RFR: 8151542: URL resources for multi-release jar files have a #runtime fragment appended to them
paul.sandoz at oracle.com
Fri Apr 29 16:52:48 UTC 2016
> On 29 Apr 2016, at 07:32, Alan Bateman <Alan.Bateman at oracle.com> wrote:
> On 28/04/2016 22:18, Steve Drach wrote:
>> I’ve updated the webrev to change all instances of the word “reified” to “real” as in getRealName().
>> Issue: https://bugs.openjdk.java.net/browse/JDK-8151542
>> Webrev: http://cr.openjdk.java.net/~sdrach/8151542/webrev.01/
> The src changes looks okay but did we come to a conclusion on URLClassLoader spec? If not, can we revert the change to URLClassPath getLoader and deal with it separately?
AFAICT this fix does not really change the existing implementation behaviour of URLClassLoader and URLClassPath. The patch redirects the processing of “jar:…/!” from the more general URLClassPath.Loader to the URLClassPath.JarLoader.
But i think the cracking of the URL within the jar URL can be made more robust. It probably should use
to check there is just one occurrence at the end.
Separately the spec of URLClassLoader could be clarified as to the current implementation behaviour.
More information about the core-libs-dev