On 17/05/2013 11:54, Chris Hegarty wrote:
There are a few networking tests failing recently. All have bugs
against them. These bugs will be address shortly, but since tl is
integrating into master, for b91, next Tuesday, it is best to add them
to the ProblemList so that the failures don't escape into master.
diff -r 06b410feed49 -r 1e860864089a test/ProblemList.txt
--- a/test/ProblemList.txt Fri May 17 11:05:43 2013 +0100
+++ b/test/ProblemList.txt Fri May 17 11:40:54 2013 +0100
@@ -205,6 +205,18 @@ java/net/MulticastSocket/Test.java
#7143960
java/net/DatagramSocket/SendDatagramToBadAddress.java macosx-all
+#8014783
+java/net/HttpURLPermission/HttpURLPermissionTest.java windows-all
+
+#8014720
+java/net/ResponseCache/B6181108.java generic-all
+
+#8014723
+sun/misc/URLClassPath/ClassnameCharTest.java generic-all
+
+#8014719
+sun/net/www/http/HttpClient/ProxyTest.java generic-all
+
-Chris.
For java/net/HttpURLPermission/HttpURLPermissionTest.java then I assume
that using try-with-resources around the code that opens the .ser files
will sort this out (meaning might be as quick to just fix this one).
For the rest then I agree, these need to be excluded until
HttpURLPermission can support proxying of ftp connections.
The long standing format has been to put a space before the bug number,
I don't know if we have any grep-like tools that depend on that.
As an aside, we need to re-format this file at some point to put the bug
numbers onto the same line as the test. That's the format that jtreg
prefers and would it be generally more grep-friendly. This needs to be
done in conjunction with taking an axe to jdk/test/Makefile and I think
on Mike's list.
-Alan
