According to both the JLS [1] and the JVM Spec [2], neither '>' or '<'
are legal within an identifier; The various Character.isJava***()
methods return false for those characters.
As for the original problem, I was able to reproduce it on some
Windows IBM JVMs [3] and I verified that applying the
If you apply the patch I have added at https://www.assembla.com/spaces/clojure/tickets/104-gc-issue-100--gen-class-creating-non-compliant-field-names-
you should be able to tell pretty quickly.
If you aren't comfortable with git patches, building clojure, etc.
ping me offline and I can just s
I've fallen victim to #104 in an IBM zSeries environment. I can't
offer much in the way of Java literacy, but I can certainly test
anything you'd care to toss my way.
--
You received this message because you are subscribed to the Google
Groups "Clojure" group.
To post to this group, send email t
I am preparing a patch to fix issue #104 [1], a.k.a. "IBM JVM rejects
member names with dashes (-) that Sun JVMs like just fine." The
original thread [2] included a patch that uses $ as a delimiter, e.g.
"who$likes$reading$this". My preference is to use underscores, e.g.
"isnt_this_much_bet