anyone around?

@tankthinks yes

hey @nathanmarz!

first and foremost … specter is awesome

now that I have that out of the way

thanks

I need to transform this data structure (I’ve elided unnecessary bits for now):

(def plans
  {:plans '({:planId "1"
             :disbursements ({:history ({:amount "1" :disbursementDate "2017-01-01"}
                                        {:amount "1" :disbursementDate "2017-01-01"}
                                        {:amount "2" :disbursementDate "2017-02-01"}
                                        {:amount "3" :disbursementDate "2017-03-01"})})})})

by removing duplicate entries from the :history list

this basically does what I want:

(transform [:plans ALL #(= "1" (:planId %)) :disbursements ALL :history]
  dedupe
  plans)

but what if I want a predicate to ensure that I’m only dedupeing entries from January, i.e. “2017-01-01”

hmm

gonna have to do it manually for now in your transform fn

it's possible subselect could be extended to support that

so you could do:

(transform [:plans ALL #(= "1" (:planId %)) :disbursements ALL :history (filterer :disbursementDate #(= % “2017-01-01”))]
  dedupe
  plans)

right now subselect/`filterer` require the output sequence to be the same size as the input

the logic change would be to fill in removed entires with NONE to trigger removal

that would be a good issue to open on github

ah … cool

Thank you for the quick response … I’ll play around a bit to make sure I completely understand, then I’ll submit an issue

actually it's fine

I was curious so I just implemented it

haha

@tankthinks https://github.com/nathanmarz/specter/commit/be30e46960c879d6177d8ab8c2d3f4450bab0095

quite elegant I think

so … uh … you’re welcome?

😃

😉

thanks again @nathanmarz, not sure how I would have done that without specter … cheers!