Ping again.
If there is no more suggestion, I'll push my fix in the first webrev.
Here I'm more concerned about the build failure than the behavior of the
input streams.
Thanks
Max
On 11/12/2012 06:22 PM, Weijun Wang wrote:
In fact, it looks like the resources object here is of type
InputStreamEnumeration (in
com/sun/naming/internal/VersionHelper12.java), with an
open-on-nextElement style. This means we can completely remove the while
(resources.hasMore()) block.
Lance and Andrew added for confirmation.
Thanks
Max
On 11/12/2012 05:42 PM, Alan Bateman wrote:
On 12/11/2012 08:43, Weijun Wang wrote:
A small webrev:
http://cr.openjdk.java.net/~weijun/8003263/webrev.00/
The reason is that 8003120 added a new line
InputStream istream = (InputStream)resources.next();
but resources is already of type NamingEnumeration<InputStream>.
Building now shows
../../../src/share/classes/com/sun/naming/internal/ResourceManager.java:563:
warning: [cast] redundant cast to InputStream
InputStream istream =
(InputStream)resources.next();
Noreg-trivial.
Looks okay to me, just wondering if we should do a more complete fix for
8003120 while you are there (meaning it should attempt to close all
resources even if there is an IOException is thrown during close).
-Alan
