digits problem.

[ribelo]

(let [s "word100word"]
  (= (->snake_case (->kebab-case s))
     "word100word"))
=> false

becaouse

(->kebab-case "word100word")
=> "word-100-word"

(->snake_case "word-100-word")
=> "word_100_word"

(->snake_case "word100word")
=> "word_100word"

In my opinion numbers case should be recognised as a case of the preceding word

[qerub]

As discussed in #22, it's non-trivial to do a perfect job of generically splitting strings in the presence of digits. My current recommendation is to use the :separator argument if the generic splitter is causing problems:

(-> "word100word"
    (->kebab-case :separator \_)
    (->snake_case :separator \-))
; => "word100word"

My long-term plan is to revise the library to be slighly more flexible.

In my opinion numbers case should be recognised as a case of the preceding word

So you'd like (->kebab-case "word100word") to be "word100-word"?

[mdaley]

Hello, well I know this is a rather old issue but as I found it when trying to solve this precise problem, I thought I would suggest my solution. Basically, follow up ->kebab-case with this function:

(defn- fixup-dash-number
  "Fixup any instances of -[0-9] to be [0-9]."
  [s]
  (replace s #"\-([0-9])" "$1"))

So, perhaps this avoids the need to make something complex inside CSK? Or perhaps it provides an idea for an internal fix? OK, it's not very clever or optimal (being a fix after the event) but it is nice and simple.

[lotuc]

I prefer numbers case should be recognised as a case of the preceding word also.

Here we did a dirty alter-var-root patch on our code base which change the line which defines the behavior.

https://gist.github.com/lotuc/fba421f74ea8610494118bc899f9f1b5#file-camel_snake_kebab_utils-clj-L35