Does this code reflect that?
Rich
> --
> You received this message because you are subscribed to the Google Groups "Clojure Dev" group.
> To post to this group, send email to cloju...@googlegroups.com.
> To unsubscribe from this group, send email to clojure-dev...@googlegroups.com.
> For more options, visit this group at http://groups.google.com/group/clojure-dev?hl=en.
>
Rich,
Do you mean in cljs.compiler.macroexpand-1? I have yet to touch that, but I plan to look at it whenever I get home tonight. However, in the cases of (.-p o) -> (. o -p) and (.m o) -> (. o m) the relative action semantics (i.e. lookup and call) are the same. Do you know of a representational case that I should be mindful of?
I have a branch at https://github.com/clojure/clojurescript/tree/prop-lookup
with support for the new ClojureScript property lookup syntax[1]. If
you have some time please have a look at the implementation and run
through some scenarios that you have in your own code to make sure
that it works (and breaks) as expected.
Fixed. Thank you for catching that.
:F
Thank you for looking. I just pushed a fix to the branch.
:F
:F
Me and Alan are working through making this more generic and should
have something out later today.
Thanks
:F
Alan
- general announcement regarding its breaking nature and how to fix it
in the general case
- a co-release in some 1.4-alpha version of the same prop-lookup syntax
- (ideally) a utility or compiler util that can identify and warn the
old-style lookup cases in people's code
+ this might not be 100% comprehensive, but it might be worth exploring
Am I missing anything else?
:F
I just fixed apply. The only thing keeping us from pushing this
prop-lookup behavior is:
- general announcement regarding its breaking nature and how to fix it
in the general case
- a co-release in some 1.4-alpha version of the same prop-lookup syntax
- (ideally) a utility or compiler util that can identify and warn the
old-style lookup cases in people's code
+ this might not be 100% comprehensive, but it might be worth exploring
Am I missing anything else?
:F
Define necessary.
> Clojure isn't changing in any breaking way, and ClojureScript is in alpha -
> do we need this?
If we can take some of the burden off people with existing code then
that's worthwhile. Minimally we can print a warning whenever the
compiler sees a (. target something) pattern that simply raises
awareness that it might not mean what it once meant.
>> - a co-release in some 1.4-alpha version of the same prop-lookup syntaxDefine necessary.
>
> Is this necessary?
> Clojure isn't changing in any breaking way, and ClojureScript is in alpha -If we can take some of the burden off people with existing code then
> do we need this?
that's worthwhile. Minimally we can print a warning whenever the
compiler sees a (. target something) pattern that simply raises
awareness that it might not mean what it once meant.
>
> - a co-release in some 1.4-alpha version of the same prop-lookup syntax
>
> Is this necessary?
Yes. It's necessary so that ClojureScript implements Clojure on JS, and not some new language.
But, it is also not a big deal on the Clojure side, nor is generating an alpha.
Rich
Nothing on my end is needed. I'm ready for the required merge.