Oleksandr Yakushev

This is a bug, sorry. `:delayed-transform` was originally intended for values that you **do** provide but want to do some post-processing on them. I thought it also supported what I've...

Damn, I must have had unclean state when testing. This seems to work though: ```clojure (cfg/define {:hdfs-ip {:type :string} :hdfs-port {:type :number} :hdfs-url {:type :string :default ::none ;; Needed for...

Sorry! That is another bug. I've just pushed a fix to it as `0.2.8-SNAPSHOT`, please try :).

That's because you put a function definition into the file. When you do `read-string` it does not get compiled but treated as plain s-expressions. You may `eval` it, but I...

From a logical standpoint, defining such values as `:edn` is probably the correct solution, because you are not giving a separate identity to each element of the collection. Perhaps, it...

I will leave this issue open until I figure out if having a nested type for lists is doable and worth doing. Thanks for the idea, Anuj!

Hi Nicolas, I think there is no problem with Omniconf supporting such feature. I would accept this as a PR if you don't mind implementing it non-breakingly, e.g. by introducing...

This is a good point. However, if we are to go this way I prefer making a comprehensive munging/unmunging suite for each config source (env, cli opts, etc.), not just...

There is munging, but it is for making Clojure literals Java (JVM) friendly. We'll need custom munging/unmunging for our purposes.

Hi Cristopher, I've indeed been thinking about making it possible in Omniconf to have separate configuration maps. I think it is possible to do so without breaking the backwards compatibility....