[
http://jira.codehaus.org/browse/JANINO-90?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#action_96417 ]
Adam Heath commented on JANINO-90:
----------------------------------
Um, you *can't* write ncfde.getClass(). It's static. There's no danger of recursion. It's a class from java.lang; it will not be loaded by janino.
As for it being needed, if there is an error loading the class, then it's this *new* error that is the cause of the problem. Just throwing a NCFDE here will not help in solving the real problem. Consider the case where the class that was initially requested parsed, compiled, and was loaded, but then it's static initialization block throws an Error or RuntimeException, when it tries to access another class in it's package, that is also being handled by janino. In this case, the other class parses and compiled fine, but fails verification, because janino is not a valid compiler.
That verification failure most often occurs because janino doesn't have proper variable/method resolution, when it comes to implicit methods access by 'this'.
> Make use of initCause in class$, if available.
> ----------------------------------------------
>
> Key: JANINO-90
> URL:
http://jira.codehaus.org/browse/JANINO-90> Project: Janino
> Issue Type: New Feature
> Reporter: Adam Heath
> Assignee: Arno Unkrig
> Priority: Trivial
> Attachments: feature_use_initCause_in_class$_method.patch
>
>
> If initCause is available(java 1.4), the use that to initialize chained exceptions in class$ method.
--
This message is automatically generated by JIRA.
-
If you think it was sent incorrectly contact one of the administrators:
http://jira.codehaus.org/secure/Administrators.jspa-
For more information on JIRA, see:
http://www.atlassian.com/software/jira
---------------------------------------------------------------------
To unsubscribe from this list please visit:
http://xircles.codehaus.org/manage_email
