ramda/ramda

Practical functional Javascript

Dimitri Mitropoulos

@dimitropoulos

const DEBUG_MODE = true;
let debugTimestamps: { [label: string]: number }[] = [];
const debugPipe = (label: string) => <T>(input: T) => {
  if (DEBUG_MODE) {
    debugTimestamps = [
      ...debugTimestamps,
      {
        [label]: performance.now(),
      },
    ];
    console.log({ [label]: input });
  }
  return input;
}

it's basically the same thing, but allows for some recording of timestamps, and then I have a startDebug or endDebug that basically just grabs the debugTimestamps and prints the performance numbers

Hyunseok Oh

@algoshipda

how can i make this function in ramda way?

const fn = x => id => R.prop(id, x);

Ian Hofmann-Hicks

@evilsoft

Try R.flip(R.prop), that should flip the arguments to R.prop

The C combinator I believe it is called

Ian Hofmann-Hicks

@evilsoft

const fn =
  R.flip(R.prop)

flip is very useful when you have to use other libs/apis that are not set up for easy composition with the data last format (or in cases where your data is the key name). But very handy tool in your belt, none the less

Hyunseok Oh

@algoshipda

@evilsoft wow thanks

Charles Hughes

@chughes87

@algoshipda you could also use _ which i think is more readable: https://ramdajs.com/docs/#__

so const fn = R.prop(R._, x) I believe?

Ian Hofmann-Hicks

@evilsoft

Personally I find the placholders a little less readable than using combinators like flip which provide descriptive names for how the combinator acts on the functions. But I think it is really just a taste thing.

Charles Hughes

@chughes87

Yeah I could go either way in this case. I think for functions that take 3+ arguments though there is little choice

Ian Hofmann-Hicks

@evilsoft

also would that x need to be provided, like const fn = x => R.prop(R._, x). and that would loose the nice auto curry that flip would of provided. so you would probably wanna wrap that in an R.curry

Charles Hughes

@chughes87

yah good point

wangzengdi

@adispring

what's other purpose of internal api _arity, except for limiting arguments'length to less than or equals to 10

https://github.com/ramda/ramda/blob/v0.26.1/source/internal/_arity.js

for being easy to curry the function returned by other functions ?such as compose and pipe

Long Dao

@longebane

Awww yeah

Ian Hofmann-Hicks

@evilsoft

:wave: @longebane how the heck are ya

kapilpipaliya

@kapilpipaliya

onDestroy(() => { service.send('DESTROY'); });

how can i write this in ramda?

Twizzes

@Twizzes

@kapilpipaliya what are you trying to do

ramda isn't really a language, but more of a utility belt

and in that example, I don't see anything that could take advantage of ramda without seeing more context

kapilpipaliya

@kapilpipaliya

thanks

dattran1232003

@dattran1232003

Hey, I have one idea.

What if we could do this:

const appendLeft = (arr) => {}

dattran1232003

@dattran1232003

Const appendLeft = (elm, arr) => R.append(elm, arr.reverse()).reverse()

Brad Compton (he/him)

@Bradcomp

What is the difference between appendLeft and R.prepend?

dattran1232003

@dattran1232003

Is it exists?

Oh i don't no, sorry hehe

Brad Compton (he/him)

@Bradcomp

No prob! Happy to help :smile:

Pierre-Antoine Mills

@millsp

Hi all!

I come here to ask you what do you think is the best way to reduce two lists at once?

Supposing that the two lists are exactly the same length, how could I reduce over list1, access the current item (with index usually) and get the value at that same index in list2.

In plain JS, I would usually use the indexfrom the callback, but it see that ramda works differently. I'd like to What do you think?

Rob Grant

@robgrant

zip first, then reduce

mitcho

@mitchoSR_twitter

someone told me a good practice for composing sorter functions is that if they return 0 it delegates to the next function, i defined this as

const pipeCritera = (...fns) => (a, b) => (fns.find(f => f(a, b) !== 0) || (() => 0))(a, b);

essentially returning the value of the first sorting function that doesnt return 0,

now im wondering if ramda already has this covered

Pierre-Antoine Mills

@millsp

@robgrant this could double the iteration time on very long lists. Wouldn't it be wiser to have the index provided by the reducer, in that case of repeated operations/long lists?

I wonder what you think about this @Bradcomp

Ben Briggs

@ben-eb

@mitchoSR_twitter @pirix-gh Probably better to do this with reduce, here's one such implementation: https://stackoverflow.com/a/55990011

mitcho

@mitchoSR_twitter

ahh cool, thats a bit cleaner

thanks ben

Brad Compton (he/him)

@Bradcomp

What about R.sortWith?

mitcho

@mitchoSR_twitter

bingo

ty ramda tour guides XD

thanks brad

Brad Compton (he/him)

@Bradcomp

:bowtie:

Long Dao

@longebane

@evilsoft when you said javascript started out as a FP language, what did you mean?

Ian Hofmann-Hicks

@evilsoft

As I understand it, when it was being created, it was intended to be a LISP dialect, but near the end of development he added features like prototypical inheritance, this and other bit to make it more Java-like. That is why, from my understanding things like closure scope and first class functions are at the core. Which are the only (2) features that are required to be a functional language

Trying to track down my sources, one was an interview with Brendan Eich and another was a book. Gonna see if I can hunt those down and post them here

Brad Compton (he/him)

@Bradcomp

I think I read the same interview

Where communities thrive

People

Repo info

Activity