Does anyone have an idea for what I might do to diagnose/fix this? Is this
just a misunderstanding on my part? If not I'll file a bug against
definline.
On 21 June 2013 18:04, Colin Fleming wrote:
> So this has just happened to me again:
>
> Clojure 1.5.1
> (plugin.psi/symbol? 2)
> => false
> (
So this has just happened to me again:
Clojure 1.5.1
(plugin.psi/symbol? 2)
=> false
(filter plugin.psi/symbol? [1 2 3])
=> (1 2 3)
((var-get #'plugin.psi/symbol?) 2)
=> (clojure.core/instance?
org.jetbrains.plugins.clojure.psi.api.symbols.ClSymbol 2)
What that looks like to me is that the macro
ClSymbol is a Java class. I don't get the replacement warning because I've
excluded that symbol explicitly in my ns declaration using :refer-clojure
:exclude.
I haven't done a 'lein clean' because I'm not using lein, but I have
rebuilt various times. However, sometimes it will work and sometimes i
On 20/06/13 10:59, Colin Fleming wrote:
Because this tests for something different - that the element is an
instance of ClSymbol. It's not testing the same thing as the core
version. I qualify it (psi/symbol? in the examples above) to
distinguish it from the core one.
Basically, I'm trying to
Because this tests for something different - that the element is an
instance of ClSymbol. It's not testing the same thing as the core version.
I qualify it (psi/symbol? in the examples above) to distinguish it from the
core one.
Basically, I'm trying to use definline to allow me to have a more Clo
There is already a symbol? predicate in core. Why are you defining your
own? Does your problem disappear when you use the one from core?
What exactly are you trying to do? I use definline quite frequently and
have never encountered such problems...
Jim
On 20/06/13 10:35, Colin Fleming wrote:
Hi all,
I'm having a problem where a definline'd function is not working when used
as a predicate for filtering. It seems to work sporadically - occasionally
it works, occasionally it doesn't. I'm thinking it's probably a compile
problem since it seems to work or not work consistently for each com