On Sun, Dec 2, 2012 at 2:11 AM, Evan Mezeske wrote:
>
> That said what's the compelling reason these days for lein-cljsbuild to
>> depend on a specific version of ClojureScript? Are you relying on certain
>> aspects of the analyzer or compiler's API and find that they change quite
>> frequently?
> That said what's the compelling reason these days for lein-cljsbuild to
> depend on a specific version of ClojureScript? Are you relying on certain
> aspects of the analyzer or compiler's API and find that they change quite
> frequently?
I guess the (debatably compelling) reason for that t
Can we move forward on this? Option 1 seems best to me. That said what's
the compelling reason these days for lein-cljsbuild to depend on a specific
version of ClojureScript? Are you relying on certain aspects of the
analyzer or compiler's API and find that they change quite frequently?
On Fri, N
Tracking this at http://dev.clojure.org/jira/browse/CLJS-418
-S
--
You received this message because you are subscribed to the Google
Groups "Clojure" group.
To post to this group, send email to clojure@googlegroups.com
Note that posts from new members are moderated - please be patient with your
I agree with Herwig, in that if the ClojureScript JAR makes use of
goog.async.Deferred, which is from the third party library, then the JAR
really needs to depend on the third-party library.
I doubt that lein-cljsbuild is the only thing broken by this -- I expect
this would break things like pi
I concur, but maybe we could make a second release of the closure lib main
jar, with a dependency on third-party?
That would reflect the actual dependency.
I know, you asked for testing before the release. Sorry for not catching
this one.
2012/11/13 Stuart Sierra
> My original reason for split
My original reason for splitting the third-party JAR was the licensing: the
main G.Closure library is under the Apache license; the third-party
libraries are under a variety of different licenses.
One simple solution would be to make ClojureScript itself have a dependency
on the third-party lib
I've debugged a bit, the underlying cause seems to be that
goog.net.xpc.CrosspageChannel uses goog.async.Deferred, which lives in
third_party.
In my project, I could indeed successfully compile again with the browser
repl, by adding the third-party jar.
This leaves two open issues:
Why the NPE in
I've seen this same NPE in my project. I managed to work around it by
temporarily throwing out the browser repl. It seems to trigger the NPE. Not
sure how and why.
2012/11/13 Evan Mezeske
> The policy is: update the default version as quickly as possible when a
> new ClojureScript compiler revi
The policy is: update the default version as quickly as possible when a new
ClojureScript compiler revision is released. So, feel free to poke me if
it seems like I haven't noticed a new release (as I might not have!). :)
Unfortunately, after setting the default ClojureScript version to 0.0-15
On Sun, Nov 11, 2012 at 8:49 PM, Evan Mezeske wrote:
> Oops, I've been out of town and did not see this release! I'll cut a new
> lein-cljsbuild tomorrow night.
Wasn't trying to make you feel guilty! Just wondered if there was a
policy of tracking builds.
Welcome back - and thank you!
--
Sean
Oops, I've been out of town and did not see this release! I'll cut a new
lein-cljsbuild tomorrow night.
-Evan
On Friday, November 9, 2012 9:51:01 AM UTC-7, Sean Corfield wrote:
>
> Nice!
>
> I wonder how quickly lein-cljsbuild will get updated for this new
> release? (Evan?)
>
> Nice to see
Nice!
I wonder how quickly lein-cljsbuild will get updated for this new
release? (Evan?)
Nice to see Leiningen and lein-cljsbuild introduced up front in your
book as the quickest way to get up and running, Stuart!
Sean
On Fri, Nov 9, 2012 at 6:17 AM, Stuart Sierra
wrote:
> ClojureScript releas
ClojureScript release 0.0-1535 is out. Get it in Leiningen:
[org.clojure/clojurescript "0.0-1535"]
Change log for this release:
http://build.clojure.org/job/clojurescript-release/19/
This release depends on the latest version of the Google Closure Library,
r2029. You can also depend on it e
14 matches
Mail list logo