Clojurians Log v2

finally finished it, but as a beginner, I would really appreciate some reviews on it, if anyone has the time and inclination https://github.com/ashnur/nimble-tt

in valid-rank? you've written (and (< 1 rank) (> 15 rank)) which you could write as (< 1 rank 15) ... but shouldn't that be (< 0 rank 15)? I would have thought that 1 was a valid card rank to have.... ah ... no ... sorry ... just spotted that you're turning 1 into 14 in hand-result forcing aces to be high ... does that mean you can't have a flush of 1 2 3 4 5?

Also, I would have expected the namespace to be main rather than src.main

(mapv (partial vector) ranks suits) is the same as (mapv vector ranks suits)

(seq (filter (complement valid-rank?) ranks)) is can be shortened to (seq (remove valid-rank?) ranks) but I think it would be clearer to write (every? valid-rank? ranks)

Ed, that's a good catch about the flush

and I never understand the namespaces

literally any time I try anything, it's wrong

I can't even find any documentation on it, like different dir structures used differently, and what is the appropriate namespacing for them

every? valid-rank? ranks is great suggestion too, not sure how i missed it

err... not flush, straight, but I got your meaning

The file layout convention comes from the Java world where it's common to have a src directory with all your application code in and a test directory with your tests in. Often in Clojure land you'd describe the directory layout and dependencies for your project in a deps.edn file (https://clojure.org/reference/deps_and_cli#_paths) The relationship between the namespace and the filename is generally a directory for each of the parts of the namespace when separated with a . except the last one, which will be a file. So (ns foo.bar.baz ,,,) will be in src/foo/bar/baz.clj and you'd probably put tests for that ns in (ns foo.bar.baz-test ,,,) / test/foo/bar/baz_test.clj - and there we find the next wrinkle. For namespaces that have a - in them we need to replace it with a _ in the filename. This is because of a restriction in the jvm, and ClojureScript has followed the same convention to stay compatible. There's some more info here: https://www.clojure.org/guides/learn/namespaces#_loading ... does that make sense?

> err... not flush, straight, but I got your meaning yeah ... sorry - I'm not so good at poker 😉

Yeah, it is an easy fix in clojure, I just added an (or (= ranks [2 3 4 5 14])), since it's ordered, should be fine

the part I don't get in namespaces is exactly what you also don't mention btw 😄

if I have src and test, how do I refer between them?

often it appears to me that whole directories are created for no reason at all

I think I'm general the src directory is used to keep the code you're intending to ship to production and the test directory is for your tests that you probably won't include in the deployable artifacts. In development they both paths should be on the list of places that would get searched for namespaces.

thanks, that's helpful hint

> Maybe call pprint and leverage quoting so it treats the outer (def some-data ...) as a list? That's what I'd do. Though I think I'd quote the whole EDN thing as well, producing (def some-data (quote literal-edn)) This will fail when trying to read :items:

user=> (clojure.pprint/pprint (list 'def 'some-data {:point {:x 1 :y 2} :name "Harold" :items (range 20)}))
(def
 some-data
 {:point {:x 1, :y 2},
  :name "Harold",
  :items (0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19)})

user=> (clojure.pprint/pprint (list 'def 'some-data (list 'quote {:point {:x 1 :y 2} :name "Harold" :items (range 20)})))
(def
 some-data
 '{:point {:x 1, :y 2},
   :name "Harold",
   :items (0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19)})

if you can survive

(def
 some-data

Landed on the following:

(pprint-str `(def ~'cloud-services ~cloud-services-json))