Fork me on GitHub
#cljs-dev
<
2016-02-25
>
mfikes15:02:32

Bug or not? : char? fails on nil in ClojureScript but returns false in Clojure. I ask because ClojureScript is understandably forced to differ from Clojure with respect to (char? “x”)

mfikes15:02:47

Seems cut-n-dry based on the docstring that (char? nil) should return false in ClojureScript.

dnolen15:02:33

@mfikes: hrm but what is the behavior for other kinds of values?

mfikes15:02:54

@dnolen: I tried a few and also became worried about that

dnolen15:02:28

right if there are too many cases

dnolen15:02:32

then no we don’t want to do anything about it

dnolen15:02:36

too host-y

mfikes15:02:02

@dnolen: This happened for me in bootstrap when macroexpanding cond where one of the tests is nil. Perhaps a conservative fix would be to just limit the scope of a fix to bootstrap (i.e., use fnil if :cljs).

mfikes15:02:33

@dnolen: No rush to decide. I can log a ticket and let it sit.

dnolen15:02:42

sounds good

mfikes15:02:36

@dnolen: As an aside, a really good thing happened: I can now run the cljs.core-test suite in bootstrap. This will let me bang out the last few oddities like this. simple_smile

dnolen15:02:32

@mfikes: wow awesome!