41

Carthag Tuek: Oct 15, 2005; Tider skal komme,
tider skal henrulle,
sl�gt skal f�lge sl�gters gang

get the monad, scholar scholar chill tall

# ? Apr 30, 2015 09:39

cowboy beepboop: Feb 24, 2001

VikingofRock posted:

Here's a cool article about the Option monad in Rust. It's nice seeing monads (and the clean code that they create) getting some play in imperative languages.

looking forward to writing some slow python and php stuff in rust once it hits 1 or maybe 1.1

# ? Apr 30, 2015 11:01

Vanadium: Jan 8, 2005

VikingofRock posted:

Here's a cool article about the Option monad in Rust. It's nice seeing monads (and the clean code that they create) getting some play in imperative languages.

Please stop trying to make Rust look bad by associating it with that weirdo Haskell stuff :argh:

# ? Apr 30, 2015 12:02

gonadic io: Feb 16, 2011; >>=

no monads here guv, just putting a value inside a container and then accessing it with callbacks. completely different and not scary.

# ? Apr 30, 2015 12:12

Dicky B: Mar 23, 2004

a blog post about monad

# ? Apr 30, 2015 12:17

HappyHippo: Nov 19, 2003; Do you have an Air Miles Card?

VikingofRock posted:

Here's a cool article about the Option monad in Rust. It's nice seeing monads (and the clean code that they create) getting some play in imperative languages.

as someone who knows nothing about rust, is the lambda here necessary?

code:

.and_then(|shortest| find_user_by_name(shortest))

could you just do

code:

.and_then(find_user_by_name)

# ? Apr 30, 2015 13:23

fart simpson: Jul 2, 2005; DEATH TO AMERICA

HappyHippo posted:

as someone who knows nothing about rust, is the lambda here necessary?
code:
.and_then(|shortest| find_user_by_name(shortest))
could you just do
code:
.and_then(find_user_by_name)
?

i just tried it out and yes, your way works too

# ? Apr 30, 2015 13:29

gonadic io: Feb 16, 2011; >>=

fart simpson posted:

i just tried it out and yes, your way works too

partial application can be hard to read if you're not used to it, they probably just wanted to keep it simpler

# ? Apr 30, 2015 13:34

Max Facetime: Apr 18, 2009

christ this is so stupid

code:

fn get_user_with_shortest_name(names: Vec<&str>) -> Option<User> {
   get_shortest(names)
    .and_then(|shortest| find_user_by_name(shortest))
    .and_then(|user| json_to_user(user))
}

just make ?. the default dereferencing operator for nullable values and make foo(object) and object.foo() mean the same thing

now you can simply write the thing you ACTUALLY want to write:

code:

fn get_user_with_shortest_name(names: Vec<&str>) -> User {
   names.get_shortest().find_user_by_name().json_to_user()
}

and none of the Option crtap

# ? Apr 30, 2015 15:42

Sapozhnik: Jan 2, 2005; Nap Ghost

i guess instead of having a NULL that's built into the language they created an Option<> type that isn't. About as good an idea as a non-built-in String type imo (i.e. not a good idea).

Still this looks just like java.util.Optional<>, except "and_then" is called flatMap().

# ? Apr 30, 2015 15:47

Sapozhnik: Jan 2, 2005; Nap Ghost

Suspicious Dish mentioned that Red Hat has an employee whose surname is Null, and his name makes their HR systems gently caress up in entertaining ways

this amuses me greatly

# ? Apr 30, 2015 15:50

Subjunctive: Sep 12, 2006; ✨sparkle and shine✨

Mr Dog posted:

Still this looks just like java.util.Optional<>, except "and_then" is called flatMap().

is that bad?

# ? Apr 30, 2015 15:50

Max Facetime: Apr 18, 2009

java.util.Optional is so stupid too

worst new addition of java 8

# ? Apr 30, 2015 15:51

my homie dhall: Dec 9, 2010; honey, oh please, it's just a machine

Mr Dog posted:

i guess instead of having a NULL that's built into the language they created an Option<> type that isn't. About as good an idea as a non-built-in String type imo (i.e. not a good idea).

Optional typing is much better than NULL, hth

# ? Apr 30, 2015 15:51

Sapozhnik: Jan 2, 2005; Nap Ghost

Optional typing is great but it should be built-in, just like strings should be built in, because both of those things are used constantly

java.util.Optional<> is a band-aid for a language that already has null references in it, Rust shouldn't need an identical copy of it if they're starting from scratch.

I use java.util.Optional in new code everywhere. It's clunky but it's a drat sight better than constant "hmm is this parameter nullable or not"

# ? Apr 30, 2015 15:55

Subjunctive: Sep 12, 2006; ✨sparkle and shine✨

what would be different in Rust if it were built in? can you not rely on its presence? is it just the dereference sugar that you miss?

# ? Apr 30, 2015 16:05

Sapozhnik: Jan 2, 2005; Nap Ghost

something like c# 4 (or is it 5?)'s conditional-dereference operator ?? or whatever would make the optional value gonad a lot less verbose.

if you're focusing on a page of code then you shouldn't have a bunch of extra chaff in the way of the meat of the operation.

# ? Apr 30, 2015 16:07

leftist heap: Feb 28, 2013; Fun Shoe

hell, go one step further and bake some sort of generalization of the option type into the language. and then like, some special syntax for doing stuff i dunno i'm just an ideas guy.

# ? Apr 30, 2015 16:16

CPColin: Sep 9, 2003; Big ol' smile.

The worst part of java.util.Optional is that Eclipse still can't apply @NonNull and @Nullable to it. Java 9 had better bring those annotations into the language specification.

# ? Apr 30, 2015 16:24

giogadi: Oct 27, 2009

rrrrrrrrrrrt posted:

hell, go one step further and bake some sort of generalization of the option type into the language. and then like, some special syntax for doing stuff i dunno i'm just an ideas guy.

# ? Apr 30, 2015 16:28

fart simpson: Jul 2, 2005; DEATH TO AMERICA

Max Facetime posted:

christ this is so stupid
code:
fn get_user_with_shortest_name(names: Vec<&str>) -> Option<User> {
   get_shortest(names)
    .and_then(|shortest| find_user_by_name(shortest))
    .and_then(|user| json_to_user(user))
}
just make ?. the default dereferencing operator for nullable values and make foo(object) and object.foo() mean the same thing

now you can simply write the thing you ACTUALLY want to write:
code:
fn get_user_with_shortest_name(names: Vec<&str>) -> User {
   names.get_shortest().find_user_by_name().json_to_user()
}
and none of the Option crtap

but what if get_shortest() returned null? then you need to build explicit null handling into every function? they could have written the option example like this

code:

fn get_user_with_shortest_name(names: Vec<&str>) -> Option<User> {
   get_shortest(names).and_then(find_user_by_name).and_then(json_to_user)
}

# ? Apr 30, 2015 16:29

Arcsech: Aug 5, 2008

Mr Dog posted:

something like c# 4 (or is it 5?)'s conditional-dereference operator ?? or whatever would make the optional value gonad a lot less verbose.

if you're focusing on a page of code then you shouldn't have a bunch of extra chaff in the way of the meat of the operation.

we could always make and_then into an operator like... Hmm, how about >>=

# ? Apr 30, 2015 16:31

fart simpson: Jul 2, 2005; DEATH TO AMERICA

Arcsech posted:

we could always make and_then into an operator like... Hmm, how about >>=

i like this idea, i wonder if there;s any languages out there that do it

# ? Apr 30, 2015 16:31

Sapozhnik: Jan 2, 2005; Nap Ghost

Arcsech posted:

we could always make and_then into an operator like... Hmm, how about >>=

making it infix seems pretty cool

not sure about replacing a nice idiomatic "and_then" with gobbledygook symbols though

# ? Apr 30, 2015 16:36

Subjunctive: Sep 12, 2006; ✨sparkle and shine✨

fart simpson posted:

but what if get_shortest() returned null? then you need to build explicit null handling into every function? they could have written the option example like this
code:
fn get_user_with_shortest_name(names: Vec<&str>) -> Option<User> {
   get_shortest(names).and_then(find_user_by_name).and_then(json_to_user)
}

it also only works if you have existing functions with the right signature and behaviour, and don't require any fields to be extracted or anything to be computed specific to this application.

still, I see the appeal of optional-friendly syntax. wonder why they didn't go there; too much "put it in the library"?

# ? Apr 30, 2015 16:39

Adbot: ADBOT LOVES YOU

# ? Jun 1, 2024 09:23

Bloody: Mar 3, 2013

this all looks bad and worse than doing it in c#

# ? Apr 30, 2015 16:40

The Something Awful Forums > Discussion > Serious Hardware/Software Crap > YOSPOS > p-lang thread: (now (have you (problems two)))

«‹›1784 »