[ANN] Clojure 1.7.0-beta3

186 views
Skip to first unread message

Alex Miller

unread,
May 13, 2015, 9:55:15 AM5/13/15
to clo...@googlegroups.com, cloju...@googlegroups.com, clojur...@googlegroups.com
Clojure 1.7.0-beta3 is now available.

Try it via
- Leiningen: [org.clojure/clojure "1.7.0-beta3"]

Additional enhancements to new features since 1.7.0-beta2:

1) CLJ-1728 - `source` fn now works for vars in cljc files
2) CLJ-1716 - the new print format for Throwables now includes ex-data
3) CLJ-1726 - restored IPending/realized? support for iterate and cycle sequences to detect delayed computation

Regression and bug fixes since 1.7.0-beta2:

1) CLJ-1723 - expanding transducers (cat, mapcat) that produced nil values would throw NPE
2) CLJ-1727 - new long-optimized range had bugs related to overflow
3) CLJ-1648 - use equals() instead of == when resolving Symbol in compiler
4) CLJ-1195 - deftype should expand to ns-qualified invocation of fn to allow overrides (this issue was blocking Prismatic Schema via Plumbing issue #74)
 
For a full list of changes since 1.6.0, see:

Please give it a try and let us know if things are working (or not)!

- Alex

Andy Fingerhut

unread,
Jun 30, 2015, 6:13:39 PM6/30/15
to clojur...@googlegroups.com, clo...@googlegroups.com, cloju...@googlegroups.com
Is anyone interested in filing a bug against IBM JDK 1.8.0 where it appears that its JIT compiler fails about half of the time when doing 'mvn clean test' in a Clojure 1.7.0 source tree?

You can find some attachments in an earlier message of this thread that can get you started, but producing your own error logs with specific versions of OS, JDK, and Clojure source code would probably be best.  Let me know if you can't reproduce it for some reason.

I don't use IBM JDK very much, other than to see if anything goes wrong when you try to build and test Clojure using it.  I'm hoping someone more motivated in seeing this fixed for the IBM JDK will take up the torch.

Andy

On Sun, May 17, 2015 at 12:53 AM, Herwig Hochleitner <hhochl...@gmail.com> wrote:
Given that the JVM shouldn't segfault in any case, that's cleary looks to me like an ibm jdk bug, that could hit other JVM languages aswell. Maybe IBM is interested in fixing it?
In my case, the VM most likely has the excuse of not expecting classpath-jar files to be swapped out, but the JIT?

--
Note that posts from new members are moderated - please be patient with your first post.
---
You received this message because you are subscribed to the Google Groups "ClojureScript" group.
To unsubscribe from this group and stop receiving emails from it, send an email to clojurescrip...@googlegroups.com.
To post to this group, send email to clojur...@googlegroups.com.
Visit this group at http://groups.google.com/group/clojurescript.

Reply all
Reply to author
Forward
0 new messages